commit d820665f6c420edec33eecb215eb619b678f92e4 Author: Luna Date: Sun Feb 9 18:48:25 2025 +0100 Initial commit :3 diff --git a/Makefile b/Makefile new file mode 100644 index 0000000..89a7de4 --- /dev/null +++ b/Makefile @@ -0,0 +1,32 @@ +PKG_CONFIG?=pkg-config + +PKGS=wayland-client xkbcommon vulkan +CFLAGS+=$(shell $(PKG_CONFIG) --cflags $(PKGS)) +LIBS=$(shell $(PKG_CONFIG) --libs $(PKGS)) +LDFLAGS+=-lm -lglfw -rdynamic + +FLAGS=-g $(CFLAGS) $(LDFLAGS) $(LIBS) + +S=main.c +S+=comp.c wayland.c glfw.c +S+=vulkan.c kitty.c +S+=hashmap.c io.c matrix.c dynarray.c image.c types.c allocator.c log.c +S+=object.c register.c +SO=$(addprefix build/,$(S:.c=.o)) +# SC=$(addprefix src/,$(S)) +SC=$S + +clean: + rm -f build/* + +install: + ./main + +build/%.o: %.c + gcc -c $(FLAGS) -o $@ $< + +main: $(SO) Wayland/xdg-shell-protocol.c + gcc $(FLAGS) -o $@ $(SO) Wayland/xdg-shell-protocol.c + +.DEFAULT_GOAL=main +.PHONY: clean diff --git a/Shaders/shader.frag b/Shaders/shader.frag new file mode 100644 index 0000000..ed9d88e --- /dev/null +++ b/Shaders/shader.frag @@ -0,0 +1,11 @@ +#version 450 +layout(location = 0) in vec3 fragColor; +layout(location = 1) in vec2 fragUV; +layout(location = 0) out vec4 outColor; + +layout(binding = 1) uniform sampler2D texSampler; + +void main() { + outColor = texture(texSampler, fragUV) * vec4(fragColor, 1.0); +} + diff --git a/Shaders/shader.vert b/Shaders/shader.vert new file mode 100644 index 0000000..c87c266 --- /dev/null +++ b/Shaders/shader.vert @@ -0,0 +1,19 @@ +#version 450 + +layout(location = 0) in vec2 inPosition; +layout(location = 1) in vec3 inColor; + +layout(set = 0, binding = 0) uniform UBO { + mat3 model; + mat3 view; + mat3 proj; +} ubo; + +layout(location = 0) out vec3 fragColor; +layout(location = 1) out vec2 fragUV; + +void main() { + gl_Position = vec4(ubo.proj * ubo.view * ubo.model * vec3(inPosition, 1.0), 1.0); + fragColor = inColor; + fragUV = inPosition; +} diff --git a/Wayland/xdg-shell-client-protocol.h b/Wayland/xdg-shell-client-protocol.h new file mode 100644 index 0000000..708898c --- /dev/null +++ b/Wayland/xdg-shell-client-protocol.h @@ -0,0 +1,2325 @@ +/* Generated by wayland-scanner 1.23.0 */ + +#ifndef XDG_SHELL_CLIENT_PROTOCOL_H +#define XDG_SHELL_CLIENT_PROTOCOL_H + +#include +#include +#include "wayland-client.h" + +#ifdef __cplusplus +extern "C" { +#endif + +/** + * @page page_xdg_shell The xdg_shell protocol + * @section page_ifaces_xdg_shell Interfaces + * - @subpage page_iface_xdg_wm_base - create desktop-style surfaces + * - @subpage page_iface_xdg_positioner - child surface positioner + * - @subpage page_iface_xdg_surface - desktop user interface surface base interface + * - @subpage page_iface_xdg_toplevel - toplevel surface + * - @subpage page_iface_xdg_popup - short-lived, popup surfaces for menus + * @section page_copyright_xdg_shell Copyright + *
+ *
+ * Copyright © 2008-2013 Kristian Høgsberg
+ * Copyright © 2013      Rafael Antognolli
+ * Copyright © 2013      Jasper St. Pierre
+ * Copyright © 2010-2013 Intel Corporation
+ * Copyright © 2015-2017 Samsung Electronics Co., Ltd
+ * Copyright © 2015-2017 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ * 
+ */ +struct wl_output; +struct wl_seat; +struct wl_surface; +struct xdg_popup; +struct xdg_positioner; +struct xdg_surface; +struct xdg_toplevel; +struct xdg_wm_base; + +#ifndef XDG_WM_BASE_INTERFACE +#define XDG_WM_BASE_INTERFACE +/** + * @page page_iface_xdg_wm_base xdg_wm_base + * @section page_iface_xdg_wm_base_desc Description + * + * The xdg_wm_base interface is exposed as a global object enabling clients + * to turn their wl_surfaces into windows in a desktop environment. It + * defines the basic functionality needed for clients and the compositor to + * create windows that can be dragged, resized, maximized, etc, as well as + * creating transient windows such as popup menus. + * @section page_iface_xdg_wm_base_api API + * See @ref iface_xdg_wm_base. + */ +/** + * @defgroup iface_xdg_wm_base The xdg_wm_base interface + * + * The xdg_wm_base interface is exposed as a global object enabling clients + * to turn their wl_surfaces into windows in a desktop environment. It + * defines the basic functionality needed for clients and the compositor to + * create windows that can be dragged, resized, maximized, etc, as well as + * creating transient windows such as popup menus. + */ +extern const struct wl_interface xdg_wm_base_interface; +#endif +#ifndef XDG_POSITIONER_INTERFACE +#define XDG_POSITIONER_INTERFACE +/** + * @page page_iface_xdg_positioner xdg_positioner + * @section page_iface_xdg_positioner_desc Description + * + * The xdg_positioner provides a collection of rules for the placement of a + * child surface relative to a parent surface. Rules can be defined to ensure + * the child surface remains within the visible area's borders, and to + * specify how the child surface changes its position, such as sliding along + * an axis, or flipping around a rectangle. These positioner-created rules are + * constrained by the requirement that a child surface must intersect with or + * be at least partially adjacent to its parent surface. + * + * See the various requests for details about possible rules. + * + * At the time of the request, the compositor makes a copy of the rules + * specified by the xdg_positioner. Thus, after the request is complete the + * xdg_positioner object can be destroyed or reused; further changes to the + * object will have no effect on previous usages. + * + * For an xdg_positioner object to be considered complete, it must have a + * non-zero size set by set_size, and a non-zero anchor rectangle set by + * set_anchor_rect. Passing an incomplete xdg_positioner object when + * positioning a surface raises an invalid_positioner error. + * @section page_iface_xdg_positioner_api API + * See @ref iface_xdg_positioner. + */ +/** + * @defgroup iface_xdg_positioner The xdg_positioner interface + * + * The xdg_positioner provides a collection of rules for the placement of a + * child surface relative to a parent surface. Rules can be defined to ensure + * the child surface remains within the visible area's borders, and to + * specify how the child surface changes its position, such as sliding along + * an axis, or flipping around a rectangle. These positioner-created rules are + * constrained by the requirement that a child surface must intersect with or + * be at least partially adjacent to its parent surface. + * + * See the various requests for details about possible rules. + * + * At the time of the request, the compositor makes a copy of the rules + * specified by the xdg_positioner. Thus, after the request is complete the + * xdg_positioner object can be destroyed or reused; further changes to the + * object will have no effect on previous usages. + * + * For an xdg_positioner object to be considered complete, it must have a + * non-zero size set by set_size, and a non-zero anchor rectangle set by + * set_anchor_rect. Passing an incomplete xdg_positioner object when + * positioning a surface raises an invalid_positioner error. + */ +extern const struct wl_interface xdg_positioner_interface; +#endif +#ifndef XDG_SURFACE_INTERFACE +#define XDG_SURFACE_INTERFACE +/** + * @page page_iface_xdg_surface xdg_surface + * @section page_iface_xdg_surface_desc Description + * + * An interface that may be implemented by a wl_surface, for + * implementations that provide a desktop-style user interface. + * + * It provides a base set of functionality required to construct user + * interface elements requiring management by the compositor, such as + * toplevel windows, menus, etc. The types of functionality are split into + * xdg_surface roles. + * + * Creating an xdg_surface does not set the role for a wl_surface. In order + * to map an xdg_surface, the client must create a role-specific object + * using, e.g., get_toplevel, get_popup. The wl_surface for any given + * xdg_surface can have at most one role, and may not be assigned any role + * not based on xdg_surface. + * + * A role must be assigned before any other requests are made to the + * xdg_surface object. + * + * The client must call wl_surface.commit on the corresponding wl_surface + * for the xdg_surface state to take effect. + * + * Creating an xdg_surface from a wl_surface which has a buffer attached or + * committed is a client error, and any attempts by a client to attach or + * manipulate a buffer prior to the first xdg_surface.configure call must + * also be treated as errors. + * + * After creating a role-specific object and setting it up (e.g. by sending + * the title, app ID, size constraints, parent, etc), the client must + * perform an initial commit without any buffer attached. The compositor + * will reply with initial wl_surface state such as + * wl_surface.preferred_buffer_scale followed by an xdg_surface.configure + * event. The client must acknowledge it and is then allowed to attach a + * buffer to map the surface. + * + * Mapping an xdg_surface-based role surface is defined as making it + * possible for the surface to be shown by the compositor. Note that + * a mapped surface is not guaranteed to be visible once it is mapped. + * + * For an xdg_surface to be mapped by the compositor, the following + * conditions must be met: + * (1) the client has assigned an xdg_surface-based role to the surface + * (2) the client has set and committed the xdg_surface state and the + * role-dependent state to the surface + * (3) the client has committed a buffer to the surface + * + * A newly-unmapped surface is considered to have met condition (1) out + * of the 3 required conditions for mapping a surface if its role surface + * has not been destroyed, i.e. the client must perform the initial commit + * again before attaching a buffer. + * @section page_iface_xdg_surface_api API + * See @ref iface_xdg_surface. + */ +/** + * @defgroup iface_xdg_surface The xdg_surface interface + * + * An interface that may be implemented by a wl_surface, for + * implementations that provide a desktop-style user interface. + * + * It provides a base set of functionality required to construct user + * interface elements requiring management by the compositor, such as + * toplevel windows, menus, etc. The types of functionality are split into + * xdg_surface roles. + * + * Creating an xdg_surface does not set the role for a wl_surface. In order + * to map an xdg_surface, the client must create a role-specific object + * using, e.g., get_toplevel, get_popup. The wl_surface for any given + * xdg_surface can have at most one role, and may not be assigned any role + * not based on xdg_surface. + * + * A role must be assigned before any other requests are made to the + * xdg_surface object. + * + * The client must call wl_surface.commit on the corresponding wl_surface + * for the xdg_surface state to take effect. + * + * Creating an xdg_surface from a wl_surface which has a buffer attached or + * committed is a client error, and any attempts by a client to attach or + * manipulate a buffer prior to the first xdg_surface.configure call must + * also be treated as errors. + * + * After creating a role-specific object and setting it up (e.g. by sending + * the title, app ID, size constraints, parent, etc), the client must + * perform an initial commit without any buffer attached. The compositor + * will reply with initial wl_surface state such as + * wl_surface.preferred_buffer_scale followed by an xdg_surface.configure + * event. The client must acknowledge it and is then allowed to attach a + * buffer to map the surface. + * + * Mapping an xdg_surface-based role surface is defined as making it + * possible for the surface to be shown by the compositor. Note that + * a mapped surface is not guaranteed to be visible once it is mapped. + * + * For an xdg_surface to be mapped by the compositor, the following + * conditions must be met: + * (1) the client has assigned an xdg_surface-based role to the surface + * (2) the client has set and committed the xdg_surface state and the + * role-dependent state to the surface + * (3) the client has committed a buffer to the surface + * + * A newly-unmapped surface is considered to have met condition (1) out + * of the 3 required conditions for mapping a surface if its role surface + * has not been destroyed, i.e. the client must perform the initial commit + * again before attaching a buffer. + */ +extern const struct wl_interface xdg_surface_interface; +#endif +#ifndef XDG_TOPLEVEL_INTERFACE +#define XDG_TOPLEVEL_INTERFACE +/** + * @page page_iface_xdg_toplevel xdg_toplevel + * @section page_iface_xdg_toplevel_desc Description + * + * This interface defines an xdg_surface role which allows a surface to, + * among other things, set window-like properties such as maximize, + * fullscreen, and minimize, set application-specific metadata like title and + * id, and well as trigger user interactive operations such as interactive + * resize and move. + * + * A xdg_toplevel by default is responsible for providing the full intended + * visual representation of the toplevel, which depending on the window + * state, may mean things like a title bar, window controls and drop shadow. + * + * Unmapping an xdg_toplevel means that the surface cannot be shown + * by the compositor until it is explicitly mapped again. + * All active operations (e.g., move, resize) are canceled and all + * attributes (e.g. title, state, stacking, ...) are discarded for + * an xdg_toplevel surface when it is unmapped. The xdg_toplevel returns to + * the state it had right after xdg_surface.get_toplevel. The client + * can re-map the toplevel by performing a commit without any buffer + * attached, waiting for a configure event and handling it as usual (see + * xdg_surface description). + * + * Attaching a null buffer to a toplevel unmaps the surface. + * @section page_iface_xdg_toplevel_api API + * See @ref iface_xdg_toplevel. + */ +/** + * @defgroup iface_xdg_toplevel The xdg_toplevel interface + * + * This interface defines an xdg_surface role which allows a surface to, + * among other things, set window-like properties such as maximize, + * fullscreen, and minimize, set application-specific metadata like title and + * id, and well as trigger user interactive operations such as interactive + * resize and move. + * + * A xdg_toplevel by default is responsible for providing the full intended + * visual representation of the toplevel, which depending on the window + * state, may mean things like a title bar, window controls and drop shadow. + * + * Unmapping an xdg_toplevel means that the surface cannot be shown + * by the compositor until it is explicitly mapped again. + * All active operations (e.g., move, resize) are canceled and all + * attributes (e.g. title, state, stacking, ...) are discarded for + * an xdg_toplevel surface when it is unmapped. The xdg_toplevel returns to + * the state it had right after xdg_surface.get_toplevel. The client + * can re-map the toplevel by performing a commit without any buffer + * attached, waiting for a configure event and handling it as usual (see + * xdg_surface description). + * + * Attaching a null buffer to a toplevel unmaps the surface. + */ +extern const struct wl_interface xdg_toplevel_interface; +#endif +#ifndef XDG_POPUP_INTERFACE +#define XDG_POPUP_INTERFACE +/** + * @page page_iface_xdg_popup xdg_popup + * @section page_iface_xdg_popup_desc Description + * + * A popup surface is a short-lived, temporary surface. It can be used to + * implement for example menus, popovers, tooltips and other similar user + * interface concepts. + * + * A popup can be made to take an explicit grab. See xdg_popup.grab for + * details. + * + * When the popup is dismissed, a popup_done event will be sent out, and at + * the same time the surface will be unmapped. See the xdg_popup.popup_done + * event for details. + * + * Explicitly destroying the xdg_popup object will also dismiss the popup and + * unmap the surface. Clients that want to dismiss the popup when another + * surface of their own is clicked should dismiss the popup using the destroy + * request. + * + * A newly created xdg_popup will be stacked on top of all previously created + * xdg_popup surfaces associated with the same xdg_toplevel. + * + * The parent of an xdg_popup must be mapped (see the xdg_surface + * description) before the xdg_popup itself. + * + * The client must call wl_surface.commit on the corresponding wl_surface + * for the xdg_popup state to take effect. + * @section page_iface_xdg_popup_api API + * See @ref iface_xdg_popup. + */ +/** + * @defgroup iface_xdg_popup The xdg_popup interface + * + * A popup surface is a short-lived, temporary surface. It can be used to + * implement for example menus, popovers, tooltips and other similar user + * interface concepts. + * + * A popup can be made to take an explicit grab. See xdg_popup.grab for + * details. + * + * When the popup is dismissed, a popup_done event will be sent out, and at + * the same time the surface will be unmapped. See the xdg_popup.popup_done + * event for details. + * + * Explicitly destroying the xdg_popup object will also dismiss the popup and + * unmap the surface. Clients that want to dismiss the popup when another + * surface of their own is clicked should dismiss the popup using the destroy + * request. + * + * A newly created xdg_popup will be stacked on top of all previously created + * xdg_popup surfaces associated with the same xdg_toplevel. + * + * The parent of an xdg_popup must be mapped (see the xdg_surface + * description) before the xdg_popup itself. + * + * The client must call wl_surface.commit on the corresponding wl_surface + * for the xdg_popup state to take effect. + */ +extern const struct wl_interface xdg_popup_interface; +#endif + +#ifndef XDG_WM_BASE_ERROR_ENUM +#define XDG_WM_BASE_ERROR_ENUM +enum xdg_wm_base_error { + /** + * given wl_surface has another role + */ + XDG_WM_BASE_ERROR_ROLE = 0, + /** + * xdg_wm_base was destroyed before children + */ + XDG_WM_BASE_ERROR_DEFUNCT_SURFACES = 1, + /** + * the client tried to map or destroy a non-topmost popup + */ + XDG_WM_BASE_ERROR_NOT_THE_TOPMOST_POPUP = 2, + /** + * the client specified an invalid popup parent surface + */ + XDG_WM_BASE_ERROR_INVALID_POPUP_PARENT = 3, + /** + * the client provided an invalid surface state + */ + XDG_WM_BASE_ERROR_INVALID_SURFACE_STATE = 4, + /** + * the client provided an invalid positioner + */ + XDG_WM_BASE_ERROR_INVALID_POSITIONER = 5, + /** + * the client didn’t respond to a ping event in time + */ + XDG_WM_BASE_ERROR_UNRESPONSIVE = 6, +}; +#endif /* XDG_WM_BASE_ERROR_ENUM */ + +/** + * @ingroup iface_xdg_wm_base + * @struct xdg_wm_base_listener + */ +struct xdg_wm_base_listener { + /** + * check if the client is alive + * + * The ping event asks the client if it's still alive. Pass the + * serial specified in the event back to the compositor by sending + * a "pong" request back with the specified serial. See + * xdg_wm_base.pong. + * + * Compositors can use this to determine if the client is still + * alive. It's unspecified what will happen if the client doesn't + * respond to the ping request, or in what timeframe. Clients + * should try to respond in a reasonable amount of time. The + * “unresponsive” error is provided for compositors that wish + * to disconnect unresponsive clients. + * + * A compositor is free to ping in any way it wants, but a client + * must always respond to any xdg_wm_base object it created. + * @param serial pass this to the pong request + */ + void (*ping)(void *data, + struct xdg_wm_base *xdg_wm_base, + uint32_t serial); +}; + +/** + * @ingroup iface_xdg_wm_base + */ +static inline int +xdg_wm_base_add_listener(struct xdg_wm_base *xdg_wm_base, + const struct xdg_wm_base_listener *listener, void *data) +{ + return wl_proxy_add_listener((struct wl_proxy *) xdg_wm_base, + (void (**)(void)) listener, data); +} + +#define XDG_WM_BASE_DESTROY 0 +#define XDG_WM_BASE_CREATE_POSITIONER 1 +#define XDG_WM_BASE_GET_XDG_SURFACE 2 +#define XDG_WM_BASE_PONG 3 + +/** + * @ingroup iface_xdg_wm_base + */ +#define XDG_WM_BASE_PING_SINCE_VERSION 1 + +/** + * @ingroup iface_xdg_wm_base + */ +#define XDG_WM_BASE_DESTROY_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_wm_base + */ +#define XDG_WM_BASE_CREATE_POSITIONER_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_wm_base + */ +#define XDG_WM_BASE_GET_XDG_SURFACE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_wm_base + */ +#define XDG_WM_BASE_PONG_SINCE_VERSION 1 + +/** @ingroup iface_xdg_wm_base */ +static inline void +xdg_wm_base_set_user_data(struct xdg_wm_base *xdg_wm_base, void *user_data) +{ + wl_proxy_set_user_data((struct wl_proxy *) xdg_wm_base, user_data); +} + +/** @ingroup iface_xdg_wm_base */ +static inline void * +xdg_wm_base_get_user_data(struct xdg_wm_base *xdg_wm_base) +{ + return wl_proxy_get_user_data((struct wl_proxy *) xdg_wm_base); +} + +static inline uint32_t +xdg_wm_base_get_version(struct xdg_wm_base *xdg_wm_base) +{ + return wl_proxy_get_version((struct wl_proxy *) xdg_wm_base); +} + +/** + * @ingroup iface_xdg_wm_base + * + * Destroy this xdg_wm_base object. + * + * Destroying a bound xdg_wm_base object while there are surfaces + * still alive created by this xdg_wm_base object instance is illegal + * and will result in a defunct_surfaces error. + */ +static inline void +xdg_wm_base_destroy(struct xdg_wm_base *xdg_wm_base) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_wm_base, + XDG_WM_BASE_DESTROY, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_wm_base), WL_MARSHAL_FLAG_DESTROY); +} + +/** + * @ingroup iface_xdg_wm_base + * + * Create a positioner object. A positioner object is used to position + * surfaces relative to some parent surface. See the interface description + * and xdg_surface.get_popup for details. + */ +static inline struct xdg_positioner * +xdg_wm_base_create_positioner(struct xdg_wm_base *xdg_wm_base) +{ + struct wl_proxy *id; + + id = wl_proxy_marshal_flags((struct wl_proxy *) xdg_wm_base, + XDG_WM_BASE_CREATE_POSITIONER, &xdg_positioner_interface, wl_proxy_get_version((struct wl_proxy *) xdg_wm_base), 0, NULL); + + return (struct xdg_positioner *) id; +} + +/** + * @ingroup iface_xdg_wm_base + * + * This creates an xdg_surface for the given surface. While xdg_surface + * itself is not a role, the corresponding surface may only be assigned + * a role extending xdg_surface, such as xdg_toplevel or xdg_popup. It is + * illegal to create an xdg_surface for a wl_surface which already has an + * assigned role and this will result in a role error. + * + * This creates an xdg_surface for the given surface. An xdg_surface is + * used as basis to define a role to a given surface, such as xdg_toplevel + * or xdg_popup. It also manages functionality shared between xdg_surface + * based surface roles. + * + * See the documentation of xdg_surface for more details about what an + * xdg_surface is and how it is used. + */ +static inline struct xdg_surface * +xdg_wm_base_get_xdg_surface(struct xdg_wm_base *xdg_wm_base, struct wl_surface *surface) +{ + struct wl_proxy *id; + + id = wl_proxy_marshal_flags((struct wl_proxy *) xdg_wm_base, + XDG_WM_BASE_GET_XDG_SURFACE, &xdg_surface_interface, wl_proxy_get_version((struct wl_proxy *) xdg_wm_base), 0, NULL, surface); + + return (struct xdg_surface *) id; +} + +/** + * @ingroup iface_xdg_wm_base + * + * A client must respond to a ping event with a pong request or + * the client may be deemed unresponsive. See xdg_wm_base.ping + * and xdg_wm_base.error.unresponsive. + */ +static inline void +xdg_wm_base_pong(struct xdg_wm_base *xdg_wm_base, uint32_t serial) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_wm_base, + XDG_WM_BASE_PONG, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_wm_base), 0, serial); +} + +#ifndef XDG_POSITIONER_ERROR_ENUM +#define XDG_POSITIONER_ERROR_ENUM +enum xdg_positioner_error { + /** + * invalid input provided + */ + XDG_POSITIONER_ERROR_INVALID_INPUT = 0, +}; +#endif /* XDG_POSITIONER_ERROR_ENUM */ + +#ifndef XDG_POSITIONER_ANCHOR_ENUM +#define XDG_POSITIONER_ANCHOR_ENUM +enum xdg_positioner_anchor { + XDG_POSITIONER_ANCHOR_NONE = 0, + XDG_POSITIONER_ANCHOR_TOP = 1, + XDG_POSITIONER_ANCHOR_BOTTOM = 2, + XDG_POSITIONER_ANCHOR_LEFT = 3, + XDG_POSITIONER_ANCHOR_RIGHT = 4, + XDG_POSITIONER_ANCHOR_TOP_LEFT = 5, + XDG_POSITIONER_ANCHOR_BOTTOM_LEFT = 6, + XDG_POSITIONER_ANCHOR_TOP_RIGHT = 7, + XDG_POSITIONER_ANCHOR_BOTTOM_RIGHT = 8, +}; +#endif /* XDG_POSITIONER_ANCHOR_ENUM */ + +#ifndef XDG_POSITIONER_GRAVITY_ENUM +#define XDG_POSITIONER_GRAVITY_ENUM +enum xdg_positioner_gravity { + XDG_POSITIONER_GRAVITY_NONE = 0, + XDG_POSITIONER_GRAVITY_TOP = 1, + XDG_POSITIONER_GRAVITY_BOTTOM = 2, + XDG_POSITIONER_GRAVITY_LEFT = 3, + XDG_POSITIONER_GRAVITY_RIGHT = 4, + XDG_POSITIONER_GRAVITY_TOP_LEFT = 5, + XDG_POSITIONER_GRAVITY_BOTTOM_LEFT = 6, + XDG_POSITIONER_GRAVITY_TOP_RIGHT = 7, + XDG_POSITIONER_GRAVITY_BOTTOM_RIGHT = 8, +}; +#endif /* XDG_POSITIONER_GRAVITY_ENUM */ + +#ifndef XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_ENUM +#define XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_ENUM +/** + * @ingroup iface_xdg_positioner + * constraint adjustments + * + * The constraint adjustment value define ways the compositor will adjust + * the position of the surface, if the unadjusted position would result + * in the surface being partly constrained. + * + * Whether a surface is considered 'constrained' is left to the compositor + * to determine. For example, the surface may be partly outside the + * compositor's defined 'work area', thus necessitating the child surface's + * position be adjusted until it is entirely inside the work area. + * + * The adjustments can be combined, according to a defined precedence: 1) + * Flip, 2) Slide, 3) Resize. + */ +enum xdg_positioner_constraint_adjustment { + /** + * don't move the child surface when constrained + * + * Don't alter the surface position even if it is constrained on + * some axis, for example partially outside the edge of an output. + */ + XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_NONE = 0, + /** + * move along the x axis until unconstrained + * + * Slide the surface along the x axis until it is no longer + * constrained. + * + * First try to slide towards the direction of the gravity on the x + * axis until either the edge in the opposite direction of the + * gravity is unconstrained or the edge in the direction of the + * gravity is constrained. + * + * Then try to slide towards the opposite direction of the gravity + * on the x axis until either the edge in the direction of the + * gravity is unconstrained or the edge in the opposite direction + * of the gravity is constrained. + */ + XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_SLIDE_X = 1, + /** + * move along the y axis until unconstrained + * + * Slide the surface along the y axis until it is no longer + * constrained. + * + * First try to slide towards the direction of the gravity on the y + * axis until either the edge in the opposite direction of the + * gravity is unconstrained or the edge in the direction of the + * gravity is constrained. + * + * Then try to slide towards the opposite direction of the gravity + * on the y axis until either the edge in the direction of the + * gravity is unconstrained or the edge in the opposite direction + * of the gravity is constrained. + */ + XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_SLIDE_Y = 2, + /** + * invert the anchor and gravity on the x axis + * + * Invert the anchor and gravity on the x axis if the surface is + * constrained on the x axis. For example, if the left edge of the + * surface is constrained, the gravity is 'left' and the anchor is + * 'left', change the gravity to 'right' and the anchor to 'right'. + * + * If the adjusted position also ends up being constrained, the + * resulting position of the flip_x adjustment will be the one + * before the adjustment. + */ + XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_FLIP_X = 4, + /** + * invert the anchor and gravity on the y axis + * + * Invert the anchor and gravity on the y axis if the surface is + * constrained on the y axis. For example, if the bottom edge of + * the surface is constrained, the gravity is 'bottom' and the + * anchor is 'bottom', change the gravity to 'top' and the anchor + * to 'top'. + * + * The adjusted position is calculated given the original anchor + * rectangle and offset, but with the new flipped anchor and + * gravity values. + * + * If the adjusted position also ends up being constrained, the + * resulting position of the flip_y adjustment will be the one + * before the adjustment. + */ + XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_FLIP_Y = 8, + /** + * horizontally resize the surface + * + * Resize the surface horizontally so that it is completely + * unconstrained. + */ + XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_RESIZE_X = 16, + /** + * vertically resize the surface + * + * Resize the surface vertically so that it is completely + * unconstrained. + */ + XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_RESIZE_Y = 32, +}; +#endif /* XDG_POSITIONER_CONSTRAINT_ADJUSTMENT_ENUM */ + +#define XDG_POSITIONER_DESTROY 0 +#define XDG_POSITIONER_SET_SIZE 1 +#define XDG_POSITIONER_SET_ANCHOR_RECT 2 +#define XDG_POSITIONER_SET_ANCHOR 3 +#define XDG_POSITIONER_SET_GRAVITY 4 +#define XDG_POSITIONER_SET_CONSTRAINT_ADJUSTMENT 5 +#define XDG_POSITIONER_SET_OFFSET 6 +#define XDG_POSITIONER_SET_REACTIVE 7 +#define XDG_POSITIONER_SET_PARENT_SIZE 8 +#define XDG_POSITIONER_SET_PARENT_CONFIGURE 9 + + +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_DESTROY_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_SIZE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_ANCHOR_RECT_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_ANCHOR_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_GRAVITY_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_CONSTRAINT_ADJUSTMENT_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_OFFSET_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_REACTIVE_SINCE_VERSION 3 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_PARENT_SIZE_SINCE_VERSION 3 +/** + * @ingroup iface_xdg_positioner + */ +#define XDG_POSITIONER_SET_PARENT_CONFIGURE_SINCE_VERSION 3 + +/** @ingroup iface_xdg_positioner */ +static inline void +xdg_positioner_set_user_data(struct xdg_positioner *xdg_positioner, void *user_data) +{ + wl_proxy_set_user_data((struct wl_proxy *) xdg_positioner, user_data); +} + +/** @ingroup iface_xdg_positioner */ +static inline void * +xdg_positioner_get_user_data(struct xdg_positioner *xdg_positioner) +{ + return wl_proxy_get_user_data((struct wl_proxy *) xdg_positioner); +} + +static inline uint32_t +xdg_positioner_get_version(struct xdg_positioner *xdg_positioner) +{ + return wl_proxy_get_version((struct wl_proxy *) xdg_positioner); +} + +/** + * @ingroup iface_xdg_positioner + * + * Notify the compositor that the xdg_positioner will no longer be used. + */ +static inline void +xdg_positioner_destroy(struct xdg_positioner *xdg_positioner) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_DESTROY, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), WL_MARSHAL_FLAG_DESTROY); +} + +/** + * @ingroup iface_xdg_positioner + * + * Set the size of the surface that is to be positioned with the positioner + * object. The size is in surface-local coordinates and corresponds to the + * window geometry. See xdg_surface.set_window_geometry. + * + * If a zero or negative size is set the invalid_input error is raised. + */ +static inline void +xdg_positioner_set_size(struct xdg_positioner *xdg_positioner, int32_t width, int32_t height) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_SIZE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0, width, height); +} + +/** + * @ingroup iface_xdg_positioner + * + * Specify the anchor rectangle within the parent surface that the child + * surface will be placed relative to. The rectangle is relative to the + * window geometry as defined by xdg_surface.set_window_geometry of the + * parent surface. + * + * When the xdg_positioner object is used to position a child surface, the + * anchor rectangle may not extend outside the window geometry of the + * positioned child's parent surface. + * + * If a negative size is set the invalid_input error is raised. + */ +static inline void +xdg_positioner_set_anchor_rect(struct xdg_positioner *xdg_positioner, int32_t x, int32_t y, int32_t width, int32_t height) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_ANCHOR_RECT, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0, x, y, width, height); +} + +/** + * @ingroup iface_xdg_positioner + * + * Defines the anchor point for the anchor rectangle. The specified anchor + * is used derive an anchor point that the child surface will be + * positioned relative to. If a corner anchor is set (e.g. 'top_left' or + * 'bottom_right'), the anchor point will be at the specified corner; + * otherwise, the derived anchor point will be centered on the specified + * edge, or in the center of the anchor rectangle if no edge is specified. + */ +static inline void +xdg_positioner_set_anchor(struct xdg_positioner *xdg_positioner, uint32_t anchor) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_ANCHOR, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0, anchor); +} + +/** + * @ingroup iface_xdg_positioner + * + * Defines in what direction a surface should be positioned, relative to + * the anchor point of the parent surface. If a corner gravity is + * specified (e.g. 'bottom_right' or 'top_left'), then the child surface + * will be placed towards the specified gravity; otherwise, the child + * surface will be centered over the anchor point on any axis that had no + * gravity specified. If the gravity is not in the ‘gravity’ enum, an + * invalid_input error is raised. + */ +static inline void +xdg_positioner_set_gravity(struct xdg_positioner *xdg_positioner, uint32_t gravity) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_GRAVITY, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0, gravity); +} + +/** + * @ingroup iface_xdg_positioner + * + * Specify how the window should be positioned if the originally intended + * position caused the surface to be constrained, meaning at least + * partially outside positioning boundaries set by the compositor. The + * adjustment is set by constructing a bitmask describing the adjustment to + * be made when the surface is constrained on that axis. + * + * If no bit for one axis is set, the compositor will assume that the child + * surface should not change its position on that axis when constrained. + * + * If more than one bit for one axis is set, the order of how adjustments + * are applied is specified in the corresponding adjustment descriptions. + * + * The default adjustment is none. + */ +static inline void +xdg_positioner_set_constraint_adjustment(struct xdg_positioner *xdg_positioner, uint32_t constraint_adjustment) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_CONSTRAINT_ADJUSTMENT, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0, constraint_adjustment); +} + +/** + * @ingroup iface_xdg_positioner + * + * Specify the surface position offset relative to the position of the + * anchor on the anchor rectangle and the anchor on the surface. For + * example if the anchor of the anchor rectangle is at (x, y), the surface + * has the gravity bottom|right, and the offset is (ox, oy), the calculated + * surface position will be (x + ox, y + oy). The offset position of the + * surface is the one used for constraint testing. See + * set_constraint_adjustment. + * + * An example use case is placing a popup menu on top of a user interface + * element, while aligning the user interface element of the parent surface + * with some user interface element placed somewhere in the popup surface. + */ +static inline void +xdg_positioner_set_offset(struct xdg_positioner *xdg_positioner, int32_t x, int32_t y) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_OFFSET, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0, x, y); +} + +/** + * @ingroup iface_xdg_positioner + * + * When set reactive, the surface is reconstrained if the conditions used + * for constraining changed, e.g. the parent window moved. + * + * If the conditions changed and the popup was reconstrained, an + * xdg_popup.configure event is sent with updated geometry, followed by an + * xdg_surface.configure event. + */ +static inline void +xdg_positioner_set_reactive(struct xdg_positioner *xdg_positioner) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_REACTIVE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0); +} + +/** + * @ingroup iface_xdg_positioner + * + * Set the parent window geometry the compositor should use when + * positioning the popup. The compositor may use this information to + * determine the future state the popup should be constrained using. If + * this doesn't match the dimension of the parent the popup is eventually + * positioned against, the behavior is undefined. + * + * The arguments are given in the surface-local coordinate space. + */ +static inline void +xdg_positioner_set_parent_size(struct xdg_positioner *xdg_positioner, int32_t parent_width, int32_t parent_height) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_PARENT_SIZE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0, parent_width, parent_height); +} + +/** + * @ingroup iface_xdg_positioner + * + * Set the serial of an xdg_surface.configure event this positioner will be + * used in response to. The compositor may use this information together + * with set_parent_size to determine what future state the popup should be + * constrained using. + */ +static inline void +xdg_positioner_set_parent_configure(struct xdg_positioner *xdg_positioner, uint32_t serial) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_positioner, + XDG_POSITIONER_SET_PARENT_CONFIGURE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_positioner), 0, serial); +} + +#ifndef XDG_SURFACE_ERROR_ENUM +#define XDG_SURFACE_ERROR_ENUM +enum xdg_surface_error { + /** + * Surface was not fully constructed + */ + XDG_SURFACE_ERROR_NOT_CONSTRUCTED = 1, + /** + * Surface was already constructed + */ + XDG_SURFACE_ERROR_ALREADY_CONSTRUCTED = 2, + /** + * Attaching a buffer to an unconfigured surface + */ + XDG_SURFACE_ERROR_UNCONFIGURED_BUFFER = 3, + /** + * Invalid serial number when acking a configure event + */ + XDG_SURFACE_ERROR_INVALID_SERIAL = 4, + /** + * Width or height was zero or negative + */ + XDG_SURFACE_ERROR_INVALID_SIZE = 5, + /** + * Surface was destroyed before its role object + */ + XDG_SURFACE_ERROR_DEFUNCT_ROLE_OBJECT = 6, +}; +#endif /* XDG_SURFACE_ERROR_ENUM */ + +/** + * @ingroup iface_xdg_surface + * @struct xdg_surface_listener + */ +struct xdg_surface_listener { + /** + * suggest a surface change + * + * The configure event marks the end of a configure sequence. A + * configure sequence is a set of one or more events configuring + * the state of the xdg_surface, including the final + * xdg_surface.configure event. + * + * Where applicable, xdg_surface surface roles will during a + * configure sequence extend this event as a latched state sent as + * events before the xdg_surface.configure event. Such events + * should be considered to make up a set of atomically applied + * configuration states, where the xdg_surface.configure commits + * the accumulated state. + * + * Clients should arrange their surface for the new states, and + * then send an ack_configure request with the serial sent in this + * configure event at some point before committing the new surface. + * + * If the client receives multiple configure events before it can + * respond to one, it is free to discard all but the last event it + * received. + * @param serial serial of the configure event + */ + void (*configure)(void *data, + struct xdg_surface *xdg_surface, + uint32_t serial); +}; + +/** + * @ingroup iface_xdg_surface + */ +static inline int +xdg_surface_add_listener(struct xdg_surface *xdg_surface, + const struct xdg_surface_listener *listener, void *data) +{ + return wl_proxy_add_listener((struct wl_proxy *) xdg_surface, + (void (**)(void)) listener, data); +} + +#define XDG_SURFACE_DESTROY 0 +#define XDG_SURFACE_GET_TOPLEVEL 1 +#define XDG_SURFACE_GET_POPUP 2 +#define XDG_SURFACE_SET_WINDOW_GEOMETRY 3 +#define XDG_SURFACE_ACK_CONFIGURE 4 + +/** + * @ingroup iface_xdg_surface + */ +#define XDG_SURFACE_CONFIGURE_SINCE_VERSION 1 + +/** + * @ingroup iface_xdg_surface + */ +#define XDG_SURFACE_DESTROY_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_surface + */ +#define XDG_SURFACE_GET_TOPLEVEL_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_surface + */ +#define XDG_SURFACE_GET_POPUP_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_surface + */ +#define XDG_SURFACE_SET_WINDOW_GEOMETRY_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_surface + */ +#define XDG_SURFACE_ACK_CONFIGURE_SINCE_VERSION 1 + +/** @ingroup iface_xdg_surface */ +static inline void +xdg_surface_set_user_data(struct xdg_surface *xdg_surface, void *user_data) +{ + wl_proxy_set_user_data((struct wl_proxy *) xdg_surface, user_data); +} + +/** @ingroup iface_xdg_surface */ +static inline void * +xdg_surface_get_user_data(struct xdg_surface *xdg_surface) +{ + return wl_proxy_get_user_data((struct wl_proxy *) xdg_surface); +} + +static inline uint32_t +xdg_surface_get_version(struct xdg_surface *xdg_surface) +{ + return wl_proxy_get_version((struct wl_proxy *) xdg_surface); +} + +/** + * @ingroup iface_xdg_surface + * + * Destroy the xdg_surface object. An xdg_surface must only be destroyed + * after its role object has been destroyed, otherwise + * a defunct_role_object error is raised. + */ +static inline void +xdg_surface_destroy(struct xdg_surface *xdg_surface) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_surface, + XDG_SURFACE_DESTROY, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_surface), WL_MARSHAL_FLAG_DESTROY); +} + +/** + * @ingroup iface_xdg_surface + * + * This creates an xdg_toplevel object for the given xdg_surface and gives + * the associated wl_surface the xdg_toplevel role. + * + * See the documentation of xdg_toplevel for more details about what an + * xdg_toplevel is and how it is used. + */ +static inline struct xdg_toplevel * +xdg_surface_get_toplevel(struct xdg_surface *xdg_surface) +{ + struct wl_proxy *id; + + id = wl_proxy_marshal_flags((struct wl_proxy *) xdg_surface, + XDG_SURFACE_GET_TOPLEVEL, &xdg_toplevel_interface, wl_proxy_get_version((struct wl_proxy *) xdg_surface), 0, NULL); + + return (struct xdg_toplevel *) id; +} + +/** + * @ingroup iface_xdg_surface + * + * This creates an xdg_popup object for the given xdg_surface and gives + * the associated wl_surface the xdg_popup role. + * + * If null is passed as a parent, a parent surface must be specified using + * some other protocol, before committing the initial state. + * + * See the documentation of xdg_popup for more details about what an + * xdg_popup is and how it is used. + */ +static inline struct xdg_popup * +xdg_surface_get_popup(struct xdg_surface *xdg_surface, struct xdg_surface *parent, struct xdg_positioner *positioner) +{ + struct wl_proxy *id; + + id = wl_proxy_marshal_flags((struct wl_proxy *) xdg_surface, + XDG_SURFACE_GET_POPUP, &xdg_popup_interface, wl_proxy_get_version((struct wl_proxy *) xdg_surface), 0, NULL, parent, positioner); + + return (struct xdg_popup *) id; +} + +/** + * @ingroup iface_xdg_surface + * + * The window geometry of a surface is its "visible bounds" from the + * user's perspective. Client-side decorations often have invisible + * portions like drop-shadows which should be ignored for the + * purposes of aligning, placing and constraining windows. + * + * The window geometry is double-buffered state, see wl_surface.commit. + * + * When maintaining a position, the compositor should treat the (x, y) + * coordinate of the window geometry as the top left corner of the window. + * A client changing the (x, y) window geometry coordinate should in + * general not alter the position of the window. + * + * Once the window geometry of the surface is set, it is not possible to + * unset it, and it will remain the same until set_window_geometry is + * called again, even if a new subsurface or buffer is attached. + * + * If never set, the value is the full bounds of the surface, + * including any subsurfaces. This updates dynamically on every + * commit. This unset is meant for extremely simple clients. + * + * The arguments are given in the surface-local coordinate space of + * the wl_surface associated with this xdg_surface, and may extend outside + * of the wl_surface itself to mark parts of the subsurface tree as part of + * the window geometry. + * + * When applied, the effective window geometry will be the set window + * geometry clamped to the bounding rectangle of the combined + * geometry of the surface of the xdg_surface and the associated + * subsurfaces. + * + * The effective geometry will not be recalculated unless a new call to + * set_window_geometry is done and the new pending surface state is + * subsequently applied. + * + * The width and height of the effective window geometry must be + * greater than zero. Setting an invalid size will raise an + * invalid_size error. + */ +static inline void +xdg_surface_set_window_geometry(struct xdg_surface *xdg_surface, int32_t x, int32_t y, int32_t width, int32_t height) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_surface, + XDG_SURFACE_SET_WINDOW_GEOMETRY, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_surface), 0, x, y, width, height); +} + +/** + * @ingroup iface_xdg_surface + * + * When a configure event is received, if a client commits the + * surface in response to the configure event, then the client + * must make an ack_configure request sometime before the commit + * request, passing along the serial of the configure event. + * + * For instance, for toplevel surfaces the compositor might use this + * information to move a surface to the top left only when the client has + * drawn itself for the maximized or fullscreen state. + * + * If the client receives multiple configure events before it + * can respond to one, it only has to ack the last configure event. + * Acking a configure event that was never sent raises an invalid_serial + * error. + * + * A client is not required to commit immediately after sending + * an ack_configure request - it may even ack_configure several times + * before its next surface commit. + * + * A client may send multiple ack_configure requests before committing, but + * only the last request sent before a commit indicates which configure + * event the client really is responding to. + * + * Sending an ack_configure request consumes the serial number sent with + * the request, as well as serial numbers sent by all configure events + * sent on this xdg_surface prior to the configure event referenced by + * the committed serial. + * + * It is an error to issue multiple ack_configure requests referencing a + * serial from the same configure event, or to issue an ack_configure + * request referencing a serial from a configure event issued before the + * event identified by the last ack_configure request for the same + * xdg_surface. Doing so will raise an invalid_serial error. + */ +static inline void +xdg_surface_ack_configure(struct xdg_surface *xdg_surface, uint32_t serial) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_surface, + XDG_SURFACE_ACK_CONFIGURE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_surface), 0, serial); +} + +#ifndef XDG_TOPLEVEL_ERROR_ENUM +#define XDG_TOPLEVEL_ERROR_ENUM +enum xdg_toplevel_error { + /** + * provided value is not a valid variant of the resize_edge enum + */ + XDG_TOPLEVEL_ERROR_INVALID_RESIZE_EDGE = 0, + /** + * invalid parent toplevel + */ + XDG_TOPLEVEL_ERROR_INVALID_PARENT = 1, + /** + * client provided an invalid min or max size + */ + XDG_TOPLEVEL_ERROR_INVALID_SIZE = 2, +}; +#endif /* XDG_TOPLEVEL_ERROR_ENUM */ + +#ifndef XDG_TOPLEVEL_RESIZE_EDGE_ENUM +#define XDG_TOPLEVEL_RESIZE_EDGE_ENUM +/** + * @ingroup iface_xdg_toplevel + * edge values for resizing + * + * These values are used to indicate which edge of a surface + * is being dragged in a resize operation. + */ +enum xdg_toplevel_resize_edge { + XDG_TOPLEVEL_RESIZE_EDGE_NONE = 0, + XDG_TOPLEVEL_RESIZE_EDGE_TOP = 1, + XDG_TOPLEVEL_RESIZE_EDGE_BOTTOM = 2, + XDG_TOPLEVEL_RESIZE_EDGE_LEFT = 4, + XDG_TOPLEVEL_RESIZE_EDGE_TOP_LEFT = 5, + XDG_TOPLEVEL_RESIZE_EDGE_BOTTOM_LEFT = 6, + XDG_TOPLEVEL_RESIZE_EDGE_RIGHT = 8, + XDG_TOPLEVEL_RESIZE_EDGE_TOP_RIGHT = 9, + XDG_TOPLEVEL_RESIZE_EDGE_BOTTOM_RIGHT = 10, +}; +#endif /* XDG_TOPLEVEL_RESIZE_EDGE_ENUM */ + +#ifndef XDG_TOPLEVEL_STATE_ENUM +#define XDG_TOPLEVEL_STATE_ENUM +/** + * @ingroup iface_xdg_toplevel + * types of state on the surface + * + * The different state values used on the surface. This is designed for + * state values like maximized, fullscreen. It is paired with the + * configure event to ensure that both the client and the compositor + * setting the state can be synchronized. + * + * States set in this way are double-buffered, see wl_surface.commit. + */ +enum xdg_toplevel_state { + /** + * the surface is maximized + * the surface is maximized + * + * The surface is maximized. The window geometry specified in the + * configure event must be obeyed by the client, or the + * xdg_wm_base.invalid_surface_state error is raised. + * + * The client should draw without shadow or other decoration + * outside of the window geometry. + */ + XDG_TOPLEVEL_STATE_MAXIMIZED = 1, + /** + * the surface is fullscreen + * the surface is fullscreen + * + * The surface is fullscreen. The window geometry specified in + * the configure event is a maximum; the client cannot resize + * beyond it. For a surface to cover the whole fullscreened area, + * the geometry dimensions must be obeyed by the client. For more + * details, see xdg_toplevel.set_fullscreen. + */ + XDG_TOPLEVEL_STATE_FULLSCREEN = 2, + /** + * the surface is being resized + * the surface is being resized + * + * The surface is being resized. The window geometry specified in + * the configure event is a maximum; the client cannot resize + * beyond it. Clients that have aspect ratio or cell sizing + * configuration can use a smaller size, however. + */ + XDG_TOPLEVEL_STATE_RESIZING = 3, + /** + * the surface is now activated + * the surface is now activated + * + * Client window decorations should be painted as if the window + * is active. Do not assume this means that the window actually has + * keyboard or pointer focus. + */ + XDG_TOPLEVEL_STATE_ACTIVATED = 4, + /** + * the surface’s left edge is tiled + * + * The window is currently in a tiled layout and the left edge is + * considered to be adjacent to another part of the tiling grid. + * + * The client should draw without shadow or other decoration + * outside of the window geometry on the left edge. + * @since 2 + */ + XDG_TOPLEVEL_STATE_TILED_LEFT = 5, + /** + * the surface’s right edge is tiled + * + * The window is currently in a tiled layout and the right edge + * is considered to be adjacent to another part of the tiling grid. + * + * The client should draw without shadow or other decoration + * outside of the window geometry on the right edge. + * @since 2 + */ + XDG_TOPLEVEL_STATE_TILED_RIGHT = 6, + /** + * the surface’s top edge is tiled + * + * The window is currently in a tiled layout and the top edge is + * considered to be adjacent to another part of the tiling grid. + * + * The client should draw without shadow or other decoration + * outside of the window geometry on the top edge. + * @since 2 + */ + XDG_TOPLEVEL_STATE_TILED_TOP = 7, + /** + * the surface’s bottom edge is tiled + * + * The window is currently in a tiled layout and the bottom edge + * is considered to be adjacent to another part of the tiling grid. + * + * The client should draw without shadow or other decoration + * outside of the window geometry on the bottom edge. + * @since 2 + */ + XDG_TOPLEVEL_STATE_TILED_BOTTOM = 8, + /** + * surface repaint is suspended + * + * The surface is currently not ordinarily being repainted; for + * example because its content is occluded by another window, or + * its outputs are switched off due to screen locking. + * @since 6 + */ + XDG_TOPLEVEL_STATE_SUSPENDED = 9, +}; +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_STATE_TILED_LEFT_SINCE_VERSION 2 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_STATE_TILED_RIGHT_SINCE_VERSION 2 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_STATE_TILED_TOP_SINCE_VERSION 2 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_STATE_TILED_BOTTOM_SINCE_VERSION 2 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_STATE_SUSPENDED_SINCE_VERSION 6 +#endif /* XDG_TOPLEVEL_STATE_ENUM */ + +#ifndef XDG_TOPLEVEL_WM_CAPABILITIES_ENUM +#define XDG_TOPLEVEL_WM_CAPABILITIES_ENUM +enum xdg_toplevel_wm_capabilities { + /** + * show_window_menu is available + */ + XDG_TOPLEVEL_WM_CAPABILITIES_WINDOW_MENU = 1, + /** + * set_maximized and unset_maximized are available + */ + XDG_TOPLEVEL_WM_CAPABILITIES_MAXIMIZE = 2, + /** + * set_fullscreen and unset_fullscreen are available + */ + XDG_TOPLEVEL_WM_CAPABILITIES_FULLSCREEN = 3, + /** + * set_minimized is available + */ + XDG_TOPLEVEL_WM_CAPABILITIES_MINIMIZE = 4, +}; +#endif /* XDG_TOPLEVEL_WM_CAPABILITIES_ENUM */ + +/** + * @ingroup iface_xdg_toplevel + * @struct xdg_toplevel_listener + */ +struct xdg_toplevel_listener { + /** + * suggest a surface change + * + * This configure event asks the client to resize its toplevel + * surface or to change its state. The configured state should not + * be applied immediately. See xdg_surface.configure for details. + * + * The width and height arguments specify a hint to the window + * about how its surface should be resized in window geometry + * coordinates. See set_window_geometry. + * + * If the width or height arguments are zero, it means the client + * should decide its own window dimension. This may happen when the + * compositor needs to configure the state of the surface but + * doesn't have any information about any previous or expected + * dimension. + * + * The states listed in the event specify how the width/height + * arguments should be interpreted, and possibly how it should be + * drawn. + * + * Clients must send an ack_configure in response to this event. + * See xdg_surface.configure and xdg_surface.ack_configure for + * details. + */ + void (*configure)(void *data, + struct xdg_toplevel *xdg_toplevel, + int32_t width, + int32_t height, + struct wl_array *states); + /** + * surface wants to be closed + * + * The close event is sent by the compositor when the user wants + * the surface to be closed. This should be equivalent to the user + * clicking the close button in client-side decorations, if your + * application has any. + * + * This is only a request that the user intends to close the + * window. The client may choose to ignore this request, or show a + * dialog to ask the user to save their data, etc. + */ + void (*close)(void *data, + struct xdg_toplevel *xdg_toplevel); + /** + * recommended window geometry bounds + * + * The configure_bounds event may be sent prior to a + * xdg_toplevel.configure event to communicate the bounds a window + * geometry size is recommended to constrain to. + * + * The passed width and height are in surface coordinate space. If + * width and height are 0, it means bounds is unknown and + * equivalent to as if no configure_bounds event was ever sent for + * this surface. + * + * The bounds can for example correspond to the size of a monitor + * excluding any panels or other shell components, so that a + * surface isn't created in a way that it cannot fit. + * + * The bounds may change at any point, and in such a case, a new + * xdg_toplevel.configure_bounds will be sent, followed by + * xdg_toplevel.configure and xdg_surface.configure. + * @since 4 + */ + void (*configure_bounds)(void *data, + struct xdg_toplevel *xdg_toplevel, + int32_t width, + int32_t height); + /** + * compositor capabilities + * + * This event advertises the capabilities supported by the + * compositor. If a capability isn't supported, clients should hide + * or disable the UI elements that expose this functionality. For + * instance, if the compositor doesn't advertise support for + * minimized toplevels, a button triggering the set_minimized + * request should not be displayed. + * + * The compositor will ignore requests it doesn't support. For + * instance, a compositor which doesn't advertise support for + * minimized will ignore set_minimized requests. + * + * Compositors must send this event once before the first + * xdg_surface.configure event. When the capabilities change, + * compositors must send this event again and then send an + * xdg_surface.configure event. + * + * The configured state should not be applied immediately. See + * xdg_surface.configure for details. + * + * The capabilities are sent as an array of 32-bit unsigned + * integers in native endianness. + * @param capabilities array of 32-bit capabilities + * @since 5 + */ + void (*wm_capabilities)(void *data, + struct xdg_toplevel *xdg_toplevel, + struct wl_array *capabilities); +}; + +/** + * @ingroup iface_xdg_toplevel + */ +static inline int +xdg_toplevel_add_listener(struct xdg_toplevel *xdg_toplevel, + const struct xdg_toplevel_listener *listener, void *data) +{ + return wl_proxy_add_listener((struct wl_proxy *) xdg_toplevel, + (void (**)(void)) listener, data); +} + +#define XDG_TOPLEVEL_DESTROY 0 +#define XDG_TOPLEVEL_SET_PARENT 1 +#define XDG_TOPLEVEL_SET_TITLE 2 +#define XDG_TOPLEVEL_SET_APP_ID 3 +#define XDG_TOPLEVEL_SHOW_WINDOW_MENU 4 +#define XDG_TOPLEVEL_MOVE 5 +#define XDG_TOPLEVEL_RESIZE 6 +#define XDG_TOPLEVEL_SET_MAX_SIZE 7 +#define XDG_TOPLEVEL_SET_MIN_SIZE 8 +#define XDG_TOPLEVEL_SET_MAXIMIZED 9 +#define XDG_TOPLEVEL_UNSET_MAXIMIZED 10 +#define XDG_TOPLEVEL_SET_FULLSCREEN 11 +#define XDG_TOPLEVEL_UNSET_FULLSCREEN 12 +#define XDG_TOPLEVEL_SET_MINIMIZED 13 + +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_CONFIGURE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_CLOSE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_CONFIGURE_BOUNDS_SINCE_VERSION 4 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_WM_CAPABILITIES_SINCE_VERSION 5 + +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_DESTROY_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SET_PARENT_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SET_TITLE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SET_APP_ID_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SHOW_WINDOW_MENU_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_MOVE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_RESIZE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SET_MAX_SIZE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SET_MIN_SIZE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SET_MAXIMIZED_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_UNSET_MAXIMIZED_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SET_FULLSCREEN_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_UNSET_FULLSCREEN_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_toplevel + */ +#define XDG_TOPLEVEL_SET_MINIMIZED_SINCE_VERSION 1 + +/** @ingroup iface_xdg_toplevel */ +static inline void +xdg_toplevel_set_user_data(struct xdg_toplevel *xdg_toplevel, void *user_data) +{ + wl_proxy_set_user_data((struct wl_proxy *) xdg_toplevel, user_data); +} + +/** @ingroup iface_xdg_toplevel */ +static inline void * +xdg_toplevel_get_user_data(struct xdg_toplevel *xdg_toplevel) +{ + return wl_proxy_get_user_data((struct wl_proxy *) xdg_toplevel); +} + +static inline uint32_t +xdg_toplevel_get_version(struct xdg_toplevel *xdg_toplevel) +{ + return wl_proxy_get_version((struct wl_proxy *) xdg_toplevel); +} + +/** + * @ingroup iface_xdg_toplevel + * + * This request destroys the role surface and unmaps the surface; + * see "Unmapping" behavior in interface section for details. + */ +static inline void +xdg_toplevel_destroy(struct xdg_toplevel *xdg_toplevel) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_DESTROY, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), WL_MARSHAL_FLAG_DESTROY); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Set the "parent" of this surface. This surface should be stacked + * above the parent surface and all other ancestor surfaces. + * + * Parent surfaces should be set on dialogs, toolboxes, or other + * "auxiliary" surfaces, so that the parent is raised when the dialog + * is raised. + * + * Setting a null parent for a child surface unsets its parent. Setting + * a null parent for a surface which currently has no parent is a no-op. + * + * Only mapped surfaces can have child surfaces. Setting a parent which + * is not mapped is equivalent to setting a null parent. If a surface + * becomes unmapped, its children's parent is set to the parent of + * the now-unmapped surface. If the now-unmapped surface has no parent, + * its children's parent is unset. If the now-unmapped surface becomes + * mapped again, its parent-child relationship is not restored. + * + * The parent toplevel must not be one of the child toplevel's + * descendants, and the parent must be different from the child toplevel, + * otherwise the invalid_parent protocol error is raised. + */ +static inline void +xdg_toplevel_set_parent(struct xdg_toplevel *xdg_toplevel, struct xdg_toplevel *parent) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SET_PARENT, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, parent); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Set a short title for the surface. + * + * This string may be used to identify the surface in a task bar, + * window list, or other user interface elements provided by the + * compositor. + * + * The string must be encoded in UTF-8. + */ +static inline void +xdg_toplevel_set_title(struct xdg_toplevel *xdg_toplevel, const char *title) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SET_TITLE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, title); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Set an application identifier for the surface. + * + * The app ID identifies the general class of applications to which + * the surface belongs. The compositor can use this to group multiple + * surfaces together, or to determine how to launch a new application. + * + * For D-Bus activatable applications, the app ID is used as the D-Bus + * service name. + * + * The compositor shell will try to group application surfaces together + * by their app ID. As a best practice, it is suggested to select app + * ID's that match the basename of the application's .desktop file. + * For example, "org.freedesktop.FooViewer" where the .desktop file is + * "org.freedesktop.FooViewer.desktop". + * + * Like other properties, a set_app_id request can be sent after the + * xdg_toplevel has been mapped to update the property. + * + * See the desktop-entry specification [0] for more details on + * application identifiers and how they relate to well-known D-Bus + * names and .desktop files. + * + * [0] https://standards.freedesktop.org/desktop-entry-spec/ + */ +static inline void +xdg_toplevel_set_app_id(struct xdg_toplevel *xdg_toplevel, const char *app_id) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SET_APP_ID, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, app_id); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Clients implementing client-side decorations might want to show + * a context menu when right-clicking on the decorations, giving the + * user a menu that they can use to maximize or minimize the window. + * + * This request asks the compositor to pop up such a window menu at + * the given position, relative to the local surface coordinates of + * the parent surface. There are no guarantees as to what menu items + * the window menu contains, or even if a window menu will be drawn + * at all. + * + * This request must be used in response to some sort of user action + * like a button press, key press, or touch down event. + */ +static inline void +xdg_toplevel_show_window_menu(struct xdg_toplevel *xdg_toplevel, struct wl_seat *seat, uint32_t serial, int32_t x, int32_t y) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SHOW_WINDOW_MENU, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, seat, serial, x, y); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Start an interactive, user-driven move of the surface. + * + * This request must be used in response to some sort of user action + * like a button press, key press, or touch down event. The passed + * serial is used to determine the type of interactive move (touch, + * pointer, etc). + * + * The server may ignore move requests depending on the state of + * the surface (e.g. fullscreen or maximized), or if the passed serial + * is no longer valid. + * + * If triggered, the surface will lose the focus of the device + * (wl_pointer, wl_touch, etc) used for the move. It is up to the + * compositor to visually indicate that the move is taking place, such as + * updating a pointer cursor, during the move. There is no guarantee + * that the device focus will return when the move is completed. + */ +static inline void +xdg_toplevel_move(struct xdg_toplevel *xdg_toplevel, struct wl_seat *seat, uint32_t serial) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_MOVE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, seat, serial); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Start a user-driven, interactive resize of the surface. + * + * This request must be used in response to some sort of user action + * like a button press, key press, or touch down event. The passed + * serial is used to determine the type of interactive resize (touch, + * pointer, etc). + * + * The server may ignore resize requests depending on the state of + * the surface (e.g. fullscreen or maximized). + * + * If triggered, the client will receive configure events with the + * "resize" state enum value and the expected sizes. See the "resize" + * enum value for more details about what is required. The client + * must also acknowledge configure events using "ack_configure". After + * the resize is completed, the client will receive another "configure" + * event without the resize state. + * + * If triggered, the surface also will lose the focus of the device + * (wl_pointer, wl_touch, etc) used for the resize. It is up to the + * compositor to visually indicate that the resize is taking place, + * such as updating a pointer cursor, during the resize. There is no + * guarantee that the device focus will return when the resize is + * completed. + * + * The edges parameter specifies how the surface should be resized, and + * is one of the values of the resize_edge enum. Values not matching + * a variant of the enum will cause the invalid_resize_edge protocol error. + * The compositor may use this information to update the surface position + * for example when dragging the top left corner. The compositor may also + * use this information to adapt its behavior, e.g. choose an appropriate + * cursor image. + */ +static inline void +xdg_toplevel_resize(struct xdg_toplevel *xdg_toplevel, struct wl_seat *seat, uint32_t serial, uint32_t edges) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_RESIZE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, seat, serial, edges); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Set a maximum size for the window. + * + * The client can specify a maximum size so that the compositor does + * not try to configure the window beyond this size. + * + * The width and height arguments are in window geometry coordinates. + * See xdg_surface.set_window_geometry. + * + * Values set in this way are double-buffered, see wl_surface.commit. + * + * The compositor can use this information to allow or disallow + * different states like maximize or fullscreen and draw accurate + * animations. + * + * Similarly, a tiling window manager may use this information to + * place and resize client windows in a more effective way. + * + * The client should not rely on the compositor to obey the maximum + * size. The compositor may decide to ignore the values set by the + * client and request a larger size. + * + * If never set, or a value of zero in the request, means that the + * client has no expected maximum size in the given dimension. + * As a result, a client wishing to reset the maximum size + * to an unspecified state can use zero for width and height in the + * request. + * + * Requesting a maximum size to be smaller than the minimum size of + * a surface is illegal and will result in an invalid_size error. + * + * The width and height must be greater than or equal to zero. Using + * strictly negative values for width or height will result in a + * invalid_size error. + */ +static inline void +xdg_toplevel_set_max_size(struct xdg_toplevel *xdg_toplevel, int32_t width, int32_t height) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SET_MAX_SIZE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, width, height); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Set a minimum size for the window. + * + * The client can specify a minimum size so that the compositor does + * not try to configure the window below this size. + * + * The width and height arguments are in window geometry coordinates. + * See xdg_surface.set_window_geometry. + * + * Values set in this way are double-buffered, see wl_surface.commit. + * + * The compositor can use this information to allow or disallow + * different states like maximize or fullscreen and draw accurate + * animations. + * + * Similarly, a tiling window manager may use this information to + * place and resize client windows in a more effective way. + * + * The client should not rely on the compositor to obey the minimum + * size. The compositor may decide to ignore the values set by the + * client and request a smaller size. + * + * If never set, or a value of zero in the request, means that the + * client has no expected minimum size in the given dimension. + * As a result, a client wishing to reset the minimum size + * to an unspecified state can use zero for width and height in the + * request. + * + * Requesting a minimum size to be larger than the maximum size of + * a surface is illegal and will result in an invalid_size error. + * + * The width and height must be greater than or equal to zero. Using + * strictly negative values for width and height will result in a + * invalid_size error. + */ +static inline void +xdg_toplevel_set_min_size(struct xdg_toplevel *xdg_toplevel, int32_t width, int32_t height) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SET_MIN_SIZE, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, width, height); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Maximize the surface. + * + * After requesting that the surface should be maximized, the compositor + * will respond by emitting a configure event. Whether this configure + * actually sets the window maximized is subject to compositor policies. + * The client must then update its content, drawing in the configured + * state. The client must also acknowledge the configure when committing + * the new content (see ack_configure). + * + * It is up to the compositor to decide how and where to maximize the + * surface, for example which output and what region of the screen should + * be used. + * + * If the surface was already maximized, the compositor will still emit + * a configure event with the "maximized" state. + * + * If the surface is in a fullscreen state, this request has no direct + * effect. It may alter the state the surface is returned to when + * unmaximized unless overridden by the compositor. + */ +static inline void +xdg_toplevel_set_maximized(struct xdg_toplevel *xdg_toplevel) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SET_MAXIMIZED, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Unmaximize the surface. + * + * After requesting that the surface should be unmaximized, the compositor + * will respond by emitting a configure event. Whether this actually + * un-maximizes the window is subject to compositor policies. + * If available and applicable, the compositor will include the window + * geometry dimensions the window had prior to being maximized in the + * configure event. The client must then update its content, drawing it in + * the configured state. The client must also acknowledge the configure + * when committing the new content (see ack_configure). + * + * It is up to the compositor to position the surface after it was + * unmaximized; usually the position the surface had before maximizing, if + * applicable. + * + * If the surface was already not maximized, the compositor will still + * emit a configure event without the "maximized" state. + * + * If the surface is in a fullscreen state, this request has no direct + * effect. It may alter the state the surface is returned to when + * unmaximized unless overridden by the compositor. + */ +static inline void +xdg_toplevel_unset_maximized(struct xdg_toplevel *xdg_toplevel) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_UNSET_MAXIMIZED, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Make the surface fullscreen. + * + * After requesting that the surface should be fullscreened, the + * compositor will respond by emitting a configure event. Whether the + * client is actually put into a fullscreen state is subject to compositor + * policies. The client must also acknowledge the configure when + * committing the new content (see ack_configure). + * + * The output passed by the request indicates the client's preference as + * to which display it should be set fullscreen on. If this value is NULL, + * it's up to the compositor to choose which display will be used to map + * this surface. + * + * If the surface doesn't cover the whole output, the compositor will + * position the surface in the center of the output and compensate with + * with border fill covering the rest of the output. The content of the + * border fill is undefined, but should be assumed to be in some way that + * attempts to blend into the surrounding area (e.g. solid black). + * + * If the fullscreened surface is not opaque, the compositor must make + * sure that other screen content not part of the same surface tree (made + * up of subsurfaces, popups or similarly coupled surfaces) are not + * visible below the fullscreened surface. + */ +static inline void +xdg_toplevel_set_fullscreen(struct xdg_toplevel *xdg_toplevel, struct wl_output *output) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SET_FULLSCREEN, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0, output); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Make the surface no longer fullscreen. + * + * After requesting that the surface should be unfullscreened, the + * compositor will respond by emitting a configure event. + * Whether this actually removes the fullscreen state of the client is + * subject to compositor policies. + * + * Making a surface unfullscreen sets states for the surface based on the following: + * * the state(s) it may have had before becoming fullscreen + * * any state(s) decided by the compositor + * * any state(s) requested by the client while the surface was fullscreen + * + * The compositor may include the previous window geometry dimensions in + * the configure event, if applicable. + * + * The client must also acknowledge the configure when committing the new + * content (see ack_configure). + */ +static inline void +xdg_toplevel_unset_fullscreen(struct xdg_toplevel *xdg_toplevel) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_UNSET_FULLSCREEN, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0); +} + +/** + * @ingroup iface_xdg_toplevel + * + * Request that the compositor minimize your surface. There is no + * way to know if the surface is currently minimized, nor is there + * any way to unset minimization on this surface. + * + * If you are looking to throttle redrawing when minimized, please + * instead use the wl_surface.frame event for this, as this will + * also work with live previews on windows in Alt-Tab, Expose or + * similar compositor features. + */ +static inline void +xdg_toplevel_set_minimized(struct xdg_toplevel *xdg_toplevel) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_toplevel, + XDG_TOPLEVEL_SET_MINIMIZED, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_toplevel), 0); +} + +#ifndef XDG_POPUP_ERROR_ENUM +#define XDG_POPUP_ERROR_ENUM +enum xdg_popup_error { + /** + * tried to grab after being mapped + */ + XDG_POPUP_ERROR_INVALID_GRAB = 0, +}; +#endif /* XDG_POPUP_ERROR_ENUM */ + +/** + * @ingroup iface_xdg_popup + * @struct xdg_popup_listener + */ +struct xdg_popup_listener { + /** + * configure the popup surface + * + * This event asks the popup surface to configure itself given + * the configuration. The configured state should not be applied + * immediately. See xdg_surface.configure for details. + * + * The x and y arguments represent the position the popup was + * placed at given the xdg_positioner rule, relative to the upper + * left corner of the window geometry of the parent surface. + * + * For version 2 or older, the configure event for an xdg_popup is + * only ever sent once for the initial configuration. Starting with + * version 3, it may be sent again if the popup is setup with an + * xdg_positioner with set_reactive requested, or in response to + * xdg_popup.reposition requests. + * @param x x position relative to parent surface window geometry + * @param y y position relative to parent surface window geometry + * @param width window geometry width + * @param height window geometry height + */ + void (*configure)(void *data, + struct xdg_popup *xdg_popup, + int32_t x, + int32_t y, + int32_t width, + int32_t height); + /** + * popup interaction is done + * + * The popup_done event is sent out when a popup is dismissed by + * the compositor. The client should destroy the xdg_popup object + * at this point. + */ + void (*popup_done)(void *data, + struct xdg_popup *xdg_popup); + /** + * signal the completion of a repositioned request + * + * The repositioned event is sent as part of a popup + * configuration sequence, together with xdg_popup.configure and + * lastly xdg_surface.configure to notify the completion of a + * reposition request. + * + * The repositioned event is to notify about the completion of a + * xdg_popup.reposition request. The token argument is the token + * passed in the xdg_popup.reposition request. + * + * Immediately after this event is emitted, xdg_popup.configure and + * xdg_surface.configure will be sent with the updated size and + * position, as well as a new configure serial. + * + * The client should optionally update the content of the popup, + * but must acknowledge the new popup configuration for the new + * position to take effect. See xdg_surface.ack_configure for + * details. + * @param token reposition request token + * @since 3 + */ + void (*repositioned)(void *data, + struct xdg_popup *xdg_popup, + uint32_t token); +}; + +/** + * @ingroup iface_xdg_popup + */ +static inline int +xdg_popup_add_listener(struct xdg_popup *xdg_popup, + const struct xdg_popup_listener *listener, void *data) +{ + return wl_proxy_add_listener((struct wl_proxy *) xdg_popup, + (void (**)(void)) listener, data); +} + +#define XDG_POPUP_DESTROY 0 +#define XDG_POPUP_GRAB 1 +#define XDG_POPUP_REPOSITION 2 + +/** + * @ingroup iface_xdg_popup + */ +#define XDG_POPUP_CONFIGURE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_popup + */ +#define XDG_POPUP_POPUP_DONE_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_popup + */ +#define XDG_POPUP_REPOSITIONED_SINCE_VERSION 3 + +/** + * @ingroup iface_xdg_popup + */ +#define XDG_POPUP_DESTROY_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_popup + */ +#define XDG_POPUP_GRAB_SINCE_VERSION 1 +/** + * @ingroup iface_xdg_popup + */ +#define XDG_POPUP_REPOSITION_SINCE_VERSION 3 + +/** @ingroup iface_xdg_popup */ +static inline void +xdg_popup_set_user_data(struct xdg_popup *xdg_popup, void *user_data) +{ + wl_proxy_set_user_data((struct wl_proxy *) xdg_popup, user_data); +} + +/** @ingroup iface_xdg_popup */ +static inline void * +xdg_popup_get_user_data(struct xdg_popup *xdg_popup) +{ + return wl_proxy_get_user_data((struct wl_proxy *) xdg_popup); +} + +static inline uint32_t +xdg_popup_get_version(struct xdg_popup *xdg_popup) +{ + return wl_proxy_get_version((struct wl_proxy *) xdg_popup); +} + +/** + * @ingroup iface_xdg_popup + * + * This destroys the popup. Explicitly destroying the xdg_popup + * object will also dismiss the popup, and unmap the surface. + * + * If this xdg_popup is not the "topmost" popup, the + * xdg_wm_base.not_the_topmost_popup protocol error will be sent. + */ +static inline void +xdg_popup_destroy(struct xdg_popup *xdg_popup) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_popup, + XDG_POPUP_DESTROY, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_popup), WL_MARSHAL_FLAG_DESTROY); +} + +/** + * @ingroup iface_xdg_popup + * + * This request makes the created popup take an explicit grab. An explicit + * grab will be dismissed when the user dismisses the popup, or when the + * client destroys the xdg_popup. This can be done by the user clicking + * outside the surface, using the keyboard, or even locking the screen + * through closing the lid or a timeout. + * + * If the compositor denies the grab, the popup will be immediately + * dismissed. + * + * This request must be used in response to some sort of user action like a + * button press, key press, or touch down event. The serial number of the + * event should be passed as 'serial'. + * + * The parent of a grabbing popup must either be an xdg_toplevel surface or + * another xdg_popup with an explicit grab. If the parent is another + * xdg_popup it means that the popups are nested, with this popup now being + * the topmost popup. + * + * Nested popups must be destroyed in the reverse order they were created + * in, e.g. the only popup you are allowed to destroy at all times is the + * topmost one. + * + * When compositors choose to dismiss a popup, they may dismiss every + * nested grabbing popup as well. When a compositor dismisses popups, it + * will follow the same dismissing order as required from the client. + * + * If the topmost grabbing popup is destroyed, the grab will be returned to + * the parent of the popup, if that parent previously had an explicit grab. + * + * If the parent is a grabbing popup which has already been dismissed, this + * popup will be immediately dismissed. If the parent is a popup that did + * not take an explicit grab, an error will be raised. + * + * During a popup grab, the client owning the grab will receive pointer + * and touch events for all their surfaces as normal (similar to an + * "owner-events" grab in X11 parlance), while the top most grabbing popup + * will always have keyboard focus. + */ +static inline void +xdg_popup_grab(struct xdg_popup *xdg_popup, struct wl_seat *seat, uint32_t serial) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_popup, + XDG_POPUP_GRAB, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_popup), 0, seat, serial); +} + +/** + * @ingroup iface_xdg_popup + * + * Reposition an already-mapped popup. The popup will be placed given the + * details in the passed xdg_positioner object, and a + * xdg_popup.repositioned followed by xdg_popup.configure and + * xdg_surface.configure will be emitted in response. Any parameters set + * by the previous positioner will be discarded. + * + * The passed token will be sent in the corresponding + * xdg_popup.repositioned event. The new popup position will not take + * effect until the corresponding configure event is acknowledged by the + * client. See xdg_popup.repositioned for details. The token itself is + * opaque, and has no other special meaning. + * + * If multiple reposition requests are sent, the compositor may skip all + * but the last one. + * + * If the popup is repositioned in response to a configure event for its + * parent, the client should send an xdg_positioner.set_parent_configure + * and possibly an xdg_positioner.set_parent_size request to allow the + * compositor to properly constrain the popup. + * + * If the popup is repositioned together with a parent that is being + * resized, but not in response to a configure event, the client should + * send an xdg_positioner.set_parent_size request. + */ +static inline void +xdg_popup_reposition(struct xdg_popup *xdg_popup, struct xdg_positioner *positioner, uint32_t token) +{ + wl_proxy_marshal_flags((struct wl_proxy *) xdg_popup, + XDG_POPUP_REPOSITION, NULL, wl_proxy_get_version((struct wl_proxy *) xdg_popup), 0, positioner, token); +} + +#ifdef __cplusplus +} +#endif + +#endif diff --git a/Wayland/xdg-shell-protocol.c b/Wayland/xdg-shell-protocol.c new file mode 100644 index 0000000..b16e1b6 --- /dev/null +++ b/Wayland/xdg-shell-protocol.c @@ -0,0 +1,184 @@ +/* Generated by wayland-scanner 1.23.0 */ + +/* + * Copyright © 2008-2013 Kristian Høgsberg + * Copyright © 2013 Rafael Antognolli + * Copyright © 2013 Jasper St. Pierre + * Copyright © 2010-2013 Intel Corporation + * Copyright © 2015-2017 Samsung Electronics Co., Ltd + * Copyright © 2015-2017 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + */ + +#include +#include +#include +#include "wayland-util.h" + +#ifndef __has_attribute +# define __has_attribute(x) 0 /* Compatibility with non-clang compilers. */ +#endif + +#if (__has_attribute(visibility) || defined(__GNUC__) && __GNUC__ >= 4) +#define WL_PRIVATE __attribute__ ((visibility("hidden"))) +#else +#define WL_PRIVATE +#endif + +extern const struct wl_interface wl_output_interface; +extern const struct wl_interface wl_seat_interface; +extern const struct wl_interface wl_surface_interface; +extern const struct wl_interface xdg_popup_interface; +extern const struct wl_interface xdg_positioner_interface; +extern const struct wl_interface xdg_surface_interface; +extern const struct wl_interface xdg_toplevel_interface; + +static const struct wl_interface *xdg_shell_types[] = { + NULL, + NULL, + NULL, + NULL, + &xdg_positioner_interface, + &xdg_surface_interface, + &wl_surface_interface, + &xdg_toplevel_interface, + &xdg_popup_interface, + &xdg_surface_interface, + &xdg_positioner_interface, + &xdg_toplevel_interface, + &wl_seat_interface, + NULL, + NULL, + NULL, + &wl_seat_interface, + NULL, + &wl_seat_interface, + NULL, + NULL, + &wl_output_interface, + &wl_seat_interface, + NULL, + &xdg_positioner_interface, + NULL, +}; + +static const struct wl_message xdg_wm_base_requests[] = { + { "destroy", "", xdg_shell_types + 0 }, + { "create_positioner", "n", xdg_shell_types + 4 }, + { "get_xdg_surface", "no", xdg_shell_types + 5 }, + { "pong", "u", xdg_shell_types + 0 }, +}; + +static const struct wl_message xdg_wm_base_events[] = { + { "ping", "u", xdg_shell_types + 0 }, +}; + +WL_PRIVATE const struct wl_interface xdg_wm_base_interface = { + "xdg_wm_base", 6, + 4, xdg_wm_base_requests, + 1, xdg_wm_base_events, +}; + +static const struct wl_message xdg_positioner_requests[] = { + { "destroy", "", xdg_shell_types + 0 }, + { "set_size", "ii", xdg_shell_types + 0 }, + { "set_anchor_rect", "iiii", xdg_shell_types + 0 }, + { "set_anchor", "u", xdg_shell_types + 0 }, + { "set_gravity", "u", xdg_shell_types + 0 }, + { "set_constraint_adjustment", "u", xdg_shell_types + 0 }, + { "set_offset", "ii", xdg_shell_types + 0 }, + { "set_reactive", "3", xdg_shell_types + 0 }, + { "set_parent_size", "3ii", xdg_shell_types + 0 }, + { "set_parent_configure", "3u", xdg_shell_types + 0 }, +}; + +WL_PRIVATE const struct wl_interface xdg_positioner_interface = { + "xdg_positioner", 6, + 10, xdg_positioner_requests, + 0, NULL, +}; + +static const struct wl_message xdg_surface_requests[] = { + { "destroy", "", xdg_shell_types + 0 }, + { "get_toplevel", "n", xdg_shell_types + 7 }, + { "get_popup", "n?oo", xdg_shell_types + 8 }, + { "set_window_geometry", "iiii", xdg_shell_types + 0 }, + { "ack_configure", "u", xdg_shell_types + 0 }, +}; + +static const struct wl_message xdg_surface_events[] = { + { "configure", "u", xdg_shell_types + 0 }, +}; + +WL_PRIVATE const struct wl_interface xdg_surface_interface = { + "xdg_surface", 6, + 5, xdg_surface_requests, + 1, xdg_surface_events, +}; + +static const struct wl_message xdg_toplevel_requests[] = { + { "destroy", "", xdg_shell_types + 0 }, + { "set_parent", "?o", xdg_shell_types + 11 }, + { "set_title", "s", xdg_shell_types + 0 }, + { "set_app_id", "s", xdg_shell_types + 0 }, + { "show_window_menu", "ouii", xdg_shell_types + 12 }, + { "move", "ou", xdg_shell_types + 16 }, + { "resize", "ouu", xdg_shell_types + 18 }, + { "set_max_size", "ii", xdg_shell_types + 0 }, + { "set_min_size", "ii", xdg_shell_types + 0 }, + { "set_maximized", "", xdg_shell_types + 0 }, + { "unset_maximized", "", xdg_shell_types + 0 }, + { "set_fullscreen", "?o", xdg_shell_types + 21 }, + { "unset_fullscreen", "", xdg_shell_types + 0 }, + { "set_minimized", "", xdg_shell_types + 0 }, +}; + +static const struct wl_message xdg_toplevel_events[] = { + { "configure", "iia", xdg_shell_types + 0 }, + { "close", "", xdg_shell_types + 0 }, + { "configure_bounds", "4ii", xdg_shell_types + 0 }, + { "wm_capabilities", "5a", xdg_shell_types + 0 }, +}; + +WL_PRIVATE const struct wl_interface xdg_toplevel_interface = { + "xdg_toplevel", 6, + 14, xdg_toplevel_requests, + 4, xdg_toplevel_events, +}; + +static const struct wl_message xdg_popup_requests[] = { + { "destroy", "", xdg_shell_types + 0 }, + { "grab", "ou", xdg_shell_types + 22 }, + { "reposition", "3ou", xdg_shell_types + 24 }, +}; + +static const struct wl_message xdg_popup_events[] = { + { "configure", "iiii", xdg_shell_types + 0 }, + { "popup_done", "", xdg_shell_types + 0 }, + { "repositioned", "3u", xdg_shell_types + 0 }, +}; + +WL_PRIVATE const struct wl_interface xdg_popup_interface = { + "xdg_popup", 6, + 3, xdg_popup_requests, + 3, xdg_popup_events, +}; + diff --git a/allocator.c b/allocator.c new file mode 100644 index 0000000..1eebaf4 --- /dev/null +++ b/allocator.c @@ -0,0 +1,88 @@ +#include "log.h" + +#include +#include +#include +#include + +#define MAX_ALLOCS 256 +struct Alloc { + void *start; + uint size; + struct Alloc *next; +}; +struct Mem { + void *mem; + size_t mem_size; + uint count; + struct Alloc *alloc; + struct Alloc allocs[MAX_ALLOCS]; +}; + +struct Mem *make_mem(size_t size) { + struct Mem *mem = malloc(sizeof(struct Mem)); + mem->mem_size = size; + mem->mem = malloc(mem->mem_size); + memset(mem->mem, 0, mem->mem_size); + mem->count = 0; + memset(&mem->allocs, 0, sizeof(struct Alloc) * MAX_ALLOCS); + mem->allocs[0] = (struct Alloc){mem->mem + mem->mem_size, 0, NULL}; + mem->allocs[1] = (struct Alloc){mem->mem, 0, &mem->allocs[0]}; + mem->alloc = &mem->allocs[1]; + return mem; +} +void uninit_mem(struct Mem *mem) { + free(mem->mem); + free(mem); +} + +static struct Alloc *get_next_alloc(struct Mem *mem) { + uint index = mem->count; + mem->count++; + while (*(char *)&mem->allocs[mem->count] != 0) { + mem->count++; + if (mem->count >= MAX_ALLOCS) { + crash("out of allocs!"); + } + } + return &mem->allocs[index]; +} + +void *mem_malloc(struct Mem *mem, size_t size) { + struct Alloc *prev = mem->alloc; + while (prev->next != NULL) { + struct Alloc *next = prev->next; + if (next->start - (prev->start + prev->size) > size) { + struct Alloc *new = get_next_alloc(mem); + *new = (struct Alloc){ + prev->start + prev->size, + size, + next, + }; + prev->next = new; + meow("MALLOC %p of size %zu, prev is at %p with size %u", new->start, + size, prev->start, prev->size); + /* print_backtrace(); */ + return new->start; + } else { + prev = next; + } + } + crash("no big enaugh free space found!"); + return NULL; +} + +void mem_free(struct Mem *mem, void *p) { + struct Alloc *current = mem->alloc; + struct Alloc *prev = mem->alloc; + while (p < current->start) { + prev = current; + current = current->next; + } + prev->next = current->next; + int index = current - mem->allocs; + if (index < mem->count) { + mem->count = index; + } + memset(&mem->allocs[index], 0, sizeof(struct Alloc)); +} diff --git a/allocator.h b/allocator.h new file mode 100644 index 0000000..e6d3ee9 --- /dev/null +++ b/allocator.h @@ -0,0 +1,14 @@ +#ifndef INCLUDE_WAYLANDCLIENT_ALLOCATOR_H_ +#define INCLUDE_WAYLANDCLIENT_ALLOCATOR_H_ + +#include "sys/types.h" + +struct Mem; + +struct Mem *make_mem(size_t size); +void uninit_mem(struct Mem *mem); + +void *mem_malloc(struct Mem *mem, size_t size); +void mem_free(struct Mem *mem, void *p); + +#endif // INCLUDE_WAYLANDCLIENT_ALLOCATOR_H_ diff --git a/comp.c b/comp.c new file mode 100644 index 0000000..f0beb20 --- /dev/null +++ b/comp.c @@ -0,0 +1,27 @@ +#include "comp.h" + +#ifdef WAYLAND + +cat_Comp *cat_comp_init(const char *title, int width, int heigh, + struct Vk **vk) { + return cat_init_wl(title, width, heigh, vk); +} +void cat_comp_uninit(cat_Comp *state) { cat_uninit_wl(state); } +void cat_comp_draw(cat_Comp *state) { cat_wl_draw(state); } +bool cat_comp_should_close(cat_Comp *state) { + return cat_wl_should_close(state); +} + +#else + +cat_Comp *cat_comp_init(const char *title, int width, int heigh, + struct Vk **vk) { + return cat_init_glfw(title, width, heigh, vk); +} +void cat_comp_uninit(cat_Comp *state) { cat_uninit_glfw(state); } +void cat_comp_draw(cat_Comp *state) { cat_glfw_draw(state); } +bool cat_comp_should_close(cat_Comp *state) { + return cat_glfw_should_close(state); +} + +#endif //! WAYLAND diff --git a/comp.h b/comp.h new file mode 100644 index 0000000..f593427 --- /dev/null +++ b/comp.h @@ -0,0 +1,27 @@ +#ifndef INCLUDE_WAYLANDCLIENT_CAT_COMP_H_ +#define INCLUDE_WAYLANDCLIENT_CAT_COMP_H_ + +#define WAYLAND + +#include + +#ifdef WAYLAND + +#include "vulkan.h" +#include "wayland.h" +typedef struct cat_Wl cat_Comp; + +#else + +#include "glfw.h" +typedef struct cat_GLFW cat_Comp; + +#endif // !WAYLAND + +cat_Comp *cat_comp_init(const char *title, int width, int heigh, + struct Vk **vk); +void cat_comp_uninit(cat_Comp *state); +void cat_comp_draw(cat_Comp *state); +bool cat_comp_should_close(cat_Comp *state); + +#endif // INCLUDE_WAYLANDCLIENT_CAT_COMP_H_ diff --git a/dynarray.c b/dynarray.c new file mode 100644 index 0000000..ab84329 --- /dev/null +++ b/dynarray.c @@ -0,0 +1,20 @@ +#include +struct da_template { + void *items; + int count; + int capacity; +}; + +void *dyn_array_create() { + struct da_template *da = malloc(sizeof(struct da_template)); + da->items = malloc(0); + da->count = 0; + da->capacity = 0; + return da; +} +void dyn_array_create_inplace(void *array) { + struct da_template *da = array; + da->items = malloc(0); + da->count = 0; + da->capacity = 0; +} diff --git a/dynarray.h b/dynarray.h new file mode 100644 index 0000000..48d2dcd --- /dev/null +++ b/dynarray.h @@ -0,0 +1,42 @@ +#ifndef INCLUDE_WAYLANDCLIENT_DYNARRAY_H_ +#define INCLUDE_WAYLANDCLIENT_DYNARRAY_H_ + +#include + +#define dyn_array_append(array, item) \ + do { \ + if ((array)->count >= (array)->capacity) { \ + (array)->capacity += 10; \ + (array)->items = realloc((array)->items, \ + (array)->capacity * sizeof(*(array)->items)); \ + } \ + (array)->items[(array)->count++] = (item); \ + } while (0) +#define dyn_array_remove(array, index) \ + do { \ + (array)->items[(index)] = (array)->items[(array)->count - 1]; \ + (array)->items[(array)->count] = NULL; \ + (array)->count--; \ + } while (0) +#define dyn_array_destroy(array) free((array)->items) +#define dyn_array_reset(array) \ + do { \ + (array)->capacity = 10; \ + (array)->count = 0; \ + (array)->items = \ + realloc((array)->items, (array)->capacity * sizeof(*(array)->items)); \ + } while (0) + +#define dyn_array_define($name, $type) \ + struct $name { \ + $type *items; \ + int count; \ + int capacity; \ + } + +dyn_array_define(da_uint32_t, uint32_t); + +void *dyn_array_create(); +void dyn_array_create_inplace(void *array); + +#endif // INCLUDE_WAYLANDCLIENT_DYNARRAY_H_ diff --git a/glfw.c b/glfw.c new file mode 100644 index 0000000..7d42315 --- /dev/null +++ b/glfw.c @@ -0,0 +1,44 @@ +#include "log.h" +#include "vulkan.h" + +#include +#include +#include + +struct cat_GLFW { + GLFWwindow *window; + struct Vk *vk; +}; + +VkSurfaceKHR create_glfw_surface(void *data, VkInstance instance) { + struct cat_GLFW *state = data; + VkSurfaceKHR surface; + + meow("%d", glfwCreateWindowSurface(instance, state->window, NULL, &surface)); + meow("created a surface at %p", surface); + return surface; +} + +struct cat_GLFW *cat_init_glfw(const char *name, int width, int heigh, + struct Vk **vk) { + glfwInit(); + + glfwWindowHint(GLFW_CLIENT_API, GLFW_NO_API); + glfwWindowHint(GLFW_RESIZABLE, GLFW_FALSE); + + struct cat_GLFW *state = malloc(sizeof(struct cat_GLFW)); + state->window = glfwCreateWindow(width, heigh, "meooow", NULL, NULL); + + state->vk = init_vk(state, 500, 500, create_glfw_surface); + *vk = state->vk; + return state; +} +void cat_glfw_draw(struct cat_GLFW *state) { + glfwPollEvents(); + vk_draw(state->vk); +} +void cat_uninit_glfw(struct cat_GLFW *state) { + glfwDestroyWindow(state->window); + glfwTerminate(); +} +bool cat_glfw_should_close(struct cat_GLFW *state) { return false; } diff --git a/glfw.h b/glfw.h new file mode 100644 index 0000000..192da5a --- /dev/null +++ b/glfw.h @@ -0,0 +1,6 @@ +struct cat_GLFW; +struct cat_GLFW *cat_init_glfw(const char *name, int width, int heigh, + struct Vk **vk); +void cat_glfw_draw(struct cat_GLFW *state); +void cat_uninit_glfw(struct cat_GLFW *state); +bool cat_glfw_should_close(struct cat_GLFW *state); diff --git a/hashmap.c b/hashmap.c new file mode 100644 index 0000000..4d9b32f --- /dev/null +++ b/hashmap.c @@ -0,0 +1,89 @@ +#include +#include +#include +#include + +#include "log.h" + +#define HASH_MASK 0b111111 +#define TYPE uint32_t + +struct Node_ui32 { + TYPE v; + struct Node_ui32 *next; +}; +struct Bucket_ui32 { + struct Node_ui32 *nodes; +}; +struct Hashmap_ui32 { + struct Bucket_ui32 buckets[HASH_MASK + 1]; +}; + +struct Hashmap_ui32 *create_hashmap_ui32() { + struct Hashmap_ui32 *map = malloc(sizeof(struct Hashmap_ui32)); + for (int i = 0; i <= HASH_MASK; i++) { + map->buckets[i].nodes = NULL; + } + return map; +} +void destroy_hashmap_ui32(struct Hashmap_ui32 *map) { + for (int i = 0; i <= HASH_MASK; i++) { + struct Node_ui32 *node = map->buckets[i].nodes; + while (node != NULL) { + struct Node_ui32 *tmp = node; + node = tmp->next; + free(tmp); + } + } + memset(map, 0, sizeof(struct Hashmap_ui32)); + free(map); +} +int hash(TYPE v) { return v & HASH_MASK; } +void insert_hashmap_ui32(struct Hashmap_ui32 *map, TYPE v) { + int h = hash(v); + struct Node_ui32 *node = map->buckets[h].nodes; + while (node != NULL) { + if (node->v == v) { + meow("%d was already in hashmap", v); + return; + } + node = node->next; + } + struct Node_ui32 *n = malloc(sizeof(struct Node_ui32)); + n->v = v; + n->next = map->buckets[h].nodes; + map->buckets[h].nodes = n; +} +void remove_hashmap_ui32(struct Hashmap_ui32 *map, TYPE v) { + int h = hash(v); + struct Node_ui32 *prev = NULL; + struct Node_ui32 *node = map->buckets[h].nodes; + while (node != NULL) { + if (node->v == v) { + if (prev == NULL) { + if (node->next == NULL) { + map->buckets[h].nodes = NULL; + } else { + map->buckets[h].nodes = node->next; + } + } else { + prev->next = node->next; + } + free(node); + return; + } + node = node->next; + } + meow("did not find %d in hashmap", v); +} +bool get_hashmap_ui32(struct Hashmap_ui32 *map, TYPE v) { + int h = hash(v); + struct Node_ui32 *node = map->buckets[h].nodes; + while (node != NULL) { + if (node->v == v) { + return true; + } + node = node->next; + } + return false; +} diff --git a/hashmap.h b/hashmap.h new file mode 100644 index 0000000..858d7a3 --- /dev/null +++ b/hashmap.h @@ -0,0 +1,10 @@ +#include +#include + +struct Hashmap_ui32; + +struct Hashmap_ui32 *create_hashmap_ui32(); +void destroy_hashmap_ui32(struct Hashmap_ui32 *map); +void insert_hashmap_ui32(struct Hashmap_ui32 *map, uint32_t v); +void remove_hashmap_ui32(struct Hashmap_ui32 *map, uint32_t v); +bool get_hashmap_ui32(struct Hashmap_ui32 *map, uint32_t v); diff --git a/image.c b/image.c new file mode 100644 index 0000000..be1e901 --- /dev/null +++ b/image.c @@ -0,0 +1,9 @@ +#define STB_IMAGE_IMPLEMENTATION +#include "stb_image.h" + +#include "image.h" + +uchar *load_image(const char *path, struct IVec2 *dims) { + int _channels; + return stbi_load(path, &dims->x, &dims->y, &_channels, STBI_rgb_alpha); +} diff --git a/image.h b/image.h new file mode 100644 index 0000000..cb7f94a --- /dev/null +++ b/image.h @@ -0,0 +1,10 @@ +#ifndef INCLUDE_WAYLANDCLIENT_CAT_IMAGE_H_ +#define INCLUDE_WAYLANDCLIENT_CAT_IMAGE_H_ + +#include "types.h" + +typedef unsigned char uchar; + +uchar *load_image(const char *path, struct IVec2 *dims); + +#endif // INCLUDE_WAYLANDCLIENT_CAT_IMAGE_H_ diff --git a/image.png b/image.png new file mode 100644 index 0000000..72036ac Binary files /dev/null and b/image.png differ diff --git a/io.c b/io.c new file mode 100644 index 0000000..4b5c2ea --- /dev/null +++ b/io.c @@ -0,0 +1,47 @@ +#include "log.h" + +#include +#include +#include + +char *read_text_file(char *path) { + FILE *file = fopen(path, "r"); + if (file == NULL) { + meow("Could not open file %s", path); + return NULL; + } + fseek(file, 0, SEEK_END); + long size = ftell(file); + fseek(file, 0, SEEK_SET); + + char *content = malloc((size + 1) * sizeof(char)); + + long count = fread(content, sizeof(char), size, file); + if (count < size) { + content = realloc(content, count + 1); + } + content[count] = 0x00; + fclose(file); + return content; +} + +uint32_t *read_binary_file(char *path, int *size) { + FILE *file = fopen(path, "r"); + if (file == NULL) { + meow("Could not open file %s", path); + return NULL; + } + fseek(file, 0, SEEK_END); + long fsize = ftell(file); + *size = fsize; + fseek(file, 0, SEEK_SET); + + char *content = malloc((fsize) * sizeof(char)); + + long count = fread(content, sizeof(char), fsize, file); + if (count < fsize) { + content = realloc(content, count); + } + fclose(file); + return (uint32_t *)content; +} diff --git a/io.h b/io.h new file mode 100644 index 0000000..5a7c83e --- /dev/null +++ b/io.h @@ -0,0 +1,9 @@ +#ifndef INCLUDE_WAYLANDCLIENT_IO_H_ +#define INCLUDE_WAYLANDCLIENT_IO_H_ + +#include + +char *read_text_file(char *path); +uint32_t *read_binary_file(char *path, int *size); + +#endif // INCLUDE_WAYLANDCLIENT_IO_H_ diff --git a/kitty.c b/kitty.c new file mode 100644 index 0000000..aaf0bae --- /dev/null +++ b/kitty.c @@ -0,0 +1,579 @@ +#include +#include +#include +#include + +#include "dynarray.h" +#include "image.h" +#include "io.h" +#include "vulkan.h" +#include "vulkan_helpers.c" + +enum AttatchType { + CAT_ATTATCH_IMAGE, + CAT_ATTATCH_UBO, +}; + +struct Attatchment { + enum AttatchType type; + union { + struct { + const char *path; + uchar *pixels; + struct IVec2 dims; + VkDeviceSize size; + VkImage image; + VkImageView view; + VkSampler sampler; + VkDeviceMemory memory; + } image; + struct { + uint32_t size; + void *mapped[MAX_FRAMES_IN_FLIGHT]; + VkBuffer buffer[MAX_FRAMES_IN_FLIGHT]; + VkDeviceMemory memory[MAX_FRAMES_IN_FLIGHT]; + } ubo; + }; +}; + +dyn_array_define(da_Attatchment, struct Attatchment); + +struct Kitty { + VkPipeline pipeline; + VkPipelineLayout pipeline_layout; + + const char *vertex_path; + const char *fragment_path; + + struct VertexBuffer vertex_buffer; + struct da_Attatchment attatchments; + + uint32_t push_constant_size; + void *next_push_constant; + + VkDescriptorPool descriptor_pool; + VkDescriptorSetLayout descriptor_set_layout; + VkDescriptorSet descriptor_sets[MAX_FRAMES_IN_FLIGHT]; +}; + +struct Kitty *kitty_make() { + struct Kitty *thingy = malloc(sizeof(struct Kitty)); + memset(thingy, 0, sizeof(struct Kitty)); + dyn_array_create_inplace(&thingy->attatchments); + dyn_array_create_inplace(&thingy->vertex_buffer.format); + meow("alloced a thingy at %p", thingy); + return thingy; +} + +void kitty_set_push_constant_size(struct Kitty *thingy, uint32_t size) { + thingy->push_constant_size = size; +} +void kitty_set_vertex_shader(struct Kitty *thingy, const char *path) { + thingy->vertex_path = path; + meow("vertex path is %s", path); +} +void kitty_set_fragment_shader(struct Kitty *thingy, const char *path) { + thingy->fragment_path = path; + meow("fragment path is %s", path); +} +void kitty_set_vertex_buffer(struct Kitty *thingy, void *data, uint32_t count, + int vertex_size) { + thingy->vertex_buffer.data = data; + thingy->vertex_buffer.count = count; + thingy->vertex_buffer.vertex_size = vertex_size; +} +void kitty_add_vertex_buffer_format(struct Kitty *thingy, + enum VkFormat format) { + meow("attatched %d to the buffer", format); + dyn_array_append(&thingy->vertex_buffer.format, format); +} +void kitty_attatch_image(struct Kitty *thingy, const char *path) { + struct Attatchment attatchment = {0}; + attatchment.type = CAT_ATTATCH_IMAGE; + attatchment.image.path = path; + dyn_array_append(&thingy->attatchments, attatchment); + meow("image was attatched"); +} +int kitty_attatch_ubo(struct Kitty *thingy, uint32_t size) { + struct Attatchment attatchment = {0}; + attatchment.type = CAT_ATTATCH_UBO; + attatchment.ubo.size = size; + dyn_array_append(&thingy->attatchments, attatchment); + meow("ubo of size %d was attatched", size); + return thingy->attatchments.count - 1; +} + +void kitty_finalise(struct Vk *state, struct Kitty *kitty) { + meow("this thingy has %d ubos", kitty->attatchments.count); + + if (kitty->vertex_path == NULL) { + meow("a kitty has no vertex shader :c"); + return; + } + if (kitty->fragment_path == NULL) { + meow("a kitty has no fragment shader :c"); + return; + } + + VkDescriptorSetLayoutBinding + descriptor_set_layout_binding[kitty->attatchments.count]; + memset(descriptor_set_layout_binding, 0, + sizeof(VkDescriptorSetLayoutBinding) * kitty->attatchments.count); + for (int index = 0; index < kitty->attatchments.count; index++) { + if (kitty->attatchments.items[index].type == CAT_ATTATCH_UBO) { + descriptor_set_layout_binding[index].binding = index; + descriptor_set_layout_binding[index].descriptorType = + VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER; + descriptor_set_layout_binding[index].descriptorCount = 1; + descriptor_set_layout_binding[index].stageFlags = VK_SHADER_STAGE_ALL; + descriptor_set_layout_binding[index].pImmutableSamplers = NULL; + } else { + descriptor_set_layout_binding[index].binding = index; + descriptor_set_layout_binding[index].descriptorType = + VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER; + descriptor_set_layout_binding[index].descriptorCount = 1; + descriptor_set_layout_binding[index].stageFlags = VK_SHADER_STAGE_ALL; + descriptor_set_layout_binding[index].pImmutableSamplers = NULL; + } + } + VkDescriptorSetLayoutCreateInfo layout_info = {0}; + layout_info.sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO; + layout_info.bindingCount = kitty->attatchments.count; + layout_info.pBindings = descriptor_set_layout_binding; + CHECK_VK_RESULT(vkCreateDescriptorSetLayout(state->device, &layout_info, NULL, + &kitty->descriptor_set_layout)); + + int vert_size, frag_size; + uint32_t *vert_shader = + read_binary_file((char *)kitty->vertex_path, &vert_size); + uint32_t *frag_shader = + read_binary_file((char *)kitty->fragment_path, &frag_size); + + VkShaderModule vert_module = + create_shader_module(state, vert_shader, vert_size); + VkShaderModule frag_module = + create_shader_module(state, frag_shader, frag_size); + + VkPipelineShaderStageCreateInfo vert_stage_info = {0}; + vert_stage_info.sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO; + vert_stage_info.stage = VK_SHADER_STAGE_VERTEX_BIT; + vert_stage_info.module = vert_module; + vert_stage_info.pName = "main"; + + VkPipelineShaderStageCreateInfo frag_stage_info = {0}; + frag_stage_info.sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO; + frag_stage_info.stage = VK_SHADER_STAGE_FRAGMENT_BIT; + frag_stage_info.module = frag_module; + frag_stage_info.pName = "main"; + + VkPipelineShaderStageCreateInfo shader_stage_info[] = {vert_stage_info, + frag_stage_info}; + VkVertexInputBindingDescription vertex_binding_description = {0}; + vertex_binding_description.binding = 0; + vertex_binding_description.stride = kitty->vertex_buffer.vertex_size; + vertex_binding_description.inputRate = VK_VERTEX_INPUT_RATE_VERTEX; + + VkVertexInputAttributeDescription *vertex_attribute_descriptions = + alloca(sizeof(VkVertexInputAttributeDescription) * + kitty->vertex_buffer.format.count); + int offset = 0; + for (int i = 0; i < kitty->vertex_buffer.format.count; i++) { + vertex_attribute_descriptions[i].binding = 0; + vertex_attribute_descriptions[i].location = i; + vertex_attribute_descriptions[i].format = + kitty->vertex_buffer.format.items[i]; + vertex_attribute_descriptions[i].offset = offset; + offset += get_size_of_format(kitty->vertex_buffer.format.items[i]); + } + + VkPipelineVertexInputStateCreateInfo vertex_input_info = {0}; + vertex_input_info.sType = + VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_STATE_CREATE_INFO; + vertex_input_info.vertexBindingDescriptionCount = 1; + vertex_input_info.pVertexBindingDescriptions = &vertex_binding_description; + vertex_input_info.vertexAttributeDescriptionCount = + kitty->vertex_buffer.format.count; + vertex_input_info.pVertexAttributeDescriptions = + vertex_attribute_descriptions; + + VkPipelineInputAssemblyStateCreateInfo input_assembly_info = {0}; + input_assembly_info.sType = + VK_STRUCTURE_TYPE_PIPELINE_INPUT_ASSEMBLY_STATE_CREATE_INFO; + input_assembly_info.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + input_assembly_info.primitiveRestartEnable = VK_FALSE; + + VkPipelineViewportStateCreateInfo viewport_state_info = {0}; + viewport_state_info.sType = + VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_STATE_CREATE_INFO; + viewport_state_info.viewportCount = 1; + viewport_state_info.scissorCount = 1; + + VkPipelineRasterizationStateCreateInfo rasterizer_info = {0}; + rasterizer_info.sType = + VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_STATE_CREATE_INFO; + rasterizer_info.depthClampEnable = VK_FALSE; + rasterizer_info.rasterizerDiscardEnable = VK_FALSE; + rasterizer_info.polygonMode = VK_POLYGON_MODE_FILL; + rasterizer_info.lineWidth = 1.0f; + rasterizer_info.cullMode = VK_CULL_MODE_BACK_BIT; + rasterizer_info.frontFace = VK_FRONT_FACE_CLOCKWISE; + rasterizer_info.depthBiasEnable = VK_FALSE; + + VkPipelineMultisampleStateCreateInfo multisampling_info = {0}; + multisampling_info.sType = + VK_STRUCTURE_TYPE_PIPELINE_MULTISAMPLE_STATE_CREATE_INFO; + multisampling_info.sampleShadingEnable = VK_FALSE; + multisampling_info.rasterizationSamples = VK_SAMPLE_COUNT_1_BIT; + + VkPipelineColorBlendAttachmentState color_blend_attachment = {0}; + color_blend_attachment.colorWriteMask = + VK_COLOR_COMPONENT_R_BIT | VK_COLOR_COMPONENT_G_BIT | + VK_COLOR_COMPONENT_B_BIT | VK_COLOR_COMPONENT_A_BIT; + color_blend_attachment.blendEnable = VK_FALSE; + + VkPipelineColorBlendStateCreateInfo color_blend_info = {0}; + color_blend_info.sType = + VK_STRUCTURE_TYPE_PIPELINE_COLOR_BLEND_STATE_CREATE_INFO; + color_blend_info.logicOpEnable = VK_FALSE; + color_blend_info.logicOp = VK_LOGIC_OP_COPY; + color_blend_info.attachmentCount = 1; + color_blend_info.pAttachments = &color_blend_attachment; + + /* VkPushConstantRange push_constant_range = {0}; */ + /* push_constant_range.offset = 0; */ + /* push_constant_range.size = thingy->push_constant_size; */ + /* push_constant_range.stageFlags = VK_SHADER_STAGE_ALL; */ + + VkPipelineLayoutCreateInfo pipeline_layout_info = {0}; + pipeline_layout_info.sType = VK_STRUCTURE_TYPE_PIPELINE_LAYOUT_CREATE_INFO; + pipeline_layout_info.setLayoutCount = 1; + pipeline_layout_info.pSetLayouts = &kitty->descriptor_set_layout; + pipeline_layout_info.pushConstantRangeCount = 0; // TODO + pipeline_layout_info.pPushConstantRanges = NULL; + CHECK_VK_RESULT(vkCreatePipelineLayout(state->device, &pipeline_layout_info, + NULL, &kitty->pipeline_layout)); + + VkDynamicState dynamic_states[] = {VK_DYNAMIC_STATE_VIEWPORT, + VK_DYNAMIC_STATE_SCISSOR}; + + VkPipelineDynamicStateCreateInfo dynamic_state_info = {0}; + dynamic_state_info.sType = + VK_STRUCTURE_TYPE_PIPELINE_DYNAMIC_STATE_CREATE_INFO; + dynamic_state_info.pDynamicStates = dynamic_states; + dynamic_state_info.dynamicStateCount = + sizeof(dynamic_states) / sizeof(dynamic_states[0]); + + VkGraphicsPipelineCreateInfo pipeline_info = {0}; + pipeline_info.sType = VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO; + pipeline_info.stageCount = 2; + pipeline_info.pStages = shader_stage_info; + pipeline_info.pVertexInputState = &vertex_input_info; + pipeline_info.pInputAssemblyState = &input_assembly_info; + pipeline_info.pViewportState = &viewport_state_info; + pipeline_info.pRasterizationState = &rasterizer_info; + pipeline_info.pMultisampleState = &multisampling_info; + pipeline_info.pDepthStencilState = NULL; + pipeline_info.pColorBlendState = &color_blend_info; + pipeline_info.pDynamicState = &dynamic_state_info; + pipeline_info.layout = kitty->pipeline_layout; + pipeline_info.renderPass = state->render_pass; + pipeline_info.subpass = 0; + pipeline_info.basePipelineHandle = NULL; + pipeline_info.basePipelineIndex = -1; + + CHECK_VK_RESULT(vkCreateGraphicsPipelines( + state->device, NULL, 1, &pipeline_info, NULL, &kitty->pipeline)); + + int ubo_count = 0; + int image_count = 0; + for (int i = 0; i < kitty->attatchments.count; i++) { + switch (kitty->attatchments.items[i].type) { + case CAT_ATTATCH_UBO: + ubo_count++; + break; + case CAT_ATTATCH_IMAGE: + image_count++; + break; + } + } + + VkDescriptorPoolSize pool_sizes[2] = {0}; + pool_sizes[0].type = VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER; + pool_sizes[0].descriptorCount = MAX_FRAMES_IN_FLIGHT; + pool_sizes[1].type = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER; + pool_sizes[1].descriptorCount = MAX_FRAMES_IN_FLIGHT; + + VkDescriptorPoolCreateInfo pool_info = {0}; + pool_info.sType = VK_STRUCTURE_TYPE_DESCRIPTOR_POOL_CREATE_INFO; + pool_info.poolSizeCount = 2; + pool_info.pPoolSizes = pool_sizes; + pool_info.maxSets = 10; + + CHECK_VK_RESULT(vkCreateDescriptorPool(state->device, &pool_info, NULL, + &kitty->descriptor_pool)); + + for (int index = 0; index < kitty->attatchments.count; index++) { + if (kitty->attatchments.items[index].type != CAT_ATTATCH_IMAGE) { + continue; + } + struct Attatchment *atch = &kitty->attatchments.items[index]; + atch->image.pixels = load_image(atch->image.path, &atch->image.dims); + atch->image.size = atch->image.dims.x * atch->image.dims.y * 4; + + VkBuffer staging_buffer; + VkDeviceMemory staging_buffer_memory; + create_buffer(state, atch->image.size, VK_BUFFER_USAGE_TRANSFER_SRC_BIT, + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | + VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, + &staging_buffer, &staging_buffer_memory); + + void *data; + vkMapMemory(state->device, staging_buffer_memory, 0, atch->image.size, 0, + &data); + memcpy(data, atch->image.pixels, atch->image.size); + vkUnmapMemory(state->device, staging_buffer_memory); + + create_image(state, atch->image.dims, VK_FORMAT_R8G8B8A8_SRGB, + VK_IMAGE_TILING_OPTIMAL, + VK_IMAGE_USAGE_TRANSFER_DST_BIT | VK_IMAGE_USAGE_SAMPLED_BIT, + VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT, &atch->image.image, + &atch->image.memory); + + transition_image_layout(state, atch->image.image, VK_FORMAT_R8G8B8A8_SRGB, + VK_IMAGE_LAYOUT_UNDEFINED, + VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL); + copy_buffer_to_image(state, staging_buffer, atch->image.image, + atch->image.dims); + transition_image_layout(state, atch->image.image, VK_FORMAT_R8G8B8A8_SRGB, + VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, + VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL); + + vkDestroyBuffer(state->device, staging_buffer, NULL); + vkFreeMemory(state->device, staging_buffer_memory, NULL); + + VkImageViewCreateInfo view_info = {0}; + view_info.sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO; + view_info.image = atch->image.image; + view_info.viewType = VK_IMAGE_VIEW_TYPE_2D; + view_info.format = VK_FORMAT_R8G8B8A8_SRGB; + view_info.subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT; + view_info.subresourceRange.baseMipLevel = 0; + view_info.subresourceRange.levelCount = 1; + view_info.subresourceRange.baseArrayLayer = 0; + view_info.subresourceRange.layerCount = 1; + CHECK_VK_RESULT( + vkCreateImageView(state->device, &view_info, NULL, &atch->image.view)); + + VkSamplerCreateInfo sampler_info = {0}; + sampler_info.sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO; + sampler_info.magFilter = VK_FILTER_NEAREST; + sampler_info.minFilter = VK_FILTER_NEAREST; + sampler_info.addressModeU = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + sampler_info.addressModeV = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + sampler_info.addressModeW = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + sampler_info.anisotropyEnable = VK_FALSE; + sampler_info.maxAnisotropy = 1.0f; + sampler_info.borderColor = VK_BORDER_COLOR_INT_TRANSPARENT_BLACK; + sampler_info.unnormalizedCoordinates = VK_FALSE; + sampler_info.compareEnable = VK_FALSE; + sampler_info.compareOp = VK_COMPARE_OP_ALWAYS; + sampler_info.mipmapMode = VK_SAMPLER_MIPMAP_MODE_NEAREST; + sampler_info.mipLodBias = 0.0f; + sampler_info.minLod = 0.0; + sampler_info.maxLod = 0.0; + CHECK_VK_RESULT(vkCreateSampler(state->device, &sampler_info, NULL, + &atch->image.sampler)); + } + + for (int index = 0; index < kitty->attatchments.count; index++) { + if (kitty->attatchments.items[index].type != CAT_ATTATCH_UBO) { + continue; + } + for (int i = 0; i < MAX_FRAMES_IN_FLIGHT; i++) { + create_buffer(state, kitty->attatchments.items[index].ubo.size, + VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT, + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | + VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, + &kitty->attatchments.items[index].ubo.buffer[i], + &kitty->attatchments.items[index].ubo.memory[i]); + + void *pointer_to_mapped; + vkMapMemory(state->device, kitty->attatchments.items[index].ubo.memory[i], + 0, kitty->attatchments.items[index].ubo.size, 0, + &pointer_to_mapped); + kitty->attatchments.items[index].ubo.mapped[i] = pointer_to_mapped; + } + } + + VkDescriptorSetLayout set_layouts[MAX_FRAMES_IN_FLIGHT]; + for (int i = 0; i < MAX_FRAMES_IN_FLIGHT; i++) { + set_layouts[i] = kitty->descriptor_set_layout; + } + + VkDescriptorSetAllocateInfo alloc_info = {0}; + alloc_info.sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO; + alloc_info.descriptorPool = kitty->descriptor_pool; + alloc_info.descriptorSetCount = MAX_FRAMES_IN_FLIGHT; + meow("allocing %d desc sets", MAX_FRAMES_IN_FLIGHT); + alloc_info.pSetLayouts = set_layouts; + CHECK_VK_RESULT(vkAllocateDescriptorSets(state->device, &alloc_info, + kitty->descriptor_sets)); + + for (int index = 0; index < kitty->attatchments.count; index++) { + for (int i = 0; i < MAX_FRAMES_IN_FLIGHT; i++) { + switch (kitty->attatchments.items[index].type) { + case CAT_ATTATCH_UBO:; + VkDescriptorBufferInfo buffer_info = {0}; + buffer_info.buffer = kitty->attatchments.items[index].ubo.buffer[i]; + buffer_info.offset = 0; + buffer_info.range = kitty->attatchments.items[index].ubo.size; + + VkWriteDescriptorSet write_buffer = {0}; + write_buffer.sType = VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET; + write_buffer.dstSet = kitty->descriptor_sets[i]; + write_buffer.dstBinding = index; + write_buffer.dstArrayElement = 0; + write_buffer.descriptorType = VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER; + write_buffer.descriptorCount = 1; + write_buffer.pBufferInfo = &buffer_info; + vkUpdateDescriptorSets(state->device, 1, &write_buffer, 0, NULL); + break; + case CAT_ATTATCH_IMAGE:; + VkDescriptorImageInfo image_info = {0}; + image_info.imageLayout = VK_IMAGE_LAYOUT_READ_ONLY_OPTIMAL; + image_info.imageView = kitty->attatchments.items[index].image.view; + image_info.sampler = kitty->attatchments.items[index].image.sampler; + + VkWriteDescriptorSet write_image = {0}; + write_image.sType = VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET; + write_image.dstSet = kitty->descriptor_sets[i]; + write_image.dstBinding = index; + write_image.dstArrayElement = 0; + write_image.descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER; + write_image.descriptorCount = 1; + write_image.pImageInfo = &image_info; + vkUpdateDescriptorSets(state->device, 1, &write_image, 0, NULL); + break; + } + } + } + + VkDeviceSize size = + kitty->vertex_buffer.count * kitty->vertex_buffer.vertex_size; + VkBuffer staging_buffer; + VkDeviceMemory staging_buffer_memory; + + create_buffer(state, size, VK_BUFFER_USAGE_TRANSFER_SRC_BIT, + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | + VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, + &staging_buffer, &staging_buffer_memory); + + void *data; + vkMapMemory(state->device, staging_buffer_memory, 0, size, 0, &data); + memcpy(data, kitty->vertex_buffer.data, size); + vkUnmapMemory(state->device, staging_buffer_memory); + + create_buffer(state, size, + VK_BUFFER_USAGE_TRANSFER_DST_BIT | + VK_BUFFER_USAGE_VERTEX_BUFFER_BIT, + VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT, + &kitty->vertex_buffer.buffer, &kitty->vertex_buffer.memory); + + copy_buffer(state, staging_buffer, kitty->vertex_buffer.buffer, size); + + vkDestroyBuffer(state->device, staging_buffer, NULL); + vkFreeMemory(state->device, staging_buffer_memory, NULL); + + vkDestroyShaderModule(state->device, vert_module, NULL); + vkDestroyShaderModule(state->device, frag_module, NULL); + + free(vert_shader); + free(frag_shader); + + dyn_array_append(&state->kitties, kitty); + meow("appended a kitty, len is now %d", state->kitties.count); +} + +void free_kitty(struct Vk *state, struct Kitty *kitty) { + vkDeviceWaitIdle(state->device); + vkDestroyBuffer(state->device, kitty->vertex_buffer.buffer, NULL); + vkFreeMemory(state->device, kitty->vertex_buffer.memory, NULL); + vkDestroyDescriptorPool(state->device, kitty->descriptor_pool, NULL); + vkDestroyDescriptorSetLayout(state->device, kitty->descriptor_set_layout, + NULL); + vkDestroyPipelineLayout(state->device, kitty->pipeline_layout, NULL); + vkDestroyPipeline(state->device, kitty->pipeline, NULL); + for (int i = 0; i < kitty->attatchments.count; i++) { + switch (kitty->attatchments.items[i].type) { + case CAT_ATTATCH_UBO: + for (int j = 0; j < MAX_FRAMES_IN_FLIGHT; j++) { + vkDestroyBuffer(state->device, + kitty->attatchments.items[i].ubo.buffer[j], NULL); + vkFreeMemory(state->device, kitty->attatchments.items[i].ubo.memory[j], + NULL); + } + break; + case CAT_ATTATCH_IMAGE: + vkDestroyImage(state->device, kitty->attatchments.items[i].image.image, + NULL); + vkFreeMemory(state->device, kitty->attatchments.items[i].image.memory, + NULL); + vkDestroyImageView(state->device, kitty->attatchments.items[i].image.view, + NULL); + vkDestroySampler(state->device, + kitty->attatchments.items[i].image.sampler, NULL); + break; + } + } + + dyn_array_destroy(&kitty->vertex_buffer.format); + dyn_array_destroy(&kitty->attatchments); + + memset(kitty, 0, sizeof(struct Kitty)); + free(kitty); +} + +void kitty_set_next_push_constant(struct Kitty *thingy, void *data) { + thingy->next_push_constant = data; +} + +void kitty_set_next_ubo(struct Vk *state, struct Kitty *thingy, int index, + void *data) { + void *dst = + thingy->attatchments.items[index].ubo.mapped[state->current_frame]; + void *src = data; + int size = thingy->attatchments.items[index].ubo.size; + /* meow("index %d, number %f", index, *(float *)data); */ + memcpy(dst, src, size); +} + +void kitty_draw(struct Vk *state, uint32_t image_index, struct Kitty *thingy) { + struct InFlightObjects flight = state->flights[state->current_frame]; + vkCmdBindPipeline(flight.command_buffer, VK_PIPELINE_BIND_POINT_GRAPHICS, + thingy->pipeline); + + VkViewport viewport = {0}; + viewport.x = 0.0f; + viewport.y = 0.0f; + viewport.width = state->width; + viewport.height = state->heigh; + viewport.minDepth = 0.0f; + viewport.maxDepth = 1.0f; + vkCmdSetViewport(flight.command_buffer, 0, 1, &viewport); + + VkRect2D scissor = {0}; + scissor.offset = (VkOffset2D){0, 0}; + scissor.extent = (VkExtent2D){state->width, state->heigh}; + vkCmdSetScissor(flight.command_buffer, 0, 1, &scissor); + + VkBuffer vertex_buffer[] = {thingy->vertex_buffer.buffer}; + VkDeviceSize offsets[] = {0}; + vkCmdBindVertexBuffers(flight.command_buffer, 0, 1, vertex_buffer, offsets); + + vkCmdBindDescriptorSets( + flight.command_buffer, VK_PIPELINE_BIND_POINT_GRAPHICS, + thingy->pipeline_layout, 0, 1, + &thingy->descriptor_sets[state->current_frame], 0, NULL); + + vkCmdDraw(flight.command_buffer, thingy->vertex_buffer.count, 1, 0, 0); +} diff --git a/kitty.h b/kitty.h new file mode 100644 index 0000000..4087762 --- /dev/null +++ b/kitty.h @@ -0,0 +1,26 @@ +#ifndef INCLUDE_KITTY +#define INCLUDE_KITTY + +#include + +struct Kitty; + +struct Kitty *kitty_make(); +void kitty_set_vertex_shader(struct Kitty *thingy, const char *path); +void kitty_set_fragment_shader(struct Kitty *thingy, const char *path); +void kitty_set_vertex_buffer(struct Kitty *thingy, void *data, uint32_t count, + int vertex_size); +void kitty_add_vertex_buffer_format(struct Kitty *thingy, enum VkFormat format); +int kitty_attatch_ubo(struct Kitty *thingy, uint32_t size); +void kitty_finalise(struct Vk *state, struct Kitty *thingy); + +void kitty_set_next_push_constant(struct Kitty *thingy, void *data); +void kitty_set_next_ubo(struct Vk *state, struct Kitty *thingy, int index, + void *data); +void kitty_draw(struct Vk *state, uint32_t image_index, struct Kitty *thingy); + +void free_kitty(struct Vk *state, struct Kitty *kitty); + +void kitty_attatch_image(struct Kitty *thingy, const char *path); +void kitty_set_push_constant_size(struct Kitty *thingy, uint32_t size); +#endif // INCLUDE_KITTY diff --git a/log.c b/log.c new file mode 100644 index 0000000..99d689e --- /dev/null +++ b/log.c @@ -0,0 +1,38 @@ +#include +#include +#include +#include +#include + +static void full_write(int fd, const char *buf, size_t len) { + while (len > 0) { + ssize_t ret = write(fd, buf, len); + + if ((ret == -1) && (errno != EINTR)) + break; + + buf += (size_t)ret; + len -= (size_t)ret; + } +} + +void print_backtrace() { + static const char start[] = "BACKTRACE ------------\n"; + static const char end[] = "----------------------\n"; + + void *bt[1024]; + int bt_size; + char **bt_syms; + int i; + + bt_size = backtrace(bt, 1024); + bt_syms = backtrace_symbols(bt, bt_size); + full_write(STDERR_FILENO, start, strlen(start)); + for (i = 1; i < bt_size; i++) { + size_t len = strlen(bt_syms[i]); + full_write(STDERR_FILENO, bt_syms[i], len); + full_write(STDERR_FILENO, "\n", 1); + } + full_write(STDERR_FILENO, end, strlen(end)); + free(bt_syms); +} diff --git a/log.h b/log.h new file mode 100644 index 0000000..414391e --- /dev/null +++ b/log.h @@ -0,0 +1,14 @@ +#ifndef INCLUDE_WAYLANDCLIENT_LOG_H_ +#define INCLUDE_WAYLANDCLIENT_LOG_H_ + +#include +#include +#define meow(fmt, ...) \ + fprintf(stderr, "[%s:%d] " fmt "\n", __FILE__, __LINE__, ##__VA_ARGS__) +#define crash(fmt, ...) \ + fprintf(stderr, "[%s:%d] " fmt "\n", __FILE__, __LINE__, ##__VA_ARGS__); \ + *(int *)0 = 0; + +void print_backtrace(); + +#endif // INCLUDE_WAYLANDCLIENT_LOG_H_ diff --git a/main b/main new file mode 100755 index 0000000..89b125b Binary files /dev/null and b/main differ diff --git a/main.c b/main.c new file mode 100644 index 0000000..e114391 --- /dev/null +++ b/main.c @@ -0,0 +1,32 @@ +#include "comp.h" +#include "object.h" +#include "register.h" +#include "trig.c" + +#include + +int main() { + struct Vk *vk; + cat_Comp *state = cat_comp_init("meooow", 500, 500, &vk); + + struct Register *reg = make_register(); + reg_attatch_type(reg, "trig", OBJ_MAKE make_trig, OBJ_FREE free_trig, + OBJ_TICK trig_tick); + struct Scene *scene = make_scene(vk, reg); + + scene_queue_insert(scene, "trig", trig_make_args(scene)); + + while (!cat_comp_should_close(state)) { + scene_tick(scene); + cat_comp_draw(state); + } + + free_scene(scene); + free_register(reg); + + cat_comp_uninit(state); + + state = NULL; + reg = NULL; + scene = NULL; +} diff --git a/matrix.c b/matrix.c new file mode 100644 index 0000000..001b7e4 --- /dev/null +++ b/matrix.c @@ -0,0 +1,42 @@ +#include "matrix.h" + +#include + +struct mat3x3 multiply3x3(struct mat3x3 a, struct mat3x3 b) { + // c ij = a i1 · b 1j + a i2 · b 2j + a i3 · b 3j + struct mat3x3 res = {0}; + + res.m11 = a.m11 * b.m11 + a.m12 * b.m21 + a.m13 * b.m31; + res.m12 = a.m11 * b.m12 + a.m12 * b.m22 + a.m13 * b.m32; + res.m13 = a.m11 * b.m13 + a.m12 * b.m23 + a.m13 * b.m33; + res.m21 = a.m21 * b.m11 + a.m22 * b.m21 + a.m23 * b.m31; + res.m22 = a.m21 * b.m12 + a.m22 * b.m22 + a.m23 * b.m32; + res.m23 = a.m21 * b.m13 + a.m22 * b.m23 + a.m23 * b.m33; + res.m31 = a.m31 * b.m11 + a.m32 * b.m21 + a.m33 * b.m31; + res.m32 = a.m31 * b.m12 + a.m32 * b.m22 + a.m33 * b.m32; + res.m33 = a.m31 * b.m13 + a.m32 * b.m23 + a.m33 * b.m33; + return res; +} + +struct mat3x3 translate3x3(struct Vec2 offset) { + struct mat3x3 res = IDENT3x3; + res.m13 = offset.x; + res.m23 = offset.y; + return res; +} + +struct mat3x3 rotate3x3(float alpha) { + struct mat3x3 res = IDENT3x3; + res.m11 = cosf(alpha); + res.m12 = -sinf(alpha); + res.m21 = sinf(alpha); + res.m22 = cosf(alpha); + return res; +} + +struct mat3x3 scale3x3(struct Vec2 scale) { + struct mat3x3 res = IDENT3x3; + res.m11 = scale.x; + res.m22 = scale.y; + return res; +} diff --git a/matrix.h b/matrix.h new file mode 100644 index 0000000..98dea8a --- /dev/null +++ b/matrix.h @@ -0,0 +1,28 @@ +#ifndef INCLUDE_WAYLANDCLIENT_MATRIX_H_ +#define INCLUDE_WAYLANDCLIENT_MATRIX_H_ + +#include "types.h" + +struct mat3x3 { + float m11, m21, m31, _p1; + float m12, m22, m32, _p2; + float m13, m23, m33, _p3; +}; + +#define IDENT3x3 \ + (struct mat3x3){.m11 = 1.f, \ + .m21 = 0.f, \ + .m31 = 0.f, \ + .m12 = 0.f, \ + .m22 = 1.f, \ + .m32 = 0.f, \ + .m13 = 0.f, \ + .m23 = 0.f, \ + .m33 = 1.f} + +struct mat3x3 multiply3x3(struct mat3x3 a, struct mat3x3 b); +struct mat3x3 translate3x3(struct Vec2 offset); +struct mat3x3 rotate3x3(float alpha); +struct mat3x3 scale3x3(struct Vec2 scale); + +#endif // INCLUDE_WAYLANDCLIENT_MATRIX_H_ diff --git a/object.c b/object.c new file mode 100644 index 0000000..6c12e60 --- /dev/null +++ b/object.c @@ -0,0 +1,89 @@ +#include "object.h" +#include "allocator.h" +#include "log.h" +#include "register.h" + +#include +#include +#include + +struct Scene *make_scene(struct Vk *vk, struct Register *reg) { + struct Scene *scene = malloc(sizeof(struct Scene)); + memset(scene, 0, sizeof(struct Scene)); + scene->vk = vk; + scene->mem = make_mem(2048); + if (reg == NULL) { + scene->reg = make_register(); + } else { + scene->reg = reg; + } + + dyn_array_create_inplace(&scene->objects); + dyn_array_create_inplace(&scene->insert_queue); + + struct timespec time; + clock_gettime(CLOCK_MONOTONIC, &time); + scene->msecs = time.tv_sec * 1000000 + time.tv_nsec / 1000; + + return scene; +} + +void free_scene(struct Scene *scene) { + for (int i = 0; i < scene->objects.count; i++) { + scene->objects.items[i]->type->free(scene, scene->objects.items[i]->data); + mem_free(scene->mem, scene->objects.items[i]); + } + dyn_array_destroy(&scene->objects); + for (int i = 0; i < scene->insert_queue.count; i++) { + scene->insert_queue.items[i]->type->free( + scene, scene->insert_queue.items[i]->data); + mem_free(scene->mem, scene->insert_queue.items[i]); + } + dyn_array_destroy(&scene->insert_queue); + uninit_mem(scene->mem); + + memset(scene, 0, sizeof(struct Scene)); + free(scene); +} + +void scene_tick(struct Scene *scene) { + struct timespec time; + clock_gettime(CLOCK_MONOTONIC, &time); + long msecs = time.tv_sec * 1000000 + time.tv_nsec / 1000; + msecs *= 1.0f; + scene->delta_secs = (float)(msecs - scene->msecs) / 1000; + + if (scene->insert_queue.count > 0) { + for (int i = 0; i < scene->insert_queue.count; i++) { + dyn_array_append(&scene->objects, scene->insert_queue.items[i]); + } + dyn_array_reset(&scene->insert_queue); + } + for (int i = 0; i < scene->objects.count; i++) { + scene->objects.items[i]->type->tick(scene, scene->objects.items[i]->data); + } + if (scene->insert_queue.count > 0) { + for (int i = 0; i < scene->insert_queue.count; i++) { + dyn_array_append(&scene->objects, scene->insert_queue.items[i]); + } + dyn_array_reset(&scene->insert_queue); + } + + scene->msecs = msecs; +} + +void scene_queue_insert(struct Scene *scene, char *name, void *object_data) { + struct Type *type = reg_get_type(scene->reg, name); + struct Object *object = mem_malloc(scene->mem, sizeof(struct Object)); + object->type = type; + object->data = object_data; + dyn_array_append(&scene->insert_queue, object); +} +void scene_queue_insert_from_data(struct Scene *scene, char *name, char *data, + int len) { + struct Type *type = reg_get_type(scene->reg, name); + struct Object *object = mem_malloc(scene->mem, sizeof(struct Object)); + object->type = type; + object->data = type->make(scene, data, len); + dyn_array_append(&scene->insert_queue, object); +} diff --git a/object.h b/object.h new file mode 100644 index 0000000..c9e409b --- /dev/null +++ b/object.h @@ -0,0 +1,30 @@ +#ifndef INCLUDE_WAYLANDCLIENT_OBJECT_H_ +#define INCLUDE_WAYLANDCLIENT_OBJECT_H_ + +#include "dynarray.h" +// #include "vulkan_internal.h" + +dyn_array_define(da_Object, struct Object *); + +struct Scene { + struct Vk *vk; + struct Mem *mem; + struct da_Object objects; + struct da_Object insert_queue; + struct Register *reg; + + float delta_secs; + long msecs; +}; + +struct Object { + struct Type *type; + void *data; +}; + +struct Scene *make_scene(struct Vk *vk, struct Register *reg); +void free_scene(struct Scene *scene); +void scene_tick(struct Scene *scene); +void scene_queue_insert(struct Scene *scene, char *name, void *object_data); + +#endif // INCLUDE_WAYLANDCLIENT_OBJECT_H_ diff --git a/register.c b/register.c new file mode 100644 index 0000000..e96457b --- /dev/null +++ b/register.c @@ -0,0 +1,36 @@ +#include "register.h" +#include "dynarray.h" +#include "log.h" +#include "object.h" + +#include +#include + +struct Register *make_register() { + struct Register *reg = malloc(sizeof(struct Register)); + dyn_array_create_inplace(®->reg); + return reg; +} + +void free_register(struct Register *reg) { + dyn_array_destroy(®->reg); + free(reg); +} + +void reg_attatch_type(struct Register *reg, char *name, + void *(*make)(struct Scene *, char *, int len), + void (*free)(struct Scene *, void *), + void (*tick)(struct Scene *, void *)) { + struct Type type = {name, make, free, tick}; + dyn_array_append(®->reg, type); +} + +struct Type *reg_get_type(struct Register *reg, char *name) { + for (int i = 0; i < reg->reg.count; i++) { + if (strcmp(reg->reg.items[i].name, name) == 0) { + return ®->reg.items[i]; + } + } + meow("did not find %s in register", name); + return NULL; +} diff --git a/register.h b/register.h new file mode 100644 index 0000000..038e471 --- /dev/null +++ b/register.h @@ -0,0 +1,33 @@ +#ifndef INCLUDE_WAYLANDCLIENT_REGISTER_H_ +#define INCLUDE_WAYLANDCLIENT_REGISTER_H_ + +#include "dynarray.h" +#include "object.h" + +dyn_array_define(da_Type, struct Type); + +struct Type { + char *name; + + void *(*make)(struct Scene *, char *, int len); + void (*free)(struct Scene *, void *); + void (*tick)(struct Scene *, void *); +}; + +struct Register { + struct da_Type reg; +}; + +#define OBJ_FREE (void (*)(struct Scene *, void *)) +#define OBJ_TICK (void (*)(struct Scene *, void *)) +#define OBJ_MAKE (void *(*)(struct Scene *, char *, int len)) + +struct Register *make_register(); +void free_register(struct Register *reg); +struct Type *reg_get_type(struct Register *reg, char *name); +void reg_attatch_type(struct Register *reg, char *name, + void *(*make)(struct Scene *, char *, int len), + void (*free)(struct Scene *, void *), + void (*tick)(struct Scene *, void *)); + +#endif // INCLUDE_WAYLANDCLIENT_REGISTER_H_ diff --git a/stb_image.h b/stb_image.h new file mode 100644 index 0000000..9eedabe --- /dev/null +++ b/stb_image.h @@ -0,0 +1,7988 @@ +/* stb_image - v2.30 - public domain image loader - http://nothings.org/stb + no warranty implied; use at your own risk + + Do this: + #define STB_IMAGE_IMPLEMENTATION + before you include this file in *one* C or C++ file to create the implementation. + + // i.e. it should look like this: + #include ... + #include ... + #include ... + #define STB_IMAGE_IMPLEMENTATION + #include "stb_image.h" + + You can #define STBI_ASSERT(x) before the #include to avoid using assert.h. + And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using malloc,realloc,free + + + QUICK NOTES: + Primarily of interest to game developers and other people who can + avoid problematic images and only need the trivial interface + + JPEG baseline & progressive (12 bpc/arithmetic not supported, same as stock IJG lib) + PNG 1/2/4/8/16-bit-per-channel + + TGA (not sure what subset, if a subset) + BMP non-1bpp, non-RLE + PSD (composited view only, no extra channels, 8/16 bit-per-channel) + + GIF (*comp always reports as 4-channel) + HDR (radiance rgbE format) + PIC (Softimage PIC) + PNM (PPM and PGM binary only) + + Animated GIF still needs a proper API, but here's one way to do it: + http://gist.github.com/urraka/685d9a6340b26b830d49 + + - decode from memory or through FILE (define STBI_NO_STDIO to remove code) + - decode from arbitrary I/O callbacks + - SIMD acceleration on x86/x64 (SSE2) and ARM (NEON) + + Full documentation under "DOCUMENTATION" below. + + +LICENSE + + See end of file for license information. + +RECENT REVISION HISTORY: + + 2.30 (2024-05-31) avoid erroneous gcc warning + 2.29 (2023-05-xx) optimizations + 2.28 (2023-01-29) many error fixes, security errors, just tons of stuff + 2.27 (2021-07-11) document stbi_info better, 16-bit PNM support, bug fixes + 2.26 (2020-07-13) many minor fixes + 2.25 (2020-02-02) fix warnings + 2.24 (2020-02-02) fix warnings; thread-local failure_reason and flip_vertically + 2.23 (2019-08-11) fix clang static analysis warning + 2.22 (2019-03-04) gif fixes, fix warnings + 2.21 (2019-02-25) fix typo in comment + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) bugfix, 1-bit BMP, 16-bitness query, fix warnings + 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; bugfixes + 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE detection on GCC + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; fixes + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 + RGB-format JPEG; remove white matting in PSD; + allocate large structures on the stack; + correct channel count for PNG & BMP + 2.10 (2016-01-22) avoid warning introduced in 2.09 + 2.09 (2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED + + See end of file for full revision history. + + + ============================ Contributors ========================= + + Image formats Extensions, features + Sean Barrett (jpeg, png, bmp) Jetro Lauha (stbi_info) + Nicolas Schulz (hdr, psd) Martin "SpartanJ" Golini (stbi_info) + Jonathan Dummer (tga) James "moose2000" Brown (iPhone PNG) + Jean-Marc Lienher (gif) Ben "Disch" Wenger (io callbacks) + Tom Seddon (pic) Omar Cornut (1/2/4-bit PNG) + Thatcher Ulrich (psd) Nicolas Guillemot (vertical flip) + Ken Miller (pgm, ppm) Richard Mitton (16-bit PSD) + github:urraka (animated gif) Junggon Kim (PNM comments) + Christopher Forseth (animated gif) Daniel Gibson (16-bit TGA) + socks-the-fox (16-bit PNG) + Jeremy Sawicki (handle all ImageNet JPGs) + Optimizations & bugfixes Mikhail Morozov (1-bit BMP) + Fabian "ryg" Giesen Anael Seghezzi (is-16-bit query) + Arseny Kapoulkine Simon Breuss (16-bit PNM) + John-Mark Allen + Carmelo J Fdez-Aguera + + Bug & warning fixes + Marc LeBlanc David Woo Guillaume George Martins Mozeiko + Christpher Lloyd Jerry Jansson Joseph Thomson Blazej Dariusz Roszkowski + Phil Jordan Dave Moore Roy Eltham + Hayaki Saito Nathan Reed Won Chun + Luke Graham Johan Duparc Nick Verigakis the Horde3D community + Thomas Ruf Ronny Chevalier github:rlyeh + Janez Zemva John Bartholomew Michal Cichon github:romigrou + Jonathan Blow Ken Hamada Tero Hanninen github:svdijk + Eugene Golushkov Laurent Gomila Cort Stratton github:snagar + Aruelien Pocheville Sergio Gonzalez Thibault Reuille github:Zelex + Cass Everitt Ryamond Barbiero github:grim210 + Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw + Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus + Josh Tobin Neil Bickford Matthew Gregan github:poppolopoppo + Julian Raschke Gregory Mullen Christian Floisand github:darealshinji + Baldur Karlsson Kevin Schmidt JR Smith github:Michaelangel007 + Brad Weinberger Matvey Cherevko github:mosra + Luca Sas Alexander Veselov Zack Middleton [reserved] + Ryan C. Gordon [reserved] [reserved] + DO NOT ADD YOUR NAME HERE + + Jacko Dirks + + To add your name to the credits, pick a random blank space in the middle and fill it. + 80% of merge conflicts on stb PRs are due to people adding their name at the end + of the credits. +*/ + +#ifndef STBI_INCLUDE_STB_IMAGE_H +#define STBI_INCLUDE_STB_IMAGE_H + +// DOCUMENTATION +// +// Limitations: +// - no 12-bit-per-channel JPEG +// - no JPEGs with arithmetic coding +// - GIF always returns *comp=4 +// +// Basic usage (see HDR discussion below for HDR usage): +// int x,y,n; +// unsigned char *data = stbi_load(filename, &x, &y, &n, 0); +// // ... process data if not NULL ... +// // ... x = width, y = height, n = # 8-bit components per pixel ... +// // ... replace '0' with '1'..'4' to force that many components per pixel +// // ... but 'n' will always be the number that it would have been if you said 0 +// stbi_image_free(data); +// +// Standard parameters: +// int *x -- outputs image width in pixels +// int *y -- outputs image height in pixels +// int *channels_in_file -- outputs # of image components in image file +// int desired_channels -- if non-zero, # of image components requested in result +// +// The return value from an image loader is an 'unsigned char *' which points +// to the pixel data, or NULL on an allocation failure or if the image is +// corrupt or invalid. The pixel data consists of *y scanlines of *x pixels, +// with each pixel consisting of N interleaved 8-bit components; the first +// pixel pointed to is top-left-most in the image. There is no padding between +// image scanlines or between pixels, regardless of format. The number of +// components N is 'desired_channels' if desired_channels is non-zero, or +// *channels_in_file otherwise. If desired_channels is non-zero, +// *channels_in_file has the number of components that _would_ have been +// output otherwise. E.g. if you set desired_channels to 4, you will always +// get RGBA output, but you can check *channels_in_file to see if it's trivially +// opaque because e.g. there were only 3 channels in the source image. +// +// An output image with N components has the following components interleaved +// in this order in each pixel: +// +// N=#comp components +// 1 grey +// 2 grey, alpha +// 3 red, green, blue +// 4 red, green, blue, alpha +// +// If image loading fails for any reason, the return value will be NULL, +// and *x, *y, *channels_in_file will be unchanged. The function +// stbi_failure_reason() can be queried for an extremely brief, end-user +// unfriendly explanation of why the load failed. Define STBI_NO_FAILURE_STRINGS +// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get slightly +// more user-friendly ones. +// +// Paletted PNG, BMP, GIF, and PIC images are automatically depalettized. +// +// To query the width, height and component count of an image without having to +// decode the full file, you can use the stbi_info family of functions: +// +// int x,y,n,ok; +// ok = stbi_info(filename, &x, &y, &n); +// // returns ok=1 and sets x, y, n if image is a supported format, +// // 0 otherwise. +// +// Note that stb_image pervasively uses ints in its public API for sizes, +// including sizes of memory buffers. This is now part of the API and thus +// hard to change without causing breakage. As a result, the various image +// loaders all have certain limits on image size; these differ somewhat +// by format but generally boil down to either just under 2GB or just under +// 1GB. When the decoded image would be larger than this, stb_image decoding +// will fail. +// +// Additionally, stb_image will reject image files that have any of their +// dimensions set to a larger value than the configurable STBI_MAX_DIMENSIONS, +// which defaults to 2**24 = 16777216 pixels. Due to the above memory limit, +// the only way to have an image with such dimensions load correctly +// is for it to have a rather extreme aspect ratio. Either way, the +// assumption here is that such larger images are likely to be malformed +// or malicious. If you do need to load an image with individual dimensions +// larger than that, and it still fits in the overall size limit, you can +// #define STBI_MAX_DIMENSIONS on your own to be something larger. +// +// =========================================================================== +// +// UNICODE: +// +// If compiling for Windows and you wish to use Unicode filenames, compile +// with +// #define STBI_WINDOWS_UTF8 +// and pass utf8-encoded filenames. Call stbi_convert_wchar_to_utf8 to convert +// Windows wchar_t filenames to utf8. +// +// =========================================================================== +// +// Philosophy +// +// stb libraries are designed with the following priorities: +// +// 1. easy to use +// 2. easy to maintain +// 3. good performance +// +// Sometimes I let "good performance" creep up in priority over "easy to maintain", +// and for best performance I may provide less-easy-to-use APIs that give higher +// performance, in addition to the easy-to-use ones. Nevertheless, it's important +// to keep in mind that from the standpoint of you, a client of this library, +// all you care about is #1 and #3, and stb libraries DO NOT emphasize #3 above all. +// +// Some secondary priorities arise directly from the first two, some of which +// provide more explicit reasons why performance can't be emphasized. +// +// - Portable ("ease of use") +// - Small source code footprint ("easy to maintain") +// - No dependencies ("ease of use") +// +// =========================================================================== +// +// I/O callbacks +// +// I/O callbacks allow you to read from arbitrary sources, like packaged +// files or some other source. Data read from callbacks are processed +// through a small internal buffer (currently 128 bytes) to try to reduce +// overhead. +// +// The three functions you must define are "read" (reads some bytes of data), +// "skip" (skips some bytes of data), "eof" (reports if the stream is at the end). +// +// =========================================================================== +// +// SIMD support +// +// The JPEG decoder will try to automatically use SIMD kernels on x86 when +// supported by the compiler. For ARM Neon support, you must explicitly +// request it. +// +// (The old do-it-yourself SIMD API is no longer supported in the current +// code.) +// +// On x86, SSE2 will automatically be used when available based on a run-time +// test; if not, the generic C versions are used as a fall-back. On ARM targets, +// the typical path is to have separate builds for NEON and non-NEON devices +// (at least this is true for iOS and Android). Therefore, the NEON support is +// toggled by a build flag: define STBI_NEON to get NEON loops. +// +// If for some reason you do not want to use any of SIMD code, or if +// you have issues compiling it, you can disable it entirely by +// defining STBI_NO_SIMD. +// +// =========================================================================== +// +// HDR image support (disable by defining STBI_NO_HDR) +// +// stb_image supports loading HDR images in general, and currently the Radiance +// .HDR file format specifically. You can still load any file through the existing +// interface; if you attempt to load an HDR file, it will be automatically remapped +// to LDR, assuming gamma 2.2 and an arbitrary scale factor defaulting to 1; +// both of these constants can be reconfigured through this interface: +// +// stbi_hdr_to_ldr_gamma(2.2f); +// stbi_hdr_to_ldr_scale(1.0f); +// +// (note, do not use _inverse_ constants; stbi_image will invert them +// appropriately). +// +// Additionally, there is a new, parallel interface for loading files as +// (linear) floats to preserve the full dynamic range: +// +// float *data = stbi_loadf(filename, &x, &y, &n, 0); +// +// If you load LDR images through this interface, those images will +// be promoted to floating point values, run through the inverse of +// constants corresponding to the above: +// +// stbi_ldr_to_hdr_scale(1.0f); +// stbi_ldr_to_hdr_gamma(2.2f); +// +// Finally, given a filename (or an open file or memory block--see header +// file for details) containing image data, you can query for the "most +// appropriate" interface to use (that is, whether the image is HDR or +// not), using: +// +// stbi_is_hdr(char *filename); +// +// =========================================================================== +// +// iPhone PNG support: +// +// We optionally support converting iPhone-formatted PNGs (which store +// premultiplied BGRA) back to RGB, even though they're internally encoded +// differently. To enable this conversion, call +// stbi_convert_iphone_png_to_rgb(1). +// +// Call stbi_set_unpremultiply_on_load(1) as well to force a divide per +// pixel to remove any premultiplied alpha *only* if the image file explicitly +// says there's premultiplied data (currently only happens in iPhone images, +// and only if iPhone convert-to-rgb processing is on). +// +// =========================================================================== +// +// ADDITIONAL CONFIGURATION +// +// - You can suppress implementation of any of the decoders to reduce +// your code footprint by #defining one or more of the following +// symbols before creating the implementation. +// +// STBI_NO_JPEG +// STBI_NO_PNG +// STBI_NO_BMP +// STBI_NO_PSD +// STBI_NO_TGA +// STBI_NO_GIF +// STBI_NO_HDR +// STBI_NO_PIC +// STBI_NO_PNM (.ppm and .pgm) +// +// - You can request *only* certain decoders and suppress all other ones +// (this will be more forward-compatible, as addition of new decoders +// doesn't require you to disable them explicitly): +// +// STBI_ONLY_JPEG +// STBI_ONLY_PNG +// STBI_ONLY_BMP +// STBI_ONLY_PSD +// STBI_ONLY_TGA +// STBI_ONLY_GIF +// STBI_ONLY_HDR +// STBI_ONLY_PIC +// STBI_ONLY_PNM (.ppm and .pgm) +// +// - If you use STBI_NO_PNG (or _ONLY_ without PNG), and you still +// want the zlib decoder to be available, #define STBI_SUPPORT_ZLIB +// +// - If you define STBI_MAX_DIMENSIONS, stb_image will reject images greater +// than that size (in either width or height) without further processing. +// This is to let programs in the wild set an upper bound to prevent +// denial-of-service attacks on untrusted data, as one could generate a +// valid image of gigantic dimensions and force stb_image to allocate a +// huge block of memory and spend disproportionate time decoding it. By +// default this is set to (1 << 24), which is 16777216, but that's still +// very big. + +#ifndef STBI_NO_STDIO +#include +#endif // STBI_NO_STDIO + +#define STBI_VERSION 1 + +enum +{ + STBI_default = 0, // only used for desired_channels + + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 +}; + +#include +typedef unsigned char stbi_uc; +typedef unsigned short stbi_us; + +#ifdef __cplusplus +extern "C" { +#endif + +#ifndef STBIDEF +#ifdef STB_IMAGE_STATIC +#define STBIDEF static +#else +#define STBIDEF extern +#endif +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// PRIMARY API - works on images of any type +// + +// +// load image by filename, open file, or memory buffer +// + +typedef struct +{ + int (*read) (void *user,char *data,int size); // fill 'data' with 'size' bytes. return number of bytes actually read + void (*skip) (void *user,int n); // skip the next 'n' bytes, or 'unget' the last -n bytes if negative + int (*eof) (void *user); // returns nonzero if we are at end of file/data +} stbi_io_callbacks; + +//////////////////////////////////// +// +// 8-bits-per-channel interface +// + +STBIDEF stbi_uc *stbi_load_from_memory (stbi_uc const *buffer, int len , int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk , void *user, int *x, int *y, int *channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO +STBIDEF stbi_uc *stbi_load (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); +// for stbi_load_from_file, file pointer is left pointing immediately after image +#endif + +#ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp); +#endif + +#ifdef STBI_WINDOWS_UTF8 +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input); +#endif + +//////////////////////////////////// +// +// 16-bits-per-channel interface +// + +STBIDEF stbi_us *stbi_load_16_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO +STBIDEF stbi_us *stbi_load_16 (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); +#endif + +//////////////////////////////////// +// +// float-per-channel interface +// +#ifndef STBI_NO_LINEAR + STBIDEF float *stbi_loadf_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); + STBIDEF float *stbi_loadf_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + + #ifndef STBI_NO_STDIO + STBIDEF float *stbi_loadf (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); + STBIDEF float *stbi_loadf_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); + #endif +#endif + +#ifndef STBI_NO_HDR + STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); + STBIDEF void stbi_hdr_to_ldr_scale(float scale); +#endif // STBI_NO_HDR + +#ifndef STBI_NO_LINEAR + STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); + STBIDEF void stbi_ldr_to_hdr_scale(float scale); +#endif // STBI_NO_LINEAR + +// stbi_is_hdr is always defined, but always returns false if STBI_NO_HDR +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user); +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); +#ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr (char const *filename); +STBIDEF int stbi_is_hdr_from_file(FILE *f); +#endif // STBI_NO_STDIO + + +// get a VERY brief reason for failure +// on most compilers (and ALL modern mainstream compilers) this is threadsafe +STBIDEF const char *stbi_failure_reason (void); + +// free the loaded image -- this is just free() +STBIDEF void stbi_image_free (void *retval_from_stbi_load); + +// get image dimensions & components without fully decoding +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, void *user); + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_info (char const *filename, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_file (FILE *f, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit (char const *filename); +STBIDEF int stbi_is_16_bit_from_file(FILE *f); +#endif + + + +// for image formats that explicitly notate that they have premultiplied alpha, +// we just return the colors as stored in the file. set this flag to force +// unpremultiplication. results are undefined if the unpremultiply overflow. +STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply); + +// indicate whether we should process iphone images back to canonical format, +// or just pass them through "as-is" +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert); + +// flip the image vertically, so the first pixel in the output array is the bottom left +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip); + +// as above, but only applies to images loaded on the thread that calls the function +// this function is only available if your compiler supports thread-local variables; +// calling it will fail to link if your compiler doesn't +STBIDEF void stbi_set_unpremultiply_on_load_thread(int flag_true_if_should_unpremultiply); +STBIDEF void stbi_convert_iphone_png_to_rgb_thread(int flag_true_if_should_convert); +STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip); + +// ZLIB client - used by PNG, available for other purposes + +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen); +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header); +STBIDEF char *stbi_zlib_decode_malloc(const char *buffer, int len, int *outlen); +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + +STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, int *outlen); +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + + +#ifdef __cplusplus +} +#endif + +// +// +//// end header file ///////////////////////////////////////////////////// +#endif // STBI_INCLUDE_STB_IMAGE_H + +#ifdef STB_IMAGE_IMPLEMENTATION + +#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ + || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ + || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ + || defined(STBI_ONLY_ZLIB) + #ifndef STBI_ONLY_JPEG + #define STBI_NO_JPEG + #endif + #ifndef STBI_ONLY_PNG + #define STBI_NO_PNG + #endif + #ifndef STBI_ONLY_BMP + #define STBI_NO_BMP + #endif + #ifndef STBI_ONLY_PSD + #define STBI_NO_PSD + #endif + #ifndef STBI_ONLY_TGA + #define STBI_NO_TGA + #endif + #ifndef STBI_ONLY_GIF + #define STBI_NO_GIF + #endif + #ifndef STBI_ONLY_HDR + #define STBI_NO_HDR + #endif + #ifndef STBI_ONLY_PIC + #define STBI_NO_PIC + #endif + #ifndef STBI_ONLY_PNM + #define STBI_NO_PNM + #endif +#endif + +#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) +#define STBI_NO_ZLIB +#endif + + +#include +#include // ptrdiff_t on osx +#include +#include +#include + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +#include // ldexp, pow +#endif + +#ifndef STBI_NO_STDIO +#include +#endif + +#ifndef STBI_ASSERT +#include +#define STBI_ASSERT(x) assert(x) +#endif + +#ifdef __cplusplus +#define STBI_EXTERN extern "C" +#else +#define STBI_EXTERN extern +#endif + + +#ifndef _MSC_VER + #ifdef __cplusplus + #define stbi_inline inline + #else + #define stbi_inline + #endif +#else + #define stbi_inline __forceinline +#endif + +#ifndef STBI_NO_THREAD_LOCALS + #if defined(__cplusplus) && __cplusplus >= 201103L + #define STBI_THREAD_LOCAL thread_local + #elif defined(__GNUC__) && __GNUC__ < 5 + #define STBI_THREAD_LOCAL __thread + #elif defined(_MSC_VER) + #define STBI_THREAD_LOCAL __declspec(thread) + #elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && !defined(__STDC_NO_THREADS__) + #define STBI_THREAD_LOCAL _Thread_local + #endif + + #ifndef STBI_THREAD_LOCAL + #if defined(__GNUC__) + #define STBI_THREAD_LOCAL __thread + #endif + #endif +#endif + +#if defined(_MSC_VER) || defined(__SYMBIAN32__) +typedef unsigned short stbi__uint16; +typedef signed short stbi__int16; +typedef unsigned int stbi__uint32; +typedef signed int stbi__int32; +#else +#include +typedef uint16_t stbi__uint16; +typedef int16_t stbi__int16; +typedef uint32_t stbi__uint32; +typedef int32_t stbi__int32; +#endif + +// should produce compiler error if size is wrong +typedef unsigned char validate_uint32[sizeof(stbi__uint32)==4 ? 1 : -1]; + +#ifdef _MSC_VER +#define STBI_NOTUSED(v) (void)(v) +#else +#define STBI_NOTUSED(v) (void)sizeof(v) +#endif + +#ifdef _MSC_VER +#define STBI_HAS_LROTL +#endif + +#ifdef STBI_HAS_LROTL + #define stbi_lrot(x,y) _lrotl(x,y) +#else + #define stbi_lrot(x,y) (((x) << (y)) | ((x) >> (-(y) & 31))) +#endif + +#if defined(STBI_MALLOC) && defined(STBI_FREE) && (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) +// ok +#elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) +// ok +#else +#error "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." +#endif + +#ifndef STBI_MALLOC +#define STBI_MALLOC(sz) malloc(sz) +#define STBI_REALLOC(p,newsz) realloc(p,newsz) +#define STBI_FREE(p) free(p) +#endif + +#ifndef STBI_REALLOC_SIZED +#define STBI_REALLOC_SIZED(p,oldsz,newsz) STBI_REALLOC(p,newsz) +#endif + +// x86/x64 detection +#if defined(__x86_64__) || defined(_M_X64) +#define STBI__X64_TARGET +#elif defined(__i386) || defined(_M_IX86) +#define STBI__X86_TARGET +#endif + +#if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) +// gcc doesn't support sse2 intrinsics unless you compile with -msse2, +// which in turn means it gets to use SSE2 everywhere. This is unfortunate, +// but previous attempts to provide the SSE2 functions with runtime +// detection caused numerous issues. The way architecture extensions are +// exposed in GCC/Clang is, sadly, not really suited for one-file libs. +// New behavior: if compiled with -msse2, we use SSE2 without any +// detection; if not, we don't use it at all. +#define STBI_NO_SIMD +#endif + +#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) +// Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid STBI__X64_TARGET +// +// 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the +// Windows ABI and VC++ as well as Windows DLLs don't maintain that invariant. +// As a result, enabling SSE2 on 32-bit MinGW is dangerous when not +// simultaneously enabling "-mstackrealign". +// +// See https://github.com/nothings/stb/issues/81 for more information. +// +// So default to no SSE2 on 32-bit MinGW. If you've read this far and added +// -mstackrealign to your build settings, feel free to #define STBI_MINGW_ENABLE_SSE2. +#define STBI_NO_SIMD +#endif + +#if !defined(STBI_NO_SIMD) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) +#define STBI_SSE2 +#include + +#ifdef _MSC_VER + +#if _MSC_VER >= 1400 // not VC6 +#include // __cpuid +static int stbi__cpuid3(void) +{ + int info[4]; + __cpuid(info,1); + return info[3]; +} +#else +static int stbi__cpuid3(void) +{ + int res; + __asm { + mov eax,1 + cpuid + mov res,edx + } + return res; +} +#endif + +#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + +#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) +{ + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; +} +#endif + +#else // assume GCC-style if not VC++ +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + +#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) +{ + // If we're even attempting to compile this on GCC/Clang, that means + // -msse2 is on, which means the compiler is allowed to use SSE2 + // instructions at will, and so are we. + return 1; +} +#endif + +#endif +#endif + +// ARM NEON +#if defined(STBI_NO_SIMD) && defined(STBI_NEON) +#undef STBI_NEON +#endif + +#ifdef STBI_NEON +#include +#ifdef _MSC_VER +#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name +#else +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) +#endif +#endif + +#ifndef STBI_SIMD_ALIGN +#define STBI_SIMD_ALIGN(type, name) type name +#endif + +#ifndef STBI_MAX_DIMENSIONS +#define STBI_MAX_DIMENSIONS (1 << 24) +#endif + +/////////////////////////////////////////////// +// +// stbi__context struct and start_xxx functions + +// stbi__context structure is our basic context used by all images, so it +// contains all the IO context, plus some basic image information +typedef struct +{ + stbi__uint32 img_x, img_y; + int img_n, img_out_n; + + stbi_io_callbacks io; + void *io_user_data; + + int read_from_callbacks; + int buflen; + stbi_uc buffer_start[128]; + int callback_already_read; + + stbi_uc *img_buffer, *img_buffer_end; + stbi_uc *img_buffer_original, *img_buffer_original_end; +} stbi__context; + + +static void stbi__refill_buffer(stbi__context *s); + +// initialize a memory-decode context +static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) +{ + s->io.read = NULL; + s->read_from_callbacks = 0; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = (stbi_uc *) buffer; + s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *) buffer+len; +} + +// initialize a callback-based context +static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, void *user) +{ + s->io = *c; + s->io_user_data = user; + s->buflen = sizeof(s->buffer_start); + s->read_from_callbacks = 1; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = s->buffer_start; + stbi__refill_buffer(s); + s->img_buffer_original_end = s->img_buffer_end; +} + +#ifndef STBI_NO_STDIO + +static int stbi__stdio_read(void *user, char *data, int size) +{ + return (int) fread(data,1,size,(FILE*) user); +} + +static void stbi__stdio_skip(void *user, int n) +{ + int ch; + fseek((FILE*) user, n, SEEK_CUR); + ch = fgetc((FILE*) user); /* have to read a byte to reset feof()'s flag */ + if (ch != EOF) { + ungetc(ch, (FILE *) user); /* push byte back onto stream if valid. */ + } +} + +static int stbi__stdio_eof(void *user) +{ + return feof((FILE*) user) || ferror((FILE *) user); +} + +static stbi_io_callbacks stbi__stdio_callbacks = +{ + stbi__stdio_read, + stbi__stdio_skip, + stbi__stdio_eof, +}; + +static void stbi__start_file(stbi__context *s, FILE *f) +{ + stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *) f); +} + +//static void stop_file(stbi__context *s) { } + +#endif // !STBI_NO_STDIO + +static void stbi__rewind(stbi__context *s) +{ + // conceptually rewind SHOULD rewind to the beginning of the stream, + // but we just rewind to the beginning of the initial buffer, because + // we only use it after doing 'test', which only ever looks at at most 92 bytes + s->img_buffer = s->img_buffer_original; + s->img_buffer_end = s->img_buffer_original_end; +} + +enum +{ + STBI_ORDER_RGB, + STBI_ORDER_BGR +}; + +typedef struct +{ + int bits_per_channel; + int num_channels; + int channel_order; +} stbi__result_info; + +#ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context *s); +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context *s); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__png_is16(stbi__context *s); +#endif + +#ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context *s); +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context *s); +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc); +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__psd_is16(stbi__context *s); +#endif + +#ifndef STBI_NO_HDR +static int stbi__hdr_test(stbi__context *s); +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_test(stbi__context *s); +static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_GIF +static int stbi__gif_test(stbi__context *s); +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp); +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PNM +static int stbi__pnm_test(stbi__context *s); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__pnm_is16(stbi__context *s); +#endif + +static +#ifdef STBI_THREAD_LOCAL +STBI_THREAD_LOCAL +#endif +const char *stbi__g_failure_reason; + +STBIDEF const char *stbi_failure_reason(void) +{ + return stbi__g_failure_reason; +} + +#ifndef STBI_NO_FAILURE_STRINGS +static int stbi__err(const char *str) +{ + stbi__g_failure_reason = str; + return 0; +} +#endif + +static void *stbi__malloc(size_t size) +{ + return STBI_MALLOC(size); +} + +// stb_image uses ints pervasively, including for offset calculations. +// therefore the largest decoded image size we can support with the +// current code, even on 64-bit targets, is INT_MAX. this is not a +// significant limitation for the intended use case. +// +// we do, however, need to make sure our size calculations don't +// overflow. hence a few helper functions for size calculations that +// multiply integers together, making sure that they're non-negative +// and no overflow occurs. + +// return 1 if the sum is valid, 0 on overflow. +// negative terms are considered invalid. +static int stbi__addsizes_valid(int a, int b) +{ + if (b < 0) return 0; + // now 0 <= b <= INT_MAX, hence also + // 0 <= INT_MAX - b <= INTMAX. + // And "a + b <= INT_MAX" (which might overflow) is the + // same as a <= INT_MAX - b (no overflow) + return a <= INT_MAX - b; +} + +// returns 1 if the product is valid, 0 on overflow. +// negative factors are considered invalid. +static int stbi__mul2sizes_valid(int a, int b) +{ + if (a < 0 || b < 0) return 0; + if (b == 0) return 1; // mul-by-0 is always safe + // portable way to check for no overflows in a*b + return a <= INT_MAX/b; +} + +#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) +// returns 1 if "a*b + add" has no negative terms/factors and doesn't overflow +static int stbi__mad2sizes_valid(int a, int b, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a*b, add); +} +#endif + +// returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow +static int stbi__mad3sizes_valid(int a, int b, int c, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && + stbi__addsizes_valid(a*b*c, add); +} + +// returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't overflow +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) || !defined(STBI_NO_PNM) +static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && + stbi__mul2sizes_valid(a*b*c, d) && stbi__addsizes_valid(a*b*c*d, add); +} +#endif + +#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) +// mallocs with size overflow checking +static void *stbi__malloc_mad2(int a, int b, int add) +{ + if (!stbi__mad2sizes_valid(a, b, add)) return NULL; + return stbi__malloc(a*b + add); +} +#endif + +static void *stbi__malloc_mad3(int a, int b, int c, int add) +{ + if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; + return stbi__malloc(a*b*c + add); +} + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) || !defined(STBI_NO_PNM) +static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) +{ + if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; + return stbi__malloc(a*b*c*d + add); +} +#endif + +// returns 1 if the sum of two signed ints is valid (between -2^31 and 2^31-1 inclusive), 0 on overflow. +static int stbi__addints_valid(int a, int b) +{ + if ((a >= 0) != (b >= 0)) return 1; // a and b have different signs, so no overflow + if (a < 0 && b < 0) return a >= INT_MIN - b; // same as a + b >= INT_MIN; INT_MIN - b cannot overflow since b < 0. + return a <= INT_MAX - b; +} + +// returns 1 if the product of two ints fits in a signed short, 0 on overflow. +static int stbi__mul2shorts_valid(int a, int b) +{ + if (b == 0 || b == -1) return 1; // multiplication by 0 is always 0; check for -1 so SHRT_MIN/b doesn't overflow + if ((a >= 0) == (b >= 0)) return a <= SHRT_MAX/b; // product is positive, so similar to mul2sizes_valid + if (b < 0) return a <= SHRT_MIN / b; // same as a * b >= SHRT_MIN + return a >= SHRT_MIN / b; +} + +// stbi__err - error +// stbi__errpf - error returning pointer to float +// stbi__errpuc - error returning pointer to unsigned char + +#ifdef STBI_NO_FAILURE_STRINGS + #define stbi__err(x,y) 0 +#elif defined(STBI_FAILURE_USERMSG) + #define stbi__err(x,y) stbi__err(y) +#else + #define stbi__err(x,y) stbi__err(x) +#endif + +#define stbi__errpf(x,y) ((float *)(size_t) (stbi__err(x,y)?NULL:NULL)) +#define stbi__errpuc(x,y) ((unsigned char *)(size_t) (stbi__err(x,y)?NULL:NULL)) + +STBIDEF void stbi_image_free(void *retval_from_stbi_load) +{ + STBI_FREE(retval_from_stbi_load); +} + +#ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); +#endif + +#ifndef STBI_NO_HDR +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); +#endif + +static int stbi__vertically_flip_on_load_global = 0; + +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) +{ + stbi__vertically_flip_on_load_global = flag_true_if_should_flip; +} + +#ifndef STBI_THREAD_LOCAL +#define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global +#else +static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, stbi__vertically_flip_on_load_set; + +STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip) +{ + stbi__vertically_flip_on_load_local = flag_true_if_should_flip; + stbi__vertically_flip_on_load_set = 1; +} + +#define stbi__vertically_flip_on_load (stbi__vertically_flip_on_load_set \ + ? stbi__vertically_flip_on_load_local \ + : stbi__vertically_flip_on_load_global) +#endif // STBI_THREAD_LOCAL + +static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) +{ + memset(ri, 0, sizeof(*ri)); // make sure it's initialized if we add new fields + ri->bits_per_channel = 8; // default is 8 so most paths don't have to be changed + ri->channel_order = STBI_ORDER_RGB; // all current input & output are this, but this is here so we can add BGR order + ri->num_channels = 0; + + // test the formats with a very explicit header first (at least a FOURCC + // or distinctive magic number first) + #ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_GIF + if (stbi__gif_test(s)) return stbi__gif_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_PSD + if (stbi__psd_test(s)) return stbi__psd_load(s,x,y,comp,req_comp, ri, bpc); + #else + STBI_NOTUSED(bpc); + #endif + #ifndef STBI_NO_PIC + if (stbi__pic_test(s)) return stbi__pic_load(s,x,y,comp,req_comp, ri); + #endif + + // then the formats that can end up attempting to load with just 1 or 2 + // bytes matching expectations; these are prone to false positives, so + // try them later + #ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_test(s)) return stbi__pnm_load(s,x,y,comp,req_comp, ri); + #endif + + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float *hdr = stbi__hdr_load(s, x,y,comp,req_comp, ri); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } + #endif + + #ifndef STBI_NO_TGA + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) + return stbi__tga_load(s,x,y,comp,req_comp, ri); + #endif + + return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); +} + +static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi_uc *reduced; + + reduced = (stbi_uc *) stbi__malloc(img_len); + if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + reduced[i] = (stbi_uc)((orig[i] >> 8) & 0xFF); // top half of each byte is sufficient approx of 16->8 bit scaling + + STBI_FREE(orig); + return reduced; +} + +static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi__uint16 *enlarged; + + enlarged = (stbi__uint16 *) stbi__malloc(img_len*2); + if (enlarged == NULL) return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + enlarged[i] = (stbi__uint16)((orig[i] << 8) + orig[i]); // replicate to high and low byte, maps 0->0, 255->0xffff + + STBI_FREE(orig); + return enlarged; +} + +static void stbi__vertical_flip(void *image, int w, int h, int bytes_per_pixel) +{ + int row; + size_t bytes_per_row = (size_t)w * bytes_per_pixel; + stbi_uc temp[2048]; + stbi_uc *bytes = (stbi_uc *)image; + + for (row = 0; row < (h>>1); row++) { + stbi_uc *row0 = bytes + row*bytes_per_row; + stbi_uc *row1 = bytes + (h - row - 1)*bytes_per_row; + // swap row0 with row1 + size_t bytes_left = bytes_per_row; + while (bytes_left) { + size_t bytes_copy = (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); + memcpy(temp, row0, bytes_copy); + memcpy(row0, row1, bytes_copy); + memcpy(row1, temp, bytes_copy); + row0 += bytes_copy; + row1 += bytes_copy; + bytes_left -= bytes_copy; + } + } +} + +#ifndef STBI_NO_GIF +static void stbi__vertical_flip_slices(void *image, int w, int h, int z, int bytes_per_pixel) +{ + int slice; + int slice_size = w * h * bytes_per_pixel; + + stbi_uc *bytes = (stbi_uc *)image; + for (slice = 0; slice < z; ++slice) { + stbi__vertical_flip(bytes, w, h, bytes_per_pixel); + bytes += slice_size; + } +} +#endif + +static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); + + if (result == NULL) + return NULL; + + // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + + if (ri.bits_per_channel != 8) { + result = stbi__convert_16_to_8((stbi__uint16 *) result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 8; + } + + // @TODO: move stbi__convert_format to here + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); + } + + return (unsigned char *) result; +} + +static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); + + if (result == NULL) + return NULL; + + // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + + if (ri.bits_per_channel != 16) { + result = stbi__convert_8_to_16((stbi_uc *) result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 16; + } + + // @TODO: move stbi__convert_format16 to here + // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to keep more precision + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); + } + + return (stbi__uint16 *) result; +} + +#if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) +static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, int req_comp) +{ + if (stbi__vertically_flip_on_load && result != NULL) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); + } +} +#endif + +#ifndef STBI_NO_STDIO + +#if defined(_WIN32) && defined(STBI_WINDOWS_UTF8) +STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar(unsigned int cp, unsigned long flags, const char *str, int cbmb, wchar_t *widestr, int cchwide); +STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, char *str, int cbmb, const char *defchar, int *used_default); +#endif + +#if defined(_WIN32) && defined(STBI_WINDOWS_UTF8) +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input) +{ + return WideCharToMultiByte(65001 /* UTF8 */, 0, input, -1, buffer, (int) bufferlen, NULL, NULL); +} +#endif + +static FILE *stbi__fopen(char const *filename, char const *mode) +{ + FILE *f; +#if defined(_WIN32) && defined(STBI_WINDOWS_UTF8) + wchar_t wMode[64]; + wchar_t wFilename[1024]; + if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, filename, -1, wFilename, sizeof(wFilename)/sizeof(*wFilename))) + return 0; + + if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, mode, -1, wMode, sizeof(wMode)/sizeof(*wMode))) + return 0; + +#if defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != _wfopen_s(&f, wFilename, wMode)) + f = 0; +#else + f = _wfopen(wFilename, wMode); +#endif + +#elif defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) + f=0; +#else + f = fopen(filename, mode); +#endif + return f; +} + + +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + unsigned char *result; + if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *result; + stbi__context s; + stbi__start_file(&s,f); + result = stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; +} + +STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi__uint16 *result; + stbi__context s; + stbi__start_file(&s,f); + result = stbi__load_and_postprocess_16bit(&s,x,y,comp,req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; +} + +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + stbi__uint16 *result; + if (!f) return (stbi_us *) stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file_16(f,x,y,comp,req_comp); + fclose(f); + return result; +} + + +#endif //!STBI_NO_STDIO + +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +} + +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +} + +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +} + +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp) +{ + unsigned char *result; + stbi__context s; + stbi__start_mem(&s,buffer,len); + + result = (unsigned char*) stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); + if (stbi__vertically_flip_on_load) { + stbi__vertical_flip_slices( result, *x, *y, *z, *comp ); + } + + return result; +} +#endif + +#ifndef STBI_NO_LINEAR +static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *data; + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + stbi__result_info ri; + float *hdr_data = stbi__hdr_load(s,x,y,comp,req_comp, &ri); + if (hdr_data) + stbi__float_postprocess(hdr_data,x,y,comp,req_comp); + return hdr_data; + } + #endif + data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); + if (data) + return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); + return stbi__errpf("unknown image type", "Image not of any known type, or corrupt"); +} + +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} + +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + float *result; + FILE *f = stbi__fopen(filename, "rb"); + if (!f) return stbi__errpf("can't fopen", "Unable to open file"); + result = stbi_loadf_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_file(&s,f); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} +#endif // !STBI_NO_STDIO + +#endif // !STBI_NO_LINEAR + +// these is-hdr-or-not is defined independent of whether STBI_NO_LINEAR is +// defined, for API simplicity; if STBI_NO_LINEAR is defined, it always +// reports false! + +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(buffer); + STBI_NOTUSED(len); + return 0; + #endif +} + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr (char const *filename) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result=0; + if (f) { + result = stbi_is_hdr_from_file(f); + fclose(f); + } + return result; +} + +STBIDEF int stbi_is_hdr_from_file(FILE *f) +{ + #ifndef STBI_NO_HDR + long pos = ftell(f); + int res; + stbi__context s; + stbi__start_file(&s,f); + res = stbi__hdr_test(&s); + fseek(f, pos, SEEK_SET); + return res; + #else + STBI_NOTUSED(f); + return 0; + #endif +} +#endif // !STBI_NO_STDIO + +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(clbk); + STBI_NOTUSED(user); + return 0; + #endif +} + +#ifndef STBI_NO_LINEAR +static float stbi__l2h_gamma=2.2f, stbi__l2h_scale=1.0f; + +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } +STBIDEF void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } +#endif + +static float stbi__h2l_gamma_i=1.0f/2.2f, stbi__h2l_scale_i=1.0f; + +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1/gamma; } +STBIDEF void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1/scale; } + + +////////////////////////////////////////////////////////////////////////////// +// +// Common code used by all image loaders +// + +enum +{ + STBI__SCAN_load=0, + STBI__SCAN_type, + STBI__SCAN_header +}; + +static void stbi__refill_buffer(stbi__context *s) +{ + int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen); + s->callback_already_read += (int) (s->img_buffer - s->img_buffer_original); + if (n == 0) { + // at end of file, treat same as if from memory, but need to handle case + // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file + s->read_from_callbacks = 0; + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start+1; + *s->img_buffer = 0; + } else { + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + n; + } +} + +stbi_inline static stbi_uc stbi__get8(stbi__context *s) +{ + if (s->img_buffer < s->img_buffer_end) + return *s->img_buffer++; + if (s->read_from_callbacks) { + stbi__refill_buffer(s); + return *s->img_buffer++; + } + return 0; +} + +#if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) +// nothing +#else +stbi_inline static int stbi__at_eof(stbi__context *s) +{ + if (s->io.read) { + if (!(s->io.eof)(s->io_user_data)) return 0; + // if feof() is true, check if buffer = end + // special case: we've only got the special 0 character at the end + if (s->read_from_callbacks == 0) return 1; + } + + return s->img_buffer >= s->img_buffer_end; +} +#endif + +#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) +// nothing +#else +static void stbi__skip(stbi__context *s, int n) +{ + if (n == 0) return; // already there! + if (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int) (s->img_buffer_end - s->img_buffer); + if (blen < n) { + s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); + return; + } + } + s->img_buffer += n; +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && defined(STBI_NO_PNM) +// nothing +#else +static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) +{ + if (s->io.read) { + int blen = (int) (s->img_buffer_end - s->img_buffer); + if (blen < n) { + int res, count; + + memcpy(buffer, s->img_buffer, blen); + + count = (s->io.read)(s->io_user_data, (char*) buffer + blen, n - blen); + res = (count == (n-blen)); + s->img_buffer = s->img_buffer_end; + return res; + } + } + + if (s->img_buffer+n <= s->img_buffer_end) { + memcpy(buffer, s->img_buffer, n); + s->img_buffer += n; + return 1; + } else + return 0; +} +#endif + +#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) +// nothing +#else +static int stbi__get16be(stbi__context *s) +{ + int z = stbi__get8(s); + return (z << 8) + stbi__get8(s); +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) +// nothing +#else +static stbi__uint32 stbi__get32be(stbi__context *s) +{ + stbi__uint32 z = stbi__get16be(s); + return (z << 16) + stbi__get16be(s); +} +#endif + +#if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) +// nothing +#else +static int stbi__get16le(stbi__context *s) +{ + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); +} +#endif + +#ifndef STBI_NO_BMP +static stbi__uint32 stbi__get32le(stbi__context *s) +{ + stbi__uint32 z = stbi__get16le(s); + z += (stbi__uint32)stbi__get16le(s) << 16; + return z; +} +#endif + +#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + +#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) +// nothing +#else +////////////////////////////////////////////////////////////////////////////// +// +// generic converter from built-in img_n to req_comp +// individual types do this automatically as much as possible (e.g. jpeg +// does all cases internally since it needs to colorspace convert anyway, +// and it never has alpha, so very few cases ). png can automatically +// interleave an alpha=255 channel, but falls back to this for other cases +// +// assume data buffer is malloced, so malloc a new one and free that one +// only failure mode is malloc failing + +static stbi_uc stbi__compute_y(int r, int g, int b) +{ + return (stbi_uc) (((r*77) + (g*150) + (29*b)) >> 8); +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) +// nothing +#else +static unsigned char *stbi__convert_format(unsigned char *data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i,j; + unsigned char *good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (unsigned char *) stbi__malloc_mad3(req_comp, x, y, 0); + if (good == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } + + for (j=0; j < (int) y; ++j) { + unsigned char *src = data + j * x * img_n ; + unsigned char *dest = good + j * x * req_comp; + + #define STBI__COMBO(a,b) ((a)*8+(b)) + #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp components; + // avoid switch per pixel, so use switch per scanline and massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=255; } break; + STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=255; } break; + STBI__CASE(2,1) { dest[0]=src[0]; } break; + STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; + STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=255; } break; + STBI__CASE(3,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; + STBI__CASE(3,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = 255; } break; + STBI__CASE(4,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; + STBI__CASE(4,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = src[3]; } break; + STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; + default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return stbi__errpuc("unsupported", "Unsupported format conversion"); + } + #undef STBI__CASE + } + + STBI_FREE(data); + return good; +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) +// nothing +#else +static stbi__uint16 stbi__compute_y_16(int r, int g, int b) +{ + return (stbi__uint16) (((r*77) + (g*150) + (29*b)) >> 8); +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) +// nothing +#else +static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i,j; + stbi__uint16 *good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (stbi__uint16 *) stbi__malloc(req_comp * x * y * 2); + if (good == NULL) { + STBI_FREE(data); + return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + } + + for (j=0; j < (int) y; ++j) { + stbi__uint16 *src = data + j * x * img_n ; + stbi__uint16 *dest = good + j * x * req_comp; + + #define STBI__COMBO(a,b) ((a)*8+(b)) + #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp components; + // avoid switch per pixel, so use switch per scanline and massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=0xffff; } break; + STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=0xffff; } break; + STBI__CASE(2,1) { dest[0]=src[0]; } break; + STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; + STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=0xffff; } break; + STBI__CASE(3,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; + STBI__CASE(3,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = 0xffff; } break; + STBI__CASE(4,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; + STBI__CASE(4,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = src[3]; } break; + STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; + default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return (stbi__uint16*) stbi__errpuc("unsupported", "Unsupported format conversion"); + } + #undef STBI__CASE + } + + STBI_FREE(data); + return good; +} +#endif + +#ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) +{ + int i,k,n; + float *output; + if (!data) return NULL; + output = (float *) stbi__malloc_mad4(x, y, comp, sizeof(float), 0); + if (output == NULL) { STBI_FREE(data); return stbi__errpf("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + output[i*comp + k] = (float) (pow(data[i*comp+k]/255.0f, stbi__l2h_gamma) * stbi__l2h_scale); + } + } + if (n < comp) { + for (i=0; i < x*y; ++i) { + output[i*comp + n] = data[i*comp + n]/255.0f; + } + } + STBI_FREE(data); + return output; +} +#endif + +#ifndef STBI_NO_HDR +#define stbi__float2int(x) ((int) (x)) +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) +{ + int i,k,n; + stbi_uc *output; + if (!data) return NULL; + output = (stbi_uc *) stbi__malloc_mad3(x, y, comp, 0); + if (output == NULL) { STBI_FREE(data); return stbi__errpuc("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + float z = (float) pow(data[i*comp+k]*stbi__h2l_scale_i, stbi__h2l_gamma_i) * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (stbi_uc) stbi__float2int(z); + } + if (k < comp) { + float z = data[i*comp+k] * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (stbi_uc) stbi__float2int(z); + } + } + STBI_FREE(data); + return output; +} +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// "baseline" JPEG/JFIF decoder +// +// simple implementation +// - doesn't support delayed output of y-dimension +// - simple interface (only one output format: 8-bit interleaved RGB) +// - doesn't try to recover corrupt jpegs +// - doesn't allow partial loading, loading multiple at once +// - still fast on x86 (copying globals into locals doesn't help x86) +// - allocates lots of intermediate memory (full size of all components) +// - non-interleaved case requires this anyway +// - allows good upsampling (see next) +// high-quality +// - upsampled channels are bilinearly interpolated, even across blocks +// - quality integer IDCT derived from IJG's 'slow' +// performance +// - fast huffman; reasonable integer IDCT +// - some SIMD kernels for common paths on targets with SSE2/NEON +// - uses a lot of intermediate memory, could cache poorly + +#ifndef STBI_NO_JPEG + +// huffman decoding acceleration +#define FAST_BITS 9 // larger handles more cases; smaller stomps less cache + +typedef struct +{ + stbi_uc fast[1 << FAST_BITS]; + // weirdly, repacking this into AoS is a 10% speed loss, instead of a win + stbi__uint16 code[256]; + stbi_uc values[256]; + stbi_uc size[257]; + unsigned int maxcode[18]; + int delta[17]; // old 'firstsymbol' - old 'firstcode' +} stbi__huffman; + +typedef struct +{ + stbi__context *s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi__uint16 dequant[4][64]; + stbi__int16 fast_ac[4][1 << FAST_BITS]; + +// sizes for components, interleaved MCUs + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; + +// definition of jpeg image component + struct + { + int id; + int h,v; + int tq; + int hd,ha; + int dc_pred; + + int x,y,w2,h2; + stbi_uc *data; + void *raw_data, *raw_coeff; + stbi_uc *linebuf; + short *coeff; // progressive only + int coeff_w, coeff_h; // number of 8x8 coefficient blocks + } img_comp[4]; + + stbi__uint32 code_buffer; // jpeg entropy-coded buffer + int code_bits; // number of valid bits + unsigned char marker; // marker seen while filling entropy buffer + int nomore; // flag if we saw a marker so must stop + + int progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + int jfif; + int app14_color_transform; // Adobe APP14 tag + int rgb; + + int scan_n, order[4]; + int restart_interval, todo; + +// kernels + void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step); + stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs); +} stbi__jpeg; + +static int stbi__build_huffman(stbi__huffman *h, int *count) +{ + int i,j,k=0; + unsigned int code; + // build size list for each symbol (from JPEG spec) + for (i=0; i < 16; ++i) { + for (j=0; j < count[i]; ++j) { + h->size[k++] = (stbi_uc) (i+1); + if(k >= 257) return stbi__err("bad size list","Corrupt JPEG"); + } + } + h->size[k] = 0; + + // compute actual symbols (from jpeg spec) + code = 0; + k = 0; + for(j=1; j <= 16; ++j) { + // compute delta to add to code to compute symbol id + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (stbi__uint16) (code++); + if (code-1 >= (1u << j)) return stbi__err("bad code lengths","Corrupt JPEG"); + } + // compute largest code + 1 for this size, preshifted as needed later + h->maxcode[j] = code << (16-j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; + + // build non-spec acceleration table; 255 is flag for not-accelerated + memset(h->fast, 255, 1 << FAST_BITS); + for (i=0; i < k; ++i) { + int s = h->size[i]; + if (s <= FAST_BITS) { + int c = h->code[i] << (FAST_BITS-s); + int m = 1 << (FAST_BITS-s); + for (j=0; j < m; ++j) { + h->fast[c+j] = (stbi_uc) i; + } + } + } + return 1; +} + +// build a table that decodes both magnitude and value of small ACs in +// one go. +static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) +{ + int i; + for (i=0; i < (1 << FAST_BITS); ++i) { + stbi_uc fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; + + if (magbits && len + magbits <= FAST_BITS) { + // magnitude code followed by receive_extend code + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (~0U << magbits) + 1; + // if the result is small enough, we can fit it in fast_ac table + if (k >= -128 && k <= 127) + fast_ac[i] = (stbi__int16) ((k * 256) + (run * 16) + (len + magbits)); + } + } + } +} + +static void stbi__grow_buffer_unsafe(stbi__jpeg *j) +{ + do { + unsigned int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__get8(j->s); + while (c == 0xff) c = stbi__get8(j->s); // consume fill bytes + if (c != 0) { + j->marker = (unsigned char) c; + j->nomore = 1; + return; + } + } + j->code_buffer |= b << (24 - j->code_bits); + j->code_bits += 8; + } while (j->code_bits <= 24); +} + +// (1 << n) - 1 +static const stbi__uint32 stbi__bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; + +// decode a jpeg huffman value from the bitstream +stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) +{ + unsigned int temp; + int c,k; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + // look at the top FAST_BITS and determine what symbol ID it is, + // if the code is <= FAST_BITS + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + k = h->fast[c]; + if (k < 255) { + int s = h->size[k]; + if (s > j->code_bits) + return -1; + j->code_buffer <<= s; + j->code_bits -= s; + return h->values[k]; + } + + // naive test is to shift the code_buffer down so k bits are + // valid, then test against maxcode. To speed this up, we've + // preshifted maxcode left so that it has (16-k) 0s at the + // end; in other words, regardless of the number of bits, it + // wants to be compared against something shifted to have 16; + // that way we don't need to shift inside the loop. + temp = j->code_buffer >> 16; + for (k=FAST_BITS+1 ; ; ++k) + if (temp < h->maxcode[k]) + break; + if (k == 17) { + // error! code not found + j->code_bits -= 16; + return -1; + } + + if (k > j->code_bits) + return -1; + + // convert the huffman code to the symbol id + c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; + if(c < 0 || c >= 256) // symbol id out of bounds! + return -1; + STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & stbi__bmask[h->size[c]]) == h->code[c]); + + // convert the id to a symbol + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; +} + +// bias[n] = (-1<code_bits < n) stbi__grow_buffer_unsafe(j); + if (j->code_bits < n) return 0; // ran out of bits from stream, return 0s intead of continuing + + sgn = j->code_buffer >> 31; // sign bit always in MSB; 0 if MSB clear (positive), 1 if MSB set (negative) + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & (sgn - 1)); +} + +// get some unsigned bits +stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) +{ + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + if (j->code_bits < n) return 0; // ran out of bits from stream, return 0s intead of continuing + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; +} + +stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) +{ + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + if (j->code_bits < 1) return 0; // ran out of bits from stream, return 0s intead of continuing + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; +} + +// given a value that's at position X in the zigzag stream, +// where does it appear in the 8x8 matrix coded as row-major? +static const stbi_uc stbi__jpeg_dezigzag[64+15] = +{ + 0, 1, 8, 16, 9, 2, 3, 10, + 17, 24, 32, 25, 18, 11, 4, 5, + 12, 19, 26, 33, 40, 48, 41, 34, + 27, 20, 13, 6, 7, 14, 21, 28, + 35, 42, 49, 56, 57, 50, 43, 36, + 29, 22, 15, 23, 30, 37, 44, 51, + 58, 59, 52, 45, 38, 31, 39, 46, + 53, 60, 61, 54, 47, 55, 62, 63, + // let corrupt input sample past end + 63, 63, 63, 63, 63, 63, 63, 63, + 63, 63, 63, 63, 63, 63, 63 +}; + +// decode one 64-entry block-- +static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], stbi__huffman *hdc, stbi__huffman *hac, stbi__int16 *fac, int b, stbi__uint16 *dequant) +{ + int diff,dc,k; + int t; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0 || t > 15) return stbi__err("bad huffman code","Corrupt JPEG"); + + // 0 all the ac values now so we can do it 32-bits at a time + memset(data,0,64*sizeof(data[0])); + + diff = t ? stbi__extend_receive(j, t) : 0; + if (!stbi__addints_valid(j->img_comp[b].dc_pred, diff)) return stbi__err("bad delta","Corrupt JPEG"); + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + if (!stbi__mul2shorts_valid(dc, dequant[0])) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + data[0] = (short) (dc * dequant[0]); + + // decode AC components, see JPEG spec + k = 1; + do { + unsigned int zig; + int c,r,s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + if (s > j->code_bits) return stbi__err("bad huffman code", "Combined length longer than code bits available"); + j->code_buffer <<= s; + j->code_bits -= s; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) ((r >> 8) * dequant[zig]); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; + } else { + k += r; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) (stbi__extend_receive(j,s) * dequant[zig]); + } + } + } while (k < 64); + return 1; +} + +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], stbi__huffman *hdc, int b) +{ + int diff,dc; + int t; + if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + if (j->succ_high == 0) { + // first scan for DC coefficient, must be first + memset(data,0,64*sizeof(data[0])); // 0 all the ac values now + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0 || t > 15) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + diff = t ? stbi__extend_receive(j, t) : 0; + + if (!stbi__addints_valid(j->img_comp[b].dc_pred, diff)) return stbi__err("bad delta", "Corrupt JPEG"); + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + if (!stbi__mul2shorts_valid(dc, 1 << j->succ_low)) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + data[0] = (short) (dc * (1 << j->succ_low)); + } else { + // refinement scan for DC coefficient + if (stbi__jpeg_get_bit(j)) + data[0] += (short) (1 << j->succ_low); + } + return 1; +} + +// @OPTIMIZE: store non-zigzagged during the decode passes, +// and only de-zigzag when dequantizing +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], stbi__huffman *hac, stbi__int16 *fac) +{ + int k; + if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->succ_high == 0) { + int shift = j->succ_low; + + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c,r,s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + if (s > j->code_bits) return stbi__err("bad huffman code", "Combined length longer than code bits available"); + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) ((r >> 8) * (1 << shift)); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) (stbi__extend_receive(j,s) * (1 << shift)); + } + } + } while (k <= j->spec_end); + } else { + // refinement scan for these AC coefficients + + short bit = (short) (1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short *p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit)==0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } + } else { + k = j->spec_start; + do { + int r,s; + int rs = stbi__jpeg_huff_decode(j, hac); // @OPTIMIZE see if we can use the fast path here, advance-by-r is so slow, eh + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; // force end of block + } else { + // r=15 s=0 should write 16 0s, so we just do + // a run of 15 0s and then write s (which is 0), + // so we don't have to do anything special here + } + } else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + // sign bit + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + // advance by r + while (k <= j->spec_end) { + short *p = &data[stbi__jpeg_dezigzag[k++]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit)==0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } else { + if (r == 0) { + *p = (short) s; + break; + } + --r; + } + } + } while (k <= j->spec_end); + } + } + return 1; +} + +// take a -128..127 value and stbi__clamp it and convert to 0..255 +stbi_inline static stbi_uc stbi__clamp(int x) +{ + // trick to use a single test to catch both cases + if ((unsigned int) x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (stbi_uc) x; +} + +#define stbi__f2f(x) ((int) (((x) * 4096 + 0.5))) +#define stbi__fsh(x) ((x) * 4096) + +// derived from jidctint -- DCT_ISLOW +#define STBI__IDCT_1D(s0,s1,s2,s3,s4,s5,s6,s7) \ + int t0,t1,t2,t3,p1,p2,p3,p4,p5,x0,x1,x2,x3; \ + p2 = s2; \ + p3 = s6; \ + p1 = (p2+p3) * stbi__f2f(0.5411961f); \ + t2 = p1 + p3*stbi__f2f(-1.847759065f); \ + t3 = p1 + p2*stbi__f2f( 0.765366865f); \ + p2 = s0; \ + p3 = s4; \ + t0 = stbi__fsh(p2+p3); \ + t1 = stbi__fsh(p2-p3); \ + x0 = t0+t3; \ + x3 = t0-t3; \ + x1 = t1+t2; \ + x2 = t1-t2; \ + t0 = s7; \ + t1 = s5; \ + t2 = s3; \ + t3 = s1; \ + p3 = t0+t2; \ + p4 = t1+t3; \ + p1 = t0+t3; \ + p2 = t1+t2; \ + p5 = (p3+p4)*stbi__f2f( 1.175875602f); \ + t0 = t0*stbi__f2f( 0.298631336f); \ + t1 = t1*stbi__f2f( 2.053119869f); \ + t2 = t2*stbi__f2f( 3.072711026f); \ + t3 = t3*stbi__f2f( 1.501321110f); \ + p1 = p5 + p1*stbi__f2f(-0.899976223f); \ + p2 = p5 + p2*stbi__f2f(-2.562915447f); \ + p3 = p3*stbi__f2f(-1.961570560f); \ + p4 = p4*stbi__f2f(-0.390180644f); \ + t3 += p1+p4; \ + t2 += p2+p3; \ + t1 += p2+p4; \ + t0 += p1+p3; + +static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) +{ + int i,val[64],*v=val; + stbi_uc *o; + short *d = data; + + // columns + for (i=0; i < 8; ++i,++d, ++v) { + // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing + if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 + && d[40]==0 && d[48]==0 && d[56]==0) { + // no shortcut 0 seconds + // (1|2|3|4|5|6|7)==0 0 seconds + // all separate -0.047 seconds + // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds + int dcterm = d[0]*4; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + STBI__IDCT_1D(d[ 0],d[ 8],d[16],d[24],d[32],d[40],d[48],d[56]) + // constants scaled things up by 1<<12; let's bring them back + // down, but keep 2 extra bits of precision + x0 += 512; x1 += 512; x2 += 512; x3 += 512; + v[ 0] = (x0+t3) >> 10; + v[56] = (x0-t3) >> 10; + v[ 8] = (x1+t2) >> 10; + v[48] = (x1-t2) >> 10; + v[16] = (x2+t1) >> 10; + v[40] = (x2-t1) >> 10; + v[24] = (x3+t0) >> 10; + v[32] = (x3-t0) >> 10; + } + } + + for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) { + // no fast case since the first 1D IDCT spread components out + STBI__IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) + // constants scaled things up by 1<<12, plus we had 1<<2 from first + // loop, plus horizontal and vertical each scale by sqrt(8) so together + // we've got an extra 1<<3, so 1<<17 total we need to remove. + // so we want to round that, which means adding 0.5 * 1<<17, + // aka 65536. Also, we'll end up with -128 to 127 that we want + // to encode as 0..255 by adding 128, so we'll add that before the shift + x0 += 65536 + (128<<17); + x1 += 65536 + (128<<17); + x2 += 65536 + (128<<17); + x3 += 65536 + (128<<17); + // tried computing the shifts into temps, or'ing the temps to see + // if any were out of range, but that was slower + o[0] = stbi__clamp((x0+t3) >> 17); + o[7] = stbi__clamp((x0-t3) >> 17); + o[1] = stbi__clamp((x1+t2) >> 17); + o[6] = stbi__clamp((x1-t2) >> 17); + o[2] = stbi__clamp((x2+t1) >> 17); + o[5] = stbi__clamp((x2-t1) >> 17); + o[3] = stbi__clamp((x3+t0) >> 17); + o[4] = stbi__clamp((x3-t0) >> 17); + } +} + +#ifdef STBI_SSE2 +// sse2 integer IDCT. not the fastest possible implementation but it +// produces bit-identical results to the generic C version so it's +// fully "transparent". +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) +{ + // This is constructed to match our regular (generic) integer IDCT exactly. + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; + + // dot product constant: even elems=x, odd elems=y + #define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + + // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) + // out(1) = c1[even]*x + c1[odd]*y + #define dct_rot(out0,out1, x,y,c0,c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + + // out = in << 12 (in 16-bit, out 32-bit) + #define dct_widen(out, in) \ + __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + + // wide add + #define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + + // wide sub + #define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + + // butterfly a/b, add bias, then shift by "s" and pack + #define dct_bfly32o(out0, out1, a,b,bias,s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ + } + + // 8-bit interleave step (for transposes) + #define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) + + // 16-bit interleave step (for transposes) + #define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) + + #define dct_pass(bias,shift) \ + { \ + /* even part */ \ + dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ + dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0,row7, x0,x7,bias,shift); \ + dct_bfly32o(row1,row6, x1,x6,bias,shift); \ + dct_bfly32o(row2,row5, x2,x5,bias,shift); \ + dct_bfly32o(row3,row4, x3,x4,bias,shift); \ + } + + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f( 0.765366865f), stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); + __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f( 0.298631336f), stbi__f2f(-1.961570560f)); + __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f( 3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f( 2.053119869f), stbi__f2f(-0.390180644f)); + __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f( 1.501321110f)); + + // rounding biases in column/row passes, see stbi__idct_block for explanation. + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128<<17)); + + // load + row0 = _mm_load_si128((const __m128i *) (data + 0*8)); + row1 = _mm_load_si128((const __m128i *) (data + 1*8)); + row2 = _mm_load_si128((const __m128i *) (data + 2*8)); + row3 = _mm_load_si128((const __m128i *) (data + 3*8)); + row4 = _mm_load_si128((const __m128i *) (data + 4*8)); + row5 = _mm_load_si128((const __m128i *) (data + 5*8)); + row6 = _mm_load_si128((const __m128i *) (data + 6*8)); + row7 = _mm_load_si128((const __m128i *) (data + 7*8)); + + // column pass + dct_pass(bias_0, 10); + + { + // 16bit 8x8 transpose pass 1 + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); + + // transpose pass 2 + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); + + // transpose pass 3 + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } + + // row pass + dct_pass(bias_1, 17); + + { + // pack + __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); + + // 8bit 8x8 transpose pass 1 + dct_interleave8(p0, p2); // a0e0a1e1... + dct_interleave8(p1, p3); // c0g0c1g1... + + // transpose pass 2 + dct_interleave8(p0, p1); // a0c0e0g0... + dct_interleave8(p2, p3); // b0d0f0h0... + + // transpose pass 3 + dct_interleave8(p0, p2); // a0b0c0d0... + dct_interleave8(p1, p3); // a4b4c4d4... + + // store + _mm_storel_epi64((__m128i *) out, p0); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p2); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p1); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p3); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p3, 0x4e)); + } + +#undef dct_const +#undef dct_rot +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_interleave8 +#undef dct_interleave16 +#undef dct_pass +} + +#endif // STBI_SSE2 + +#ifdef STBI_NEON + +// NEON integer IDCT. should produce bit-identical +// results to the generic C version. +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) +{ + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; + + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f( 0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f( 1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f( 0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f( 2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f( 3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f( 1.501321110f)); + +#define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + +#define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + +#define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + +// wide add +#define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + +// wide sub +#define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + +// butterfly a/b, then shift using "shiftop" by "s" and pack +#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } + +#define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ + dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ + dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ + dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ + } + + // load + row0 = vld1q_s16(data + 0*8); + row1 = vld1q_s16(data + 1*8); + row2 = vld1q_s16(data + 2*8); + row3 = vld1q_s16(data + 3*8); + row4 = vld1q_s16(data + 4*8); + row5 = vld1q_s16(data + 5*8); + row6 = vld1q_s16(data + 6*8); + row7 = vld1q_s16(data + 7*8); + + // add DC bias + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + + // column pass + dct_pass(vrshrn_n_s32, 10); + + // 16bit 8x8 transpose + { +// these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. +// whether compilers actually get this is another story, sadly. +#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } +#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + + // pass 1 + dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); + + // pass 2 + dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); + + // pass 3 + dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); + +#undef dct_trn16 +#undef dct_trn32 +#undef dct_trn64 + } + + // row pass + // vrshrn_n_s32 only supports shifts up to 16, we need + // 17. so do a non-rounding shift of 16 first then follow + // up with a rounding shift by 1. + dct_pass(vshrn_n_s32, 16); + + { + // pack and round + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + + // again, these can translate into one instruction, but often don't. +#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } +#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + + // sadly can't use interleaved stores here since we only write + // 8 bytes to each scan line! + + // 8x8 8-bit transpose pass 1 + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); + + // pass 2 + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); + + // pass 3 + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); + + // store + vst1_u8(out, p0); out += out_stride; + vst1_u8(out, p1); out += out_stride; + vst1_u8(out, p2); out += out_stride; + vst1_u8(out, p3); out += out_stride; + vst1_u8(out, p4); out += out_stride; + vst1_u8(out, p5); out += out_stride; + vst1_u8(out, p6); out += out_stride; + vst1_u8(out, p7); + +#undef dct_trn8_8 +#undef dct_trn8_16 +#undef dct_trn8_32 + } + +#undef dct_long_mul +#undef dct_long_mac +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_pass +} + +#endif // STBI_NEON + +#define STBI__MARKER_none 0xff +// if there's a pending marker from the entropy stream, return that +// otherwise, fetch from the stream and get a marker. if there's no +// marker, return 0xff, which is never a valid marker value +static stbi_uc stbi__get_marker(stbi__jpeg *j) +{ + stbi_uc x; + if (j->marker != STBI__MARKER_none) { x = j->marker; j->marker = STBI__MARKER_none; return x; } + x = stbi__get8(j->s); + if (x != 0xff) return STBI__MARKER_none; + while (x == 0xff) + x = stbi__get8(j->s); // consume repeated 0xff fill bytes + return x; +} + +// in each scan, we'll have scan_n components, and the order +// of the components is specified by order[] +#define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + +// after a restart interval, stbi__jpeg_reset the entropy decoder and +// the dc prediction +static void stbi__jpeg_reset(stbi__jpeg *j) +{ + j->code_bits = 0; + j->code_buffer = 0; + j->nomore = 0; + j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = j->img_comp[3].dc_pred = 0; + j->marker = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; + // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, + // since we don't even allow 1<<30 pixels +} + +static int stbi__parse_entropy_coded_data(stbi__jpeg *z) +{ + stbi__jpeg_reset(z); + if (!z->progressive) { + if (z->scan_n == 1) { + int i,j; + STBI_SIMD_ALIGN(short, data[64]); + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i,j,k,x,y; + STBI_SIMD_ALIGN(short, data[64]); + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y=0; y < z->img_comp[n].v; ++y) { + for (x=0; x < z->img_comp[n].h; ++x) { + int x2 = (i*z->img_comp[n].h + x)*8; + int y2 = (j*z->img_comp[n].v + y)*8; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data); + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } else { + if (z->scan_n == 1) { + int i,j; + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) + return 0; + } + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i,j,k,x,y; + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y=0; y < z->img_comp[n].v; ++y) { + for (x=0; x < z->img_comp[n].h; ++x) { + int x2 = (i*z->img_comp[n].h + x); + int y2 = (j*z->img_comp[n].v + y); + short *data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} + +static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) +{ + int i; + for (i=0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg *z) +{ + if (z->progressive) { + // dequantize and idct the data + int i,j,n; + for (n=0; n < z->s->img_n; ++n) { + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg *z, int m) +{ + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker","Corrupt JPEG"); + + case 0xDD: // DRI - specify restart interval + if (stbi__get16be(z->s) != 4) return stbi__err("bad DRI len","Corrupt JPEG"); + z->restart_interval = stbi__get16be(z->s); + return 1; + + case 0xDB: // DQT - define quantization table + L = stbi__get16be(z->s)-2; + while (L > 0) { + int q = stbi__get8(z->s); + int p = q >> 4, sixteen = (p != 0); + int t = q & 15,i; + if (p != 0 && p != 1) return stbi__err("bad DQT type","Corrupt JPEG"); + if (t > 3) return stbi__err("bad DQT table","Corrupt JPEG"); + + for (i=0; i < 64; ++i) + z->dequant[t][stbi__jpeg_dezigzag[i]] = (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); + L -= (sixteen ? 129 : 65); + } + return L==0; + + case 0xC4: // DHT - define huffman table + L = stbi__get16be(z->s)-2; + while (L > 0) { + stbi_uc *v; + int sizes[16],i,n=0; + int q = stbi__get8(z->s); + int tc = q >> 4; + int th = q & 15; + if (tc > 1 || th > 3) return stbi__err("bad DHT header","Corrupt JPEG"); + for (i=0; i < 16; ++i) { + sizes[i] = stbi__get8(z->s); + n += sizes[i]; + } + if(n > 256) return stbi__err("bad DHT header","Corrupt JPEG"); // Loop over i < n would write past end of values! + L -= 17; + if (tc == 0) { + if (!stbi__build_huffman(z->huff_dc+th, sizes)) return 0; + v = z->huff_dc[th].values; + } else { + if (!stbi__build_huffman(z->huff_ac+th, sizes)) return 0; + v = z->huff_ac[th].values; + } + for (i=0; i < n; ++i) + v[i] = stbi__get8(z->s); + if (tc != 0) + stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L==0; + } + + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + L = stbi__get16be(z->s); + if (L < 2) { + if (m == 0xFE) + return stbi__err("bad COM len","Corrupt JPEG"); + else + return stbi__err("bad APP len","Corrupt JPEG"); + } + L -= 2; + + if (m == 0xE0 && L >= 5) { // JFIF APP0 segment + static const unsigned char tag[5] = {'J','F','I','F','\0'}; + int ok = 1; + int i; + for (i=0; i < 5; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 5; + if (ok) + z->jfif = 1; + } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment + static const unsigned char tag[6] = {'A','d','o','b','e','\0'}; + int ok = 1; + int i; + for (i=0; i < 6; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 6; + if (ok) { + stbi__get8(z->s); // version + stbi__get16be(z->s); // flags0 + stbi__get16be(z->s); // flags1 + z->app14_color_transform = stbi__get8(z->s); // color transform + L -= 6; + } + } + + stbi__skip(z->s, L); + return 1; + } + + return stbi__err("unknown marker","Corrupt JPEG"); +} + +// after we see SOS +static int stbi__process_scan_header(stbi__jpeg *z) +{ + int i; + int Ls = stbi__get16be(z->s); + z->scan_n = stbi__get8(z->s); + if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return stbi__err("bad SOS component count","Corrupt JPEG"); + if (Ls != 6+2*z->scan_n) return stbi__err("bad SOS len","Corrupt JPEG"); + for (i=0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) + break; + if (which == z->s->img_n) return 0; // no match + z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return stbi__err("bad DC huff","Corrupt JPEG"); + z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff","Corrupt JPEG"); + z->order[i] = which; + } + + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } else { + if (z->spec_start != 0) return stbi__err("bad SOS","Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS","Corrupt JPEG"); + z->spec_end = 63; + } + } + + return 1; +} + +static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) +{ + int i; + for (i=0; i < ncomp; ++i) { + if (z->img_comp[i].raw_data) { + STBI_FREE(z->img_comp[i].raw_data); + z->img_comp[i].raw_data = NULL; + z->img_comp[i].data = NULL; + } + if (z->img_comp[i].raw_coeff) { + STBI_FREE(z->img_comp[i].raw_coeff); + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].linebuf) { + STBI_FREE(z->img_comp[i].linebuf); + z->img_comp[i].linebuf = NULL; + } + } + return why; +} + +static int stbi__process_frame_header(stbi__jpeg *z, int scan) +{ + stbi__context *s = z->s; + int Lf,p,i,q, h_max=1,v_max=1,c; + Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len","Corrupt JPEG"); // JPEG + p = stbi__get8(s); if (p != 8) return stbi__err("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline + s->img_y = stbi__get16be(s); if (s->img_y == 0) return stbi__err("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG + s->img_x = stbi__get16be(s); if (s->img_x == 0) return stbi__err("0 width","Corrupt JPEG"); // JPEG requires + if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + c = stbi__get8(s); + if (c != 3 && c != 1 && c != 4) return stbi__err("bad component count","Corrupt JPEG"); + s->img_n = c; + for (i=0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } + + if (Lf != 8+3*s->img_n) return stbi__err("bad SOF len","Corrupt JPEG"); + + z->rgb = 0; + for (i=0; i < s->img_n; ++i) { + static const unsigned char rgb[3] = { 'R', 'G', 'B' }; + z->img_comp[i].id = stbi__get8(s); + if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) + ++z->rgb; + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H","Corrupt JPEG"); + z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V","Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ","Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) return stbi__err("too large", "Image too large to decode"); + + for (i=0; i < s->img_n; ++i) { + if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; + if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; + } + + // check that plane subsampling factors are integer ratios; our resamplers can't deal with fractional ratios + // and I've never seen a non-corrupted JPEG file actually use them + for (i=0; i < s->img_n; ++i) { + if (h_max % z->img_comp[i].h != 0) return stbi__err("bad H","Corrupt JPEG"); + if (v_max % z->img_comp[i].v != 0) return stbi__err("bad V","Corrupt JPEG"); + } + + // compute interleaved mcu info + z->img_h_max = h_max; + z->img_v_max = v_max; + z->img_mcu_w = h_max * 8; + z->img_mcu_h = v_max * 8; + // these sizes can't be more than 17 bits + z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w; + z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h; + + for (i=0; i < s->img_n; ++i) { + // number of effective pixels (e.g. for non-interleaved MCU) + z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max; + z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max; + // to simplify generation, we'll allocate enough memory to decode + // the bogus oversized data from using interleaved MCUs and their + // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't + // discard the extra data until colorspace conversion + // + // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked earlier) + // so these muls can't overflow with 32-bit ints (which we require) + z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; + z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; + z->img_comp[i].coeff = 0; + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].linebuf = NULL; + z->img_comp[i].raw_data = stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); + if (z->img_comp[i].raw_data == NULL) + return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); + // align blocks for idct using mmx/sse + z->img_comp[i].data = (stbi_uc*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); + if (z->progressive) { + // w2, h2 are multiples of 8 (see above) + z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; + z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; + z->img_comp[i].raw_coeff = stbi__malloc_mad3(z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); + if (z->img_comp[i].raw_coeff == NULL) + return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); + z->img_comp[i].coeff = (short*) (((size_t) z->img_comp[i].raw_coeff + 15) & ~15); + } + } + + return 1; +} + +// use comparisons since in some cases we handle more than one case (e.g. SOF) +#define stbi__DNL(x) ((x) == 0xdc) +#define stbi__SOI(x) ((x) == 0xd8) +#define stbi__EOI(x) ((x) == 0xd9) +#define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) +#define stbi__SOS(x) ((x) == 0xda) + +#define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) +{ + int m; + z->jfif = 0; + z->app14_color_transform = -1; // valid values are 0,1,2 + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI","Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) { + if (!stbi__process_marker(z,m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); + m = stbi__get_marker(z); + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; +} + +static stbi_uc stbi__skip_jpeg_junk_at_end(stbi__jpeg *j) +{ + // some JPEGs have junk at end, skip over it but if we find what looks + // like a valid marker, resume there + while (!stbi__at_eof(j->s)) { + stbi_uc x = stbi__get8(j->s); + while (x == 0xff) { // might be a marker + if (stbi__at_eof(j->s)) return STBI__MARKER_none; + x = stbi__get8(j->s); + if (x != 0x00 && x != 0xff) { + // not a stuffed zero or lead-in to another marker, looks + // like an actual marker, return it + return x; + } + // stuffed zero has x=0 now which ends the loop, meaning we go + // back to regular scan loop. + // repeated 0xff keeps trying to read the next byte of the marker. + } + } + return STBI__MARKER_none; +} + +// decode image to YCbCr format +static int stbi__decode_jpeg_image(stbi__jpeg *j) +{ + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none ) { + j->marker = stbi__skip_jpeg_junk_at_end(j); + // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 + } + m = stbi__get_marker(j); + if (STBI__RESTART(m)) + m = stbi__get_marker(j); + } else if (stbi__DNL(m)) { + int Ld = stbi__get16be(j->s); + stbi__uint32 NL = stbi__get16be(j->s); + if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); + if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); + m = stbi__get_marker(j); + } else { + if (!stbi__process_marker(j, m)) return 1; + m = stbi__get_marker(j); + } + } + if (j->progressive) + stbi__jpeg_finish(j); + return 1; +} + +// static jfif-centered resampling (across block boundaries) + +typedef stbi_uc *(*resample_row_func)(stbi_uc *out, stbi_uc *in0, stbi_uc *in1, + int w, int hs); + +#define stbi__div4(x) ((stbi_uc) ((x) >> 2)) + +static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; +} + +static stbi_uc* stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate two samples vertically for every one in input + int i; + STBI_NOTUSED(hs); + for (i=0; i < w; ++i) + out[i] = stbi__div4(3*in_near[i] + in_far[i] + 2); + return out; +} + +static stbi_uc* stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate two samples horizontally for every one in input + int i; + stbi_uc *input = in_near; + + if (w == 1) { + // if only one sample, can't do any interpolation + out[0] = out[1] = input[0]; + return out; + } + + out[0] = input[0]; + out[1] = stbi__div4(input[0]*3 + input[1] + 2); + for (i=1; i < w-1; ++i) { + int n = 3*input[i]+2; + out[i*2+0] = stbi__div4(n+input[i-1]); + out[i*2+1] = stbi__div4(n+input[i+1]); + } + out[i*2+0] = stbi__div4(input[w-2]*3 + input[w-1] + 2); + out[i*2+1] = input[w-1]; + + STBI_NOTUSED(in_far); + STBI_NOTUSED(hs); + + return out; +} + +#define stbi__div16(x) ((stbi_uc) ((x) >> 4)) + +static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate 2x2 samples for every one in input + int i,t0,t1; + if (w == 1) { + out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + out[0] = stbi__div4(t1+2); + for (i=1; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = stbi__div16(3*t0 + t1 + 8); + out[i*2 ] = stbi__div16(3*t1 + t0 + 8); + } + out[w*2-1] = stbi__div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate 2x2 samples for every one in input + int i=0,t0,t1; + + if (w == 1) { + out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + // process groups of 8 pixels for as long as we can. + // note we can't handle the last pixel in a row in this loop + // because we need to handle the filter boundary conditions. + for (; i < ((w-1) & ~7); i += 8) { +#if defined(STBI_SSE2) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i *) (in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i *) (in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = _mm_insert_epi16(nxt0, 3*in_near[i+8] + in_far[i+8], 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); + + // interleave even and odd pixels, then undo scaling. + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); + + // pack and write output + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i *) (out + i*2), outv); +#elif defined(STBI_NEON) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = vsetq_lane_s16(3*in_near[i+8] + in_far[i+8], nxt0, 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); + + // undo scaling and round, then store with even/odd phases interleaved + uint8x8x2_t o; + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i*2, o); +#endif + + // "previous" value for next iter + t1 = 3*in_near[i+7] + in_far[i+7]; + } + + t0 = t1; + t1 = 3*in_near[i] + in_far[i]; + out[i*2] = stbi__div16(3*t1 + t0 + 8); + + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = stbi__div16(3*t0 + t1 + 8); + out[i*2 ] = stbi__div16(3*t1 + t0 + 8); + } + out[w*2-1] = stbi__div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} +#endif + +static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // resample with nearest-neighbor + int i,j; + STBI_NOTUSED(in_far); + for (i=0; i < w; ++i) + for (j=0; j < hs; ++j) + out[i*hs+j] = in_near[i]; + return out; +} + +// this is a reduced-precision calculation of YCbCr-to-RGB introduced +// to make sure the code produces the same results in both SIMD and scalar +#define stbi__float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) +static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step) +{ + int i; + for (i=0; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1<<19); // rounding + int r,g,b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr* stbi__float2fixed(1.40200f); + g = y_fixed + (cr*-stbi__float2fixed(0.71414f)) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb* stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, stbi_uc const *pcb, stbi_uc const *pcr, int count, int step) +{ + int i = 0; + +#ifdef STBI_SSE2 + // step == 3 is pretty ugly on the final interleave, and i'm not convinced + // it's useful in practice (you wouldn't use it for textures, for example). + // so just accelerate step == 4 case. + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16( (short) ( 1.40200f*4096.0f+0.5f)); + __m128i cr_const1 = _mm_set1_epi16( - (short) ( 0.71414f*4096.0f+0.5f)); + __m128i cb_const0 = _mm_set1_epi16( - (short) ( 0.34414f*4096.0f+0.5f)); + __m128i cb_const1 = _mm_set1_epi16( (short) ( 1.77200f*4096.0f+0.5f)); + __m128i y_bias = _mm_set1_epi8((char) (unsigned char) 128); + __m128i xw = _mm_set1_epi16(255); // alpha channel + + for (; i+7 < count; i += 8) { + // load + __m128i y_bytes = _mm_loadl_epi64((__m128i *) (y+i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i *) (pcr+i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i *) (pcb+i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 + + // unpack to short (and left-shift cr, cb by 8) + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + + // color transform + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); + + // descale + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); + + // back to byte, set up for transpose + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); + + // transpose to interleave channels + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); + + // store + _mm_storeu_si128((__m128i *) (out + 0), o0); + _mm_storeu_si128((__m128i *) (out + 16), o1); + out += 32; + } + } +#endif + +#ifdef STBI_NEON + // in this version, step=3 support would be easy to add. but is there demand? + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16( (short) ( 1.40200f*4096.0f+0.5f)); + int16x8_t cr_const1 = vdupq_n_s16( - (short) ( 0.71414f*4096.0f+0.5f)); + int16x8_t cb_const0 = vdupq_n_s16( - (short) ( 0.34414f*4096.0f+0.5f)); + int16x8_t cb_const1 = vdupq_n_s16( (short) ( 1.77200f*4096.0f+0.5f)); + + for (; i+7 < count; i += 8) { + // load + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + + // expand to s16 + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); + + // color transform + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); + + // undo scaling, round, convert to byte + uint8x8x4_t o; + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); + + // store, interleaving r/g/b/a + vst4_u8(out, o); + out += 8*4; + } + } +#endif + + for (; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1<<19); // rounding + int r,g,b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr* stbi__float2fixed(1.40200f); + g = y_fixed + cr*-stbi__float2fixed(0.71414f) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb* stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} +#endif + +// set up the kernels +static void stbi__setup_jpeg(stbi__jpeg *j) +{ + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; + +#ifdef STBI_SSE2 + if (stbi__sse2_available()) { + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } +#endif + +#ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; +#endif +} + +// clean up the temporary component buffers +static void stbi__cleanup_jpeg(stbi__jpeg *j) +{ + stbi__free_jpeg_components(j, j->s->img_n, 0); +} + +typedef struct +{ + resample_row_func resample; + stbi_uc *line0,*line1; + int hs,vs; // expansion factor in each axis + int w_lores; // horizontal pixels pre-expansion + int ystep; // how far through vertical expansion we are + int ypos; // which pre-expansion row we're on +} stbi__resample; + +// fast 0..255 * 0..255 => 0..255 rounded multiplication +static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) +{ + unsigned int t = x*y + 128; + return (stbi_uc) ((t + (t >>8)) >> 8); +} + +static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, int *comp, int req_comp) +{ + int n, decode_n, is_rgb; + z->s->img_n = 0; // make stbi__cleanup_jpeg safe + + // validate req_comp + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + + // load a jpeg image from whichever source, but leave in YCbCr format + if (!stbi__decode_jpeg_image(z)) { stbi__cleanup_jpeg(z); return NULL; } + + // determine actual number of components to generate + n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; + + is_rgb = z->s->img_n == 3 && (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); + + if (z->s->img_n == 3 && n < 3 && !is_rgb) + decode_n = 1; + else + decode_n = z->s->img_n; + + // nothing to do if no components requested; check this now to avoid + // accessing uninitialized coutput[0] later + if (decode_n <= 0) { stbi__cleanup_jpeg(z); return NULL; } + + // resample and color-convert + { + int k; + unsigned int i,j; + stbi_uc *output; + stbi_uc *coutput[4] = { NULL, NULL, NULL, NULL }; + + stbi__resample res_comp[4]; + + for (k=0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + + // allocate line buffer big enough for upsampling off the edges + // with upsample factor of 4 + z->img_comp[k].linebuf = (stbi_uc *) stbi__malloc(z->s->img_x + 3); + if (!z->img_comp[k].linebuf) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + + r->hs = z->img_h_max / z->img_comp[k].h; + r->vs = z->img_v_max / z->img_comp[k].v; + r->ystep = r->vs >> 1; + r->w_lores = (z->s->img_x + r->hs-1) / r->hs; + r->ypos = 0; + r->line0 = r->line1 = z->img_comp[k].data; + + if (r->hs == 1 && r->vs == 1) r->resample = resample_row_1; + else if (r->hs == 1 && r->vs == 2) r->resample = stbi__resample_row_v_2; + else if (r->hs == 2 && r->vs == 1) r->resample = stbi__resample_row_h_2; + else if (r->hs == 2 && r->vs == 2) r->resample = z->resample_row_hv_2_kernel; + else r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc *) stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); + if (!output) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + + // now go ahead and resample + for (j=0; j < z->s->img_y; ++j) { + stbi_uc *out = output + n * z->s->img_x * j; + for (k=0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + int y_bot = r->ystep >= (r->vs >> 1); + coutput[k] = r->resample(z->img_comp[k].linebuf, + y_bot ? r->line1 : r->line0, + y_bot ? r->line0 : r->line1, + r->w_lores, r->hs); + if (++r->ystep >= r->vs) { + r->ystep = 0; + r->line0 = r->line1; + if (++r->ypos < z->img_comp[k].y) + r->line1 += z->img_comp[k].w2; + } + } + if (n >= 3) { + stbi_uc *y = coutput[0]; + if (z->s->img_n == 3) { + if (is_rgb) { + for (i=0; i < z->s->img_x; ++i) { + out[0] = y[i]; + out[1] = coutput[1][i]; + out[2] = coutput[2][i]; + out[3] = 255; + out += n; + } + } else { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } else if (z->s->img_n == 4) { + if (z->app14_color_transform == 0) { // CMYK + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(coutput[0][i], m); + out[1] = stbi__blinn_8x8(coutput[1][i], m); + out[2] = stbi__blinn_8x8(coutput[2][i], m); + out[3] = 255; + out += n; + } + } else if (z->app14_color_transform == 2) { // YCCK + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(255 - out[0], m); + out[1] = stbi__blinn_8x8(255 - out[1], m); + out[2] = stbi__blinn_8x8(255 - out[2], m); + out += n; + } + } else { // YCbCr + alpha? Ignore the fourth channel for now + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } else + for (i=0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } else { + if (is_rgb) { + if (n == 1) + for (i=0; i < z->s->img_x; ++i) + *out++ = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + else { + for (i=0; i < z->s->img_x; ++i, out += 2) { + out[0] = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + out[1] = 255; + } + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); + stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); + stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); + out[0] = stbi__compute_y(r, g, b); + out[1] = 255; + out += n; + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { + for (i=0; i < z->s->img_x; ++i) { + out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); + out[1] = 255; + out += n; + } + } else { + stbi_uc *y = coutput[0]; + if (n == 1) + for (i=0; i < z->s->img_x; ++i) out[i] = y[i]; + else + for (i=0; i < z->s->img_x; ++i) { *out++ = y[i]; *out++ = 255; } + } + } + } + stbi__cleanup_jpeg(z); + *out_x = z->s->img_x; + *out_y = z->s->img_y; + if (comp) *comp = z->s->img_n >= 3 ? 3 : 1; // report original components, not output + return output; + } +} + +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + unsigned char* result; + stbi__jpeg* j = (stbi__jpeg*) stbi__malloc(sizeof(stbi__jpeg)); + if (!j) return stbi__errpuc("outofmem", "Out of memory"); + memset(j, 0, sizeof(stbi__jpeg)); + STBI_NOTUSED(ri); + j->s = s; + stbi__setup_jpeg(j); + result = load_jpeg_image(j, x,y,comp,req_comp); + STBI_FREE(j); + return result; +} + +static int stbi__jpeg_test(stbi__context *s) +{ + int r; + stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); + if (!j) return stbi__err("outofmem", "Out of memory"); + memset(j, 0, sizeof(stbi__jpeg)); + j->s = s; + stbi__setup_jpeg(j); + r = stbi__decode_jpeg_header(j, STBI__SCAN_type); + stbi__rewind(s); + STBI_FREE(j); + return r; +} + +static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) +{ + if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { + stbi__rewind( j->s ); + return 0; + } + if (x) *x = j->s->img_x; + if (y) *y = j->s->img_y; + if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; + return 1; +} + +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) +{ + int result; + stbi__jpeg* j = (stbi__jpeg*) (stbi__malloc(sizeof(stbi__jpeg))); + if (!j) return stbi__err("outofmem", "Out of memory"); + memset(j, 0, sizeof(stbi__jpeg)); + j->s = s; + result = stbi__jpeg_info_raw(j, x, y, comp); + STBI_FREE(j); + return result; +} +#endif + +// public domain zlib decode v0.2 Sean Barrett 2006-11-18 +// simple implementation +// - all input must be provided in an upfront buffer +// - all output is written to a single output buffer (can malloc/realloc) +// performance +// - fast huffman + +#ifndef STBI_NO_ZLIB + +// fast-way is faster to check than jpeg huffman, but slow way is slower +#define STBI__ZFAST_BITS 9 // accelerate all cases in default tables +#define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) +#define STBI__ZNSYMS 288 // number of symbols in literal/length alphabet + +// zlib-style huffman encoding +// (jpegs packs from left, zlib from right, so can't share code) +typedef struct +{ + stbi__uint16 fast[1 << STBI__ZFAST_BITS]; + stbi__uint16 firstcode[16]; + int maxcode[17]; + stbi__uint16 firstsymbol[16]; + stbi_uc size[STBI__ZNSYMS]; + stbi__uint16 value[STBI__ZNSYMS]; +} stbi__zhuffman; + +stbi_inline static int stbi__bitreverse16(int n) +{ + n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); + n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); + n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); + n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); + return n; +} + +stbi_inline static int stbi__bit_reverse(int v, int bits) +{ + STBI_ASSERT(bits <= 16); + // to bit reverse n bits, reverse 16 and shift + // e.g. 11 bits, bit reverse and shift away 5 + return stbi__bitreverse16(v) >> (16-bits); +} + +static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int num) +{ + int i,k=0; + int code, next_code[16], sizes[17]; + + // DEFLATE spec for generating codes + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 0, sizeof(z->fast)); + for (i=0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i=1; i < 16; ++i) + if (sizes[i] > (1 << i)) + return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i=1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (stbi__uint16) code; + z->firstsymbol[i] = (stbi__uint16) k; + code = (code + sizes[i]); + if (sizes[i]) + if (code-1 >= (1 << i)) return stbi__err("bad codelengths","Corrupt PNG"); + z->maxcode[i] = code << (16-i); // preshift for inner loop + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; // sentinel + for (i=0; i < num; ++i) { + int s = sizelist[i]; + if (s) { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + stbi__uint16 fastv = (stbi__uint16) ((s << 9) | i); + z->size [c] = (stbi_uc ) s; + z->value[c] = (stbi__uint16) i; + if (s <= STBI__ZFAST_BITS) { + int j = stbi__bit_reverse(next_code[s],s); + while (j < (1 << STBI__ZFAST_BITS)) { + z->fast[j] = fastv; + j += (1 << s); + } + } + ++next_code[s]; + } + } + return 1; +} + +// zlib-from-memory implementation for PNG reading +// because PNG allows splitting the zlib stream arbitrarily, +// and it's annoying structurally to have PNG call ZLIB call PNG, +// we require PNG read all the IDATs and combine them into a single +// memory buffer + +typedef struct +{ + stbi_uc *zbuffer, *zbuffer_end; + int num_bits; + int hit_zeof_once; + stbi__uint32 code_buffer; + + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; + + stbi__zhuffman z_length, z_distance; +} stbi__zbuf; + +stbi_inline static int stbi__zeof(stbi__zbuf *z) +{ + return (z->zbuffer >= z->zbuffer_end); +} + +stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) +{ + return stbi__zeof(z) ? 0 : *z->zbuffer++; +} + +static void stbi__fill_bits(stbi__zbuf *z) +{ + do { + if (z->code_buffer >= (1U << z->num_bits)) { + z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */ + return; + } + z->code_buffer |= (unsigned int) stbi__zget8(z) << z->num_bits; + z->num_bits += 8; + } while (z->num_bits <= 24); +} + +stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) +{ + unsigned int k; + if (z->num_bits < n) stbi__fill_bits(z); + k = z->code_buffer & ((1 << n) - 1); + z->code_buffer >>= n; + z->num_bits -= n; + return k; +} + +static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s,k; + // not resolved by fast table, so compute it the slow way + // use jpeg approach, which requires MSbits at top + k = stbi__bit_reverse(a->code_buffer, 16); + for (s=STBI__ZFAST_BITS+1; ; ++s) + if (k < z->maxcode[s]) + break; + if (s >= 16) return -1; // invalid code! + // code size is s, so: + b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s]; + if (b >= STBI__ZNSYMS) return -1; // some data was corrupt somewhere! + if (z->size[b] != s) return -1; // was originally an assert, but report failure instead. + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; +} + +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s; + if (a->num_bits < 16) { + if (stbi__zeof(a)) { + if (!a->hit_zeof_once) { + // This is the first time we hit eof, insert 16 extra padding btis + // to allow us to keep going; if we actually consume any of them + // though, that is invalid data. This is caught later. + a->hit_zeof_once = 1; + a->num_bits += 16; // add 16 implicit zero bits + } else { + // We already inserted our extra 16 padding bits and are again + // out, this stream is actually prematurely terminated. + return -1; + } + } else { + stbi__fill_bits(a); + } + } + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); +} + +static int stbi__zexpand(stbi__zbuf *z, char *zout, int n) // need to make room for n bytes +{ + char *q; + unsigned int cur, limit, old_limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("output buffer limit","Corrupt PNG"); + cur = (unsigned int) (z->zout - z->zout_start); + limit = old_limit = (unsigned) (z->zout_end - z->zout_start); + if (UINT_MAX - cur < (unsigned) n) return stbi__err("outofmem", "Out of memory"); + while (cur + n > limit) { + if(limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory"); + limit *= 2; + } + q = (char *) STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); + STBI_NOTUSED(old_limit); + if (q == NULL) return stbi__err("outofmem", "Out of memory"); + z->zout_start = q; + z->zout = q + cur; + z->zout_end = q + limit; + return 1; +} + +static const int stbi__zlength_base[31] = { + 3,4,5,6,7,8,9,10,11,13, + 15,17,19,23,27,31,35,43,51,59, + 67,83,99,115,131,163,195,227,258,0,0 }; + +static const int stbi__zlength_extra[31]= +{ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0,0,0 }; + +static const int stbi__zdist_base[32] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193, +257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577,0,0}; + +static const int stbi__zdist_extra[32] = +{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; + +static int stbi__parse_huffman_block(stbi__zbuf *a) +{ + char *zout = a->zout; + for(;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); // error in huffman codes + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char) z; + } else { + stbi_uc *p; + int len,dist; + if (z == 256) { + a->zout = zout; + if (a->hit_zeof_once && a->num_bits < 16) { + // The first time we hit zeof, we inserted 16 extra zero bits into our bit + // buffer so the decoder can just do its speculative decoding. But if we + // actually consumed any of those bits (which is the case when num_bits < 16), + // the stream actually read past the end so it is malformed. + return stbi__err("unexpected end","Corrupt PNG"); + } + return 1; + } + if (z >= 286) return stbi__err("bad huffman code","Corrupt PNG"); // per DEFLATE, length codes 286 and 287 must not appear in compressed data + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0 || z >= 30) return stbi__err("bad huffman code","Corrupt PNG"); // per DEFLATE, distance codes 30 and 31 must not appear in compressed data + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) return stbi__err("bad dist","Corrupt PNG"); + if (len > a->zout_end - zout) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc *) (zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { do *zout++ = v; while (--len); } + } else { + if (len) { do *zout++ = *p++; while (--len); } + } + } + } +} + +static int stbi__compute_huffman_codes(stbi__zbuf *a) +{ + static const stbi_uc length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; + stbi__zhuffman z_codelength; + stbi_uc lencodes[286+32+137];//padding for maximum single op + stbi_uc codelength_sizes[19]; + int i,n; + + int hlit = stbi__zreceive(a,5) + 257; + int hdist = stbi__zreceive(a,5) + 1; + int hclen = stbi__zreceive(a,4) + 4; + int ntot = hlit + hdist; + + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i=0; i < hclen; ++i) { + int s = stbi__zreceive(a,3); + codelength_sizes[length_dezigzag[i]] = (stbi_uc) s; + } + if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + + n = 0; + while (n < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc) c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a,2)+3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n-1]; + } else if (c == 17) { + c = stbi__zreceive(a,3)+3; + } else if (c == 18) { + c = stbi__zreceive(a,7)+11; + } else { + return stbi__err("bad codelengths", "Corrupt PNG"); + } + if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); + memset(lencodes+n, fill, c); + n += c; + } + } + if (n != ntot) return stbi__err("bad codelengths","Corrupt PNG"); + if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0; + return 1; +} + +static int stbi__parse_uncompressed_block(stbi__zbuf *a) +{ + stbi_uc header[4]; + int len,nlen,k; + if (a->num_bits & 7) + stbi__zreceive(a, a->num_bits & 7); // discard + // drain the bit-packed data into header + k = 0; + while (a->num_bits > 0) { + header[k++] = (stbi_uc) (a->code_buffer & 255); // suppress MSVC run-time check + a->code_buffer >>= 8; + a->num_bits -= 8; + } + if (a->num_bits < 0) return stbi__err("zlib corrupt","Corrupt PNG"); + // now fill header the normal way + while (k < 4) + header[k++] = stbi__zget8(a); + len = header[1] * 256 + header[0]; + nlen = header[3] * 256 + header[2]; + if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt","Corrupt PNG"); + if (a->zbuffer + len > a->zbuffer_end) return stbi__err("read past buffer","Corrupt PNG"); + if (a->zout + len > a->zout_end) + if (!stbi__zexpand(a, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; +} + +static int stbi__parse_zlib_header(stbi__zbuf *a) +{ + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if (stbi__zeof(a)) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec + if ((cmf*256+flg) % 31 != 0) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec + if (flg & 32) return stbi__err("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png + if (cm != 8) return stbi__err("bad compression","Corrupt PNG"); // DEFLATE required for png + // window = 1 << (8 + cinfo)... but who cares, we fully buffer output + return 1; +} + +static const stbi_uc stbi__zdefault_length[STBI__ZNSYMS] = +{ + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,8 +}; +static const stbi_uc stbi__zdefault_distance[32] = +{ + 5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5 +}; +/* +Init algorithm: +{ + int i; // use <= to match clearly with spec + for (i=0; i <= 143; ++i) stbi__zdefault_length[i] = 8; + for ( ; i <= 255; ++i) stbi__zdefault_length[i] = 9; + for ( ; i <= 279; ++i) stbi__zdefault_length[i] = 7; + for ( ; i <= 287; ++i) stbi__zdefault_length[i] = 8; + + for (i=0; i <= 31; ++i) stbi__zdefault_distance[i] = 5; +} +*/ + +static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) +{ + int final, type; + if (parse_header) + if (!stbi__parse_zlib_header(a)) return 0; + a->num_bits = 0; + a->code_buffer = 0; + a->hit_zeof_once = 0; + do { + final = stbi__zreceive(a,1); + type = stbi__zreceive(a,2); + if (type == 0) { + if (!stbi__parse_uncompressed_block(a)) return 0; + } else if (type == 3) { + return 0; + } else { + if (type == 1) { + // use fixed code lengths + if (!stbi__zbuild_huffman(&a->z_length , stbi__zdefault_length , STBI__ZNSYMS)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) return 0; + } else { + if (!stbi__compute_huffman_codes(a)) return 0; + } + if (!stbi__parse_huffman_block(a)) return 0; + } + } while (!final); + return 1; +} + +static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, int parse_header) +{ + a->zout_start = obuf; + a->zout = obuf; + a->zout_end = obuf + olen; + a->z_expandable = exp; + + return stbi__parse_zlib(a, parse_header); +} + +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen) +{ + return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); +} + +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen) +{ + stbi__zbuf a; + a.zbuffer = (stbi_uc *) ibuffer; + a.zbuffer_end = (stbi_uc *) ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) + return (int) (a.zout - a.zout_start); + else + return -1; +} + +STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(16384); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer+len; + if (stbi__do_zlib(&a, p, 16384, 1, 0)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen) +{ + stbi__zbuf a; + a.zbuffer = (stbi_uc *) ibuffer; + a.zbuffer_end = (stbi_uc *) ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) + return (int) (a.zout - a.zout_start); + else + return -1; +} +#endif + +// public domain "baseline" PNG decoder v0.10 Sean Barrett 2006-11-18 +// simple implementation +// - only 8-bit samples +// - no CRC checking +// - allocates lots of intermediate memory +// - avoids problem of streaming data between subsystems +// - avoids explicit window management +// performance +// - uses stb_zlib, a PD zlib implementation with fast huffman decoding + +#ifndef STBI_NO_PNG +typedef struct +{ + stbi__uint32 length; + stbi__uint32 type; +} stbi__pngchunk; + +static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) +{ + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; +} + +static int stbi__check_png_header(stbi__context *s) +{ + static const stbi_uc png_sig[8] = { 137,80,78,71,13,10,26,10 }; + int i; + for (i=0; i < 8; ++i) + if (stbi__get8(s) != png_sig[i]) return stbi__err("bad png sig","Not a PNG"); + return 1; +} + +typedef struct +{ + stbi__context *s; + stbi_uc *idata, *expanded, *out; + int depth; +} stbi__png; + + +enum { + STBI__F_none=0, + STBI__F_sub=1, + STBI__F_up=2, + STBI__F_avg=3, + STBI__F_paeth=4, + // synthetic filter used for first scanline to avoid needing a dummy row of 0s + STBI__F_avg_first +}; + +static stbi_uc first_row_filter[5] = +{ + STBI__F_none, + STBI__F_sub, + STBI__F_none, + STBI__F_avg_first, + STBI__F_sub // Paeth with b=c=0 turns out to be equivalent to sub +}; + +static int stbi__paeth(int a, int b, int c) +{ + // This formulation looks very different from the reference in the PNG spec, but is + // actually equivalent and has favorable data dependencies and admits straightforward + // generation of branch-free code, which helps performance significantly. + int thresh = c*3 - (a + b); + int lo = a < b ? a : b; + int hi = a < b ? b : a; + int t0 = (hi <= thresh) ? lo : c; + int t1 = (thresh <= lo) ? hi : t0; + return t1; +} + +static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; + +// adds an extra all-255 alpha channel +// dest == src is legal +// img_n must be 1 or 3 +static void stbi__create_png_alpha_expand8(stbi_uc *dest, stbi_uc *src, stbi__uint32 x, int img_n) +{ + int i; + // must process data backwards since we allow dest==src + if (img_n == 1) { + for (i=x-1; i >= 0; --i) { + dest[i*2+1] = 255; + dest[i*2+0] = src[i]; + } + } else { + STBI_ASSERT(img_n == 3); + for (i=x-1; i >= 0; --i) { + dest[i*4+3] = 255; + dest[i*4+2] = src[i*3+2]; + dest[i*4+1] = src[i*3+1]; + dest[i*4+0] = src[i*3+0]; + } + } +} + +// create the png data from post-deflated data +static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, stbi__uint32 x, stbi__uint32 y, int depth, int color) +{ + int bytes = (depth == 16 ? 2 : 1); + stbi__context *s = a->s; + stbi__uint32 i,j,stride = x*out_n*bytes; + stbi__uint32 img_len, img_width_bytes; + stbi_uc *filter_buf; + int all_ok = 1; + int k; + int img_n = s->img_n; // copy it into a local for later + + int output_bytes = out_n*bytes; + int filter_bytes = img_n*bytes; + int width = x; + + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); + a->out = (stbi_uc *) stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); + + // note: error exits here don't need to clean up a->out individually, + // stbi__do_png always does on error. + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + if (!stbi__mad2sizes_valid(img_width_bytes, y, img_width_bytes)) return stbi__err("too large", "Corrupt PNG"); + img_len = (img_width_bytes + 1) * y; + + // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, + // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), + // so just check for raw_len < img_len always. + if (raw_len < img_len) return stbi__err("not enough pixels","Corrupt PNG"); + + // Allocate two scan lines worth of filter workspace buffer. + filter_buf = (stbi_uc *) stbi__malloc_mad2(img_width_bytes, 2, 0); + if (!filter_buf) return stbi__err("outofmem", "Out of memory"); + + // Filtering for low-bit-depth images + if (depth < 8) { + filter_bytes = 1; + width = img_width_bytes; + } + + for (j=0; j < y; ++j) { + // cur/prior filter buffers alternate + stbi_uc *cur = filter_buf + (j & 1)*img_width_bytes; + stbi_uc *prior = filter_buf + (~j & 1)*img_width_bytes; + stbi_uc *dest = a->out + stride*j; + int nk = width * filter_bytes; + int filter = *raw++; + + // check filter type + if (filter > 4) { + all_ok = stbi__err("invalid filter","Corrupt PNG"); + break; + } + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // perform actual filtering + switch (filter) { + case STBI__F_none: + memcpy(cur, raw, nk); + break; + case STBI__F_sub: + memcpy(cur, raw, filter_bytes); + for (k = filter_bytes; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]); + break; + case STBI__F_up: + for (k = 0; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + break; + case STBI__F_avg: + for (k = 0; k < filter_bytes; ++k) + cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1)); + for (k = filter_bytes; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1)); + break; + case STBI__F_paeth: + for (k = 0; k < filter_bytes; ++k) + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); // prior[k] == stbi__paeth(0,prior[k],0) + for (k = filter_bytes; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes], prior[k], prior[k-filter_bytes])); + break; + case STBI__F_avg_first: + memcpy(cur, raw, filter_bytes); + for (k = filter_bytes; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1)); + break; + } + + raw += nk; + + // expand decoded bits in cur to dest, also adding an extra alpha channel if desired + if (depth < 8) { + stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range + stbi_uc *in = cur; + stbi_uc *out = dest; + stbi_uc inb = 0; + stbi__uint32 nsmp = x*img_n; + + // expand bits to bytes first + if (depth == 4) { + for (i=0; i < nsmp; ++i) { + if ((i & 1) == 0) inb = *in++; + *out++ = scale * (inb >> 4); + inb <<= 4; + } + } else if (depth == 2) { + for (i=0; i < nsmp; ++i) { + if ((i & 3) == 0) inb = *in++; + *out++ = scale * (inb >> 6); + inb <<= 2; + } + } else { + STBI_ASSERT(depth == 1); + for (i=0; i < nsmp; ++i) { + if ((i & 7) == 0) inb = *in++; + *out++ = scale * (inb >> 7); + inb <<= 1; + } + } + + // insert alpha=255 values if desired + if (img_n != out_n) + stbi__create_png_alpha_expand8(dest, dest, x, img_n); + } else if (depth == 8) { + if (img_n == out_n) + memcpy(dest, cur, x*img_n); + else + stbi__create_png_alpha_expand8(dest, cur, x, img_n); + } else if (depth == 16) { + // convert the image data from big-endian to platform-native + stbi__uint16 *dest16 = (stbi__uint16*)dest; + stbi__uint32 nsmp = x*img_n; + + if (img_n == out_n) { + for (i = 0; i < nsmp; ++i, ++dest16, cur += 2) + *dest16 = (cur[0] << 8) | cur[1]; + } else { + STBI_ASSERT(img_n+1 == out_n); + if (img_n == 1) { + for (i = 0; i < x; ++i, dest16 += 2, cur += 2) { + dest16[0] = (cur[0] << 8) | cur[1]; + dest16[1] = 0xffff; + } + } else { + STBI_ASSERT(img_n == 3); + for (i = 0; i < x; ++i, dest16 += 4, cur += 6) { + dest16[0] = (cur[0] << 8) | cur[1]; + dest16[1] = (cur[2] << 8) | cur[3]; + dest16[2] = (cur[4] << 8) | cur[5]; + dest16[3] = 0xffff; + } + } + } + } + } + + STBI_FREE(filter_buf); + if (!all_ok) return 0; + + return 1; +} + +static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) +{ + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); + + // de-interlacing + final = (stbi_uc *) stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + if (!final) return stbi__err("outofmem", "Out of memory"); + for (p=0; p < 7; ++p) { + int xorig[] = { 0,4,0,2,0,1,0 }; + int yorig[] = { 0,0,4,0,2,0,1 }; + int xspc[] = { 8,8,4,4,2,2,1 }; + int yspc[] = { 8,8,8,4,4,2,2 }; + int i,j,x,y; + // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 + x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p]; + y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p]; + if (x && y) { + stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { + STBI_FREE(final); + return 0; + } + for (j=0; j < y; ++j) { + for (i=0; i < x; ++i) { + int out_y = j*yspc[p]+yorig[p]; + int out_x = i*xspc[p]+xorig[p]; + memcpy(final + out_y*a->s->img_x*out_bytes + out_x*out_bytes, + a->out + (j*x+i)*out_bytes, out_bytes); + } + } + STBI_FREE(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; + + return 1; +} + +static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; + + // compute color-based transparency, assuming we've + // already got 255 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); + + if (out_n == 2) { + for (i=0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 255); + p += 2; + } + } else { + for (i=0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) + p[3] = 0; + p += 4; + } + } + return 1; +} + +static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], int out_n) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi__uint16 *p = (stbi__uint16*) z->out; + + // compute color-based transparency, assuming we've + // already got 65535 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); + + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 65535); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) + p[3] = 0; + p += 4; + } + } + return 1; +} + +static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, int pal_img_n) +{ + stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; + stbi_uc *p, *temp_out, *orig = a->out; + + p = (stbi_uc *) stbi__malloc_mad2(pixel_count, pal_img_n, 0); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); + + // between here and free(out) below, exitting would leak + temp_out = p; + + if (pal_img_n == 3) { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p += 3; + } + } else { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p[3] = palette[n+3]; + p += 4; + } + } + STBI_FREE(a->out); + a->out = temp_out; + + STBI_NOTUSED(len); + + return 1; +} + +static int stbi__unpremultiply_on_load_global = 0; +static int stbi__de_iphone_flag_global = 0; + +STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply) +{ + stbi__unpremultiply_on_load_global = flag_true_if_should_unpremultiply; +} + +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) +{ + stbi__de_iphone_flag_global = flag_true_if_should_convert; +} + +#ifndef STBI_THREAD_LOCAL +#define stbi__unpremultiply_on_load stbi__unpremultiply_on_load_global +#define stbi__de_iphone_flag stbi__de_iphone_flag_global +#else +static STBI_THREAD_LOCAL int stbi__unpremultiply_on_load_local, stbi__unpremultiply_on_load_set; +static STBI_THREAD_LOCAL int stbi__de_iphone_flag_local, stbi__de_iphone_flag_set; + +STBIDEF void stbi_set_unpremultiply_on_load_thread(int flag_true_if_should_unpremultiply) +{ + stbi__unpremultiply_on_load_local = flag_true_if_should_unpremultiply; + stbi__unpremultiply_on_load_set = 1; +} + +STBIDEF void stbi_convert_iphone_png_to_rgb_thread(int flag_true_if_should_convert) +{ + stbi__de_iphone_flag_local = flag_true_if_should_convert; + stbi__de_iphone_flag_set = 1; +} + +#define stbi__unpremultiply_on_load (stbi__unpremultiply_on_load_set \ + ? stbi__unpremultiply_on_load_local \ + : stbi__unpremultiply_on_load_global) +#define stbi__de_iphone_flag (stbi__de_iphone_flag_set \ + ? stbi__de_iphone_flag_local \ + : stbi__de_iphone_flag_global) +#endif // STBI_THREAD_LOCAL + +static void stbi__de_iphone(stbi__png *z) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; + + if (s->img_out_n == 3) { // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 3; + } + } else { + STBI_ASSERT(s->img_out_n == 4); + if (stbi__unpremultiply_on_load) { + // convert bgr to rgb and unpremultiply + for (i=0; i < pixel_count; ++i) { + stbi_uc a = p[3]; + stbi_uc t = p[0]; + if (a) { + stbi_uc half = a / 2; + p[0] = (p[2] * 255 + half) / a; + p[1] = (p[1] * 255 + half) / a; + p[2] = ( t * 255 + half) / a; + } else { + p[0] = p[2]; + p[2] = t; + } + p += 4; + } + } else { + // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; + } + } + } +} + +#define STBI__PNG_TYPE(a,b,c,d) (((unsigned) (a) << 24) + ((unsigned) (b) << 16) + ((unsigned) (c) << 8) + (unsigned) (d)) + +static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) +{ + stbi_uc palette[1024], pal_img_n=0; + stbi_uc has_trans=0, tc[3]={0}; + stbi__uint16 tc16[3]; + stbi__uint32 ioff=0, idata_limit=0, i, pal_len=0; + int first=1,k,interlace=0, color=0, is_iphone=0; + stbi__context *s = z->s; + + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; + + if (!stbi__check_png_header(s)) return 0; + + if (scan == STBI__SCAN_type) return 1; + + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C','g','B','I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I','H','D','R'): { + int comp,filter; + if (!first) return stbi__err("multiple IHDR","Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len","Corrupt PNG"); + s->img_x = stbi__get32be(s); + s->img_y = stbi__get32be(s); + if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + z->depth = stbi__get8(s); if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && z->depth != 16) return stbi__err("1/2/4/8/16-bit only","PNG not supported: 1/2/4/8/16-bit only"); + color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype","Corrupt PNG"); + if (color == 3 && z->depth == 16) return stbi__err("bad ctype","Corrupt PNG"); + if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype","Corrupt PNG"); + comp = stbi__get8(s); if (comp) return stbi__err("bad comp method","Corrupt PNG"); + filter= stbi__get8(s); if (filter) return stbi__err("bad filter method","Corrupt PNG"); + interlace = stbi__get8(s); if (interlace>1) return stbi__err("bad interlace method","Corrupt PNG"); + if (!s->img_x || !s->img_y) return stbi__err("0-pixel image","Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("too large", "Image too large to decode"); + } else { + // if paletted, then pal_n is our final components, and + // img_n is # components to decompress/filter. + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) return stbi__err("too large","Corrupt PNG"); + } + // even with SCAN_header, have to scan to see if we have a tRNS + break; + } + + case STBI__PNG_TYPE('P','L','T','E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256*3) return stbi__err("invalid PLTE","Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) return stbi__err("invalid PLTE","Corrupt PNG"); + for (i=0; i < pal_len; ++i) { + palette[i*4+0] = stbi__get8(s); + palette[i*4+1] = stbi__get8(s); + palette[i*4+2] = stbi__get8(s); + palette[i*4+3] = 255; + } + break; + } + + case STBI__PNG_TYPE('t','R','N','S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT","Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } + if (pal_len == 0) return stbi__err("tRNS before PLTE","Corrupt PNG"); + if (c.length > pal_len) return stbi__err("bad tRNS len","Corrupt PNG"); + pal_img_n = 4; + for (i=0; i < c.length; ++i) + palette[i*4+3] = stbi__get8(s); + } else { + if (!(s->img_n & 1)) return stbi__err("tRNS with alpha","Corrupt PNG"); + if (c.length != (stbi__uint32) s->img_n*2) return stbi__err("bad tRNS len","Corrupt PNG"); + has_trans = 1; + // non-paletted with tRNS = constant alpha. if header-scanning, we can stop now. + if (scan == STBI__SCAN_header) { ++s->img_n; return 1; } + if (z->depth == 16) { + for (k = 0; k < s->img_n && k < 3; ++k) // extra loop test to suppress false GCC warning + tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is + } else { + for (k = 0; k < s->img_n && k < 3; ++k) + tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger + } + } + break; + } + + case STBI__PNG_TYPE('I','D','A','T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE","Corrupt PNG"); + if (scan == STBI__SCAN_header) { + // header scan definitely stops at first IDAT + if (pal_img_n) + s->img_n = pal_img_n; + return 1; + } + if (c.length > (1u << 30)) return stbi__err("IDAT size limit", "IDAT section larger than 2^30 bytes"); + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + stbi__uint32 idata_limit_old = idata_limit; + stbi_uc *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + STBI_NOTUSED(idata_limit_old); + p = (stbi_uc *) STBI_REALLOC_SIZED(z->idata, idata_limit_old, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata+ioff,c.length)) return stbi__err("outofdata","Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I','E','N','D'): { + stbi__uint32 raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT","Corrupt PNG"); + // initial guess for decoded data size to avoid unnecessary reallocs + bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component + raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; + z->expanded = (stbi_uc *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, raw_len, (int *) &raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + STBI_FREE(z->idata); z->idata = NULL; + if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans) + s->img_out_n = s->img_n+1; + else + s->img_out_n = s->img_n; + if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, z->depth, color, interlace)) return 0; + if (has_trans) { + if (z->depth == 16) { + if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; + } else { + if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; + } + } + if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) + stbi__de_iphone(z); + if (pal_img_n) { + // pal_img_n == 3 or 4 + s->img_n = pal_img_n; // record the actual colors we had + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } else if (has_trans) { + // non-paletted image with tRNS -> source image has (constant) alpha + ++s->img_n; + } + STBI_FREE(z->expanded); z->expanded = NULL; + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + return 1; + } + + default: + // if critical, fail + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + // not threadsafe + static char invalid_chunk[] = "XXXX PNG chunk not known"; + invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); + invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); + invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); + invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); + #endif + return stbi__err(invalid_chunk, "PNG not supported: unknown PNG chunk type"); + } + stbi__skip(s, c.length); + break; + } + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + } +} + +static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, stbi__result_info *ri) +{ + void *result=NULL; + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + if (p->depth <= 8) + ri->bits_per_channel = 8; + else if (p->depth == 16) + ri->bits_per_channel = 16; + else + return stbi__errpuc("bad bits_per_channel", "PNG not supported: unsupported color depth"); + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + if (ri->bits_per_channel == 8) + result = stbi__convert_format((unsigned char *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + else + result = stbi__convert_format16((stbi__uint16 *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n) *n = p->s->img_n; + } + STBI_FREE(p->out); p->out = NULL; + STBI_FREE(p->expanded); p->expanded = NULL; + STBI_FREE(p->idata); p->idata = NULL; + + return result; +} + +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi__png p; + p.s = s; + return stbi__do_png(&p, x,y,comp,req_comp, ri); +} + +static int stbi__png_test(stbi__context *s) +{ + int r; + r = stbi__check_png_header(s); + stbi__rewind(s); + return r; +} + +static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) +{ + if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { + stbi__rewind( p->s ); + return 0; + } + if (x) *x = p->s->img_x; + if (y) *y = p->s->img_y; + if (comp) *comp = p->s->img_n; + return 1; +} + +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__png p; + p.s = s; + return stbi__png_info_raw(&p, x, y, comp); +} + +static int stbi__png_is16(stbi__context *s) +{ + stbi__png p; + p.s = s; + if (!stbi__png_info_raw(&p, NULL, NULL, NULL)) + return 0; + if (p.depth != 16) { + stbi__rewind(p.s); + return 0; + } + return 1; +} +#endif + +// Microsoft/Windows BMP image + +#ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context *s) +{ + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + stbi__get32le(s); // discard data offset + sz = stbi__get32le(s); + r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); + return r; +} + +static int stbi__bmp_test(stbi__context *s) +{ + int r = stbi__bmp_test_raw(s); + stbi__rewind(s); + return r; +} + + +// returns 0..31 for the highest set bit +static int stbi__high_bit(unsigned int z) +{ + int n=0; + if (z == 0) return -1; + if (z >= 0x10000) { n += 16; z >>= 16; } + if (z >= 0x00100) { n += 8; z >>= 8; } + if (z >= 0x00010) { n += 4; z >>= 4; } + if (z >= 0x00004) { n += 2; z >>= 2; } + if (z >= 0x00002) { n += 1;/* >>= 1;*/ } + return n; +} + +static int stbi__bitcount(unsigned int a) +{ + a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 + a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits + a = (a + (a >> 8)); // max 16 per 8 bits + a = (a + (a >> 16)); // max 32 per 8 bits + return a & 0xff; +} + +// extract an arbitrarily-aligned N-bit value (N=bits) +// from v, and then make it 8-bits long and fractionally +// extend it to full full range. +static int stbi__shiftsigned(unsigned int v, int shift, int bits) +{ + static unsigned int mul_table[9] = { + 0, + 0xff/*0b11111111*/, 0x55/*0b01010101*/, 0x49/*0b01001001*/, 0x11/*0b00010001*/, + 0x21/*0b00100001*/, 0x41/*0b01000001*/, 0x81/*0b10000001*/, 0x01/*0b00000001*/, + }; + static unsigned int shift_table[9] = { + 0, 0,0,1,0,2,4,6,0, + }; + if (shift < 0) + v <<= -shift; + else + v >>= shift; + STBI_ASSERT(v < 256); + v >>= (8-bits); + STBI_ASSERT(bits >= 0 && bits <= 8); + return (int) ((unsigned) v * mul_table[bits]) >> shift_table[bits]; +} + +typedef struct +{ + int bpp, offset, hsz; + unsigned int mr,mg,mb,ma, all_a; + int extra_read; +} stbi__bmp_data; + +static int stbi__bmp_set_mask_defaults(stbi__bmp_data *info, int compress) +{ + // BI_BITFIELDS specifies masks explicitly, don't override + if (compress == 3) + return 1; + + if (compress == 0) { + if (info->bpp == 16) { + info->mr = 31u << 10; + info->mg = 31u << 5; + info->mb = 31u << 0; + } else if (info->bpp == 32) { + info->mr = 0xffu << 16; + info->mg = 0xffu << 8; + info->mb = 0xffu << 0; + info->ma = 0xffu << 24; + info->all_a = 0; // if all_a is 0 at end, then we loaded alpha channel but it was all 0 + } else { + // otherwise, use defaults, which is all-0 + info->mr = info->mg = info->mb = info->ma = 0; + } + return 1; + } + return 0; // error +} + +static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) +{ + int hsz; + if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') return stbi__errpuc("not BMP", "Corrupt BMP"); + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + info->offset = stbi__get32le(s); + info->hsz = hsz = stbi__get32le(s); + info->mr = info->mg = info->mb = info->ma = 0; + info->extra_read = 14; + + if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP"); + + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); + if (hsz == 12) { + s->img_x = stbi__get16le(s); + s->img_y = stbi__get16le(s); + } else { + s->img_x = stbi__get32le(s); + s->img_y = stbi__get32le(s); + } + if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); + info->bpp = stbi__get16le(s); + if (hsz != 12) { + int compress = stbi__get32le(s); + if (compress == 1 || compress == 2) return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); + if (compress >= 4) return stbi__errpuc("BMP JPEG/PNG", "BMP type not supported: unsupported compression"); // this includes PNG/JPEG modes + if (compress == 3 && info->bpp != 16 && info->bpp != 32) return stbi__errpuc("bad BMP", "bad BMP"); // bitfields requires 16 or 32 bits/pixel + stbi__get32le(s); // discard sizeof + stbi__get32le(s); // discard hres + stbi__get32le(s); // discard vres + stbi__get32le(s); // discard colorsused + stbi__get32le(s); // discard max important + if (hsz == 40 || hsz == 56) { + if (hsz == 56) { + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + } + if (info->bpp == 16 || info->bpp == 32) { + if (compress == 0) { + stbi__bmp_set_mask_defaults(info, compress); + } else if (compress == 3) { + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->extra_read += 12; + // not documented, but generated by photoshop and handled by mspaint + if (info->mr == info->mg && info->mg == info->mb) { + // ?!?!? + return stbi__errpuc("bad BMP", "bad BMP"); + } + } else + return stbi__errpuc("bad BMP", "bad BMP"); + } + } else { + // V4/V5 header + int i; + if (hsz != 108 && hsz != 124) + return stbi__errpuc("bad BMP", "bad BMP"); + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->ma = stbi__get32le(s); + if (compress != 3) // override mr/mg/mb unless in BI_BITFIELDS mode, as per docs + stbi__bmp_set_mask_defaults(info, compress); + stbi__get32le(s); // discard color space + for (i=0; i < 12; ++i) + stbi__get32le(s); // discard color space parameters + if (hsz == 124) { + stbi__get32le(s); // discard rendering intent + stbi__get32le(s); // discard offset of profile data + stbi__get32le(s); // discard size of profile data + stbi__get32le(s); // discard reserved + } + } + } + return (void *) 1; +} + + +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *out; + unsigned int mr=0,mg=0,mb=0,ma=0, all_a; + stbi_uc pal[256][4]; + int psize=0,i,j,width; + int flip_vertically, pad, target; + stbi__bmp_data info; + STBI_NOTUSED(ri); + + info.all_a = 255; + if (stbi__bmp_parse_header(s, &info) == NULL) + return NULL; // error code already set + + flip_vertically = ((int) s->img_y) > 0; + s->img_y = abs((int) s->img_y); + + if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + mr = info.mr; + mg = info.mg; + mb = info.mb; + ma = info.ma; + all_a = info.all_a; + + if (info.hsz == 12) { + if (info.bpp < 24) + psize = (info.offset - info.extra_read - 24) / 3; + } else { + if (info.bpp < 16) + psize = (info.offset - info.extra_read - info.hsz) >> 2; + } + if (psize == 0) { + // accept some number of extra bytes after the header, but if the offset points either to before + // the header ends or implies a large amount of extra data, reject the file as malformed + int bytes_read_so_far = s->callback_already_read + (int)(s->img_buffer - s->img_buffer_original); + int header_limit = 1024; // max we actually read is below 256 bytes currently. + int extra_data_limit = 256*4; // what ordinarily goes here is a palette; 256 entries*4 bytes is its max size. + if (bytes_read_so_far <= 0 || bytes_read_so_far > header_limit) { + return stbi__errpuc("bad header", "Corrupt BMP"); + } + // we established that bytes_read_so_far is positive and sensible. + // the first half of this test rejects offsets that are either too small positives, or + // negative, and guarantees that info.offset >= bytes_read_so_far > 0. this in turn + // ensures the number computed in the second half of the test can't overflow. + if (info.offset < bytes_read_so_far || info.offset - bytes_read_so_far > extra_data_limit) { + return stbi__errpuc("bad offset", "Corrupt BMP"); + } else { + stbi__skip(s, info.offset - bytes_read_so_far); + } + } + + if (info.bpp == 24 && ma == 0xff000000) + s->img_n = 3; + else + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 + target = req_comp; + else + target = s->img_n; // if they want monochrome, we'll post-convert + + // sanity-check size + if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "Corrupt BMP"); + + out = (stbi_uc *) stbi__malloc_mad3(target, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (info.bpp < 16) { + int z=0; + if (psize == 0 || psize > 256) { STBI_FREE(out); return stbi__errpuc("invalid", "Corrupt BMP"); } + for (i=0; i < psize; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + if (info.hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__skip(s, info.offset - info.extra_read - info.hsz - psize * (info.hsz == 12 ? 3 : 4)); + if (info.bpp == 1) width = (s->img_x + 7) >> 3; + else if (info.bpp == 4) width = (s->img_x + 1) >> 1; + else if (info.bpp == 8) width = s->img_x; + else { STBI_FREE(out); return stbi__errpuc("bad bpp", "Corrupt BMP"); } + pad = (-width)&3; + if (info.bpp == 1) { + for (j=0; j < (int) s->img_y; ++j) { + int bit_offset = 7, v = stbi__get8(s); + for (i=0; i < (int) s->img_x; ++i) { + int color = (v>>bit_offset)&0x1; + out[z++] = pal[color][0]; + out[z++] = pal[color][1]; + out[z++] = pal[color][2]; + if (target == 4) out[z++] = 255; + if (i+1 == (int) s->img_x) break; + if((--bit_offset) < 0) { + bit_offset = 7; + v = stbi__get8(s); + } + } + stbi__skip(s, pad); + } + } else { + for (j=0; j < (int) s->img_y; ++j) { + for (i=0; i < (int) s->img_x; i += 2) { + int v=stbi__get8(s),v2=0; + if (info.bpp == 4) { + v2 = v & 15; + v >>= 4; + } + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + if (i+1 == (int) s->img_x) break; + v = (info.bpp == 8) ? stbi__get8(s) : v2; + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + } + stbi__skip(s, pad); + } + } + } else { + int rshift=0,gshift=0,bshift=0,ashift=0,rcount=0,gcount=0,bcount=0,acount=0; + int z = 0; + int easy=0; + stbi__skip(s, info.offset - info.extra_read - info.hsz); + if (info.bpp == 24) width = 3 * s->img_x; + else if (info.bpp == 16) width = 2*s->img_x; + else /* bpp = 32 and pad = 0 */ width=0; + pad = (-width) & 3; + if (info.bpp == 24) { + easy = 1; + } else if (info.bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } + // right shift amt to put high bit in position #7 + rshift = stbi__high_bit(mr)-7; rcount = stbi__bitcount(mr); + gshift = stbi__high_bit(mg)-7; gcount = stbi__bitcount(mg); + bshift = stbi__high_bit(mb)-7; bcount = stbi__bitcount(mb); + ashift = stbi__high_bit(ma)-7; acount = stbi__bitcount(ma); + if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } + } + for (j=0; j < (int) s->img_y; ++j) { + if (easy) { + for (i=0; i < (int) s->img_x; ++i) { + unsigned char a; + out[z+2] = stbi__get8(s); + out[z+1] = stbi__get8(s); + out[z+0] = stbi__get8(s); + z += 3; + a = (easy == 2 ? stbi__get8(s) : 255); + all_a |= a; + if (target == 4) out[z++] = a; + } + } else { + int bpp = info.bpp; + for (i=0; i < (int) s->img_x; ++i) { + stbi__uint32 v = (bpp == 16 ? (stbi__uint32) stbi__get16le(s) : stbi__get32le(s)); + unsigned int a; + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); + a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); + all_a |= a; + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + + // if alpha channel is all 0s, replace with all 255s + if (target == 4 && all_a == 0) + for (i=4*s->img_x*s->img_y-1; i >= 0; i -= 4) + out[i] = 255; + + if (flip_vertically) { + stbi_uc t; + for (j=0; j < (int) s->img_y>>1; ++j) { + stbi_uc *p1 = out + j *s->img_x*target; + stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target; + for (i=0; i < (int) s->img_x*target; ++i) { + t = p1[i]; p1[i] = p2[i]; p2[i] = t; + } + } + } + + if (req_comp && req_comp != target) { + out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} +#endif + +// Targa Truevision - TGA +// by Jonathan Dummer +#ifndef STBI_NO_TGA +// returns STBI_rgb or whatever, 0 on error +static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int* is_rgb16) +{ + // only RGB or RGBA (incl. 16bit) or grey allowed + if (is_rgb16) *is_rgb16 = 0; + switch(bits_per_pixel) { + case 8: return STBI_grey; + case 16: if(is_grey) return STBI_grey_alpha; + // fallthrough + case 15: if(is_rgb16) *is_rgb16 = 1; + return STBI_rgb; + case 24: // fallthrough + case 32: return bits_per_pixel/8; + default: return 0; + } +} + +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) +{ + int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, tga_colormap_bpp; + int sz, tga_colormap_type; + stbi__get8(s); // discard Offset + tga_colormap_type = stbi__get8(s); // colormap type + if( tga_colormap_type > 1 ) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + tga_image_type = stbi__get8(s); // image type + if ( tga_colormap_type == 1 ) { // colormapped (paletted) image + if (tga_image_type != 1 && tga_image_type != 9) { + stbi__rewind(s); + return 0; + } + stbi__skip(s,4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) { + stbi__rewind(s); + return 0; + } + stbi__skip(s,4); // skip image x and y origin + tga_colormap_bpp = sz; + } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE + if ( (tga_image_type != 2) && (tga_image_type != 3) && (tga_image_type != 10) && (tga_image_type != 11) ) { + stbi__rewind(s); + return 0; // only RGB or grey allowed, +/- RLE + } + stbi__skip(s,9); // skip colormap specification and image x/y origin + tga_colormap_bpp = 0; + } + tga_w = stbi__get16le(s); + if( tga_w < 1 ) { + stbi__rewind(s); + return 0; // test width + } + tga_h = stbi__get16le(s); + if( tga_h < 1 ) { + stbi__rewind(s); + return 0; // test height + } + tga_bits_per_pixel = stbi__get8(s); // bits per pixel + stbi__get8(s); // ignore alpha bits + if (tga_colormap_bpp != 0) { + if((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { + // when using a colormap, tga_bits_per_pixel is the size of the indexes + // I don't think anything but 8 or 16bit indexes makes sense + stbi__rewind(s); + return 0; + } + tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); + } else { + tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), NULL); + } + if(!tga_comp) { + stbi__rewind(s); + return 0; + } + if (x) *x = tga_w; + if (y) *y = tga_h; + if (comp) *comp = tga_comp; + return 1; // seems to have passed everything +} + +static int stbi__tga_test(stbi__context *s) +{ + int res = 0; + int sz, tga_color_type; + stbi__get8(s); // discard Offset + tga_color_type = stbi__get8(s); // color type + if ( tga_color_type > 1 ) goto errorEnd; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if ( tga_color_type == 1 ) { // colormapped (paletted) image + if (sz != 1 && sz != 9) goto errorEnd; // colortype 1 demands image type 1 or 9 + stbi__skip(s,4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; + stbi__skip(s,4); // skip image x and y origin + } else { // "normal" image w/o colormap + if ( (sz != 2) && (sz != 3) && (sz != 10) && (sz != 11) ) goto errorEnd; // only RGB or grey allowed, +/- RLE + stbi__skip(s,9); // skip colormap specification and image x/y origin + } + if ( stbi__get16le(s) < 1 ) goto errorEnd; // test width + if ( stbi__get16le(s) < 1 ) goto errorEnd; // test height + sz = stbi__get8(s); // bits per pixel + if ( (tga_color_type == 1) && (sz != 8) && (sz != 16) ) goto errorEnd; // for colormapped images, bpp is size of an index + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; + + res = 1; // if we got this far, everything's good and we can return 1 instead of 0 + +errorEnd: + stbi__rewind(s); + return res; +} + +// read 16bit value and convert to 24bit RGB +static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc* out) +{ + stbi__uint16 px = (stbi__uint16)stbi__get16le(s); + stbi__uint16 fiveBitMask = 31; + // we have 3 channels with 5bits each + int r = (px >> 10) & fiveBitMask; + int g = (px >> 5) & fiveBitMask; + int b = px & fiveBitMask; + // Note that this saves the data in RGB(A) order, so it doesn't need to be swapped later + out[0] = (stbi_uc)((r * 255)/31); + out[1] = (stbi_uc)((g * 255)/31); + out[2] = (stbi_uc)((b * 255)/31); + + // some people claim that the most significant bit might be used for alpha + // (possibly if an alpha-bit is set in the "image descriptor byte") + // but that only made 16bit test images completely translucent.. + // so let's treat all 15 and 16bit TGAs as RGB with no alpha. +} + +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + // read in the TGA header stuff + int tga_offset = stbi__get8(s); + int tga_indexed = stbi__get8(s); + int tga_image_type = stbi__get8(s); + int tga_is_RLE = 0; + int tga_palette_start = stbi__get16le(s); + int tga_palette_len = stbi__get16le(s); + int tga_palette_bits = stbi__get8(s); + int tga_x_origin = stbi__get16le(s); + int tga_y_origin = stbi__get16le(s); + int tga_width = stbi__get16le(s); + int tga_height = stbi__get16le(s); + int tga_bits_per_pixel = stbi__get8(s); + int tga_comp, tga_rgb16=0; + int tga_inverted = stbi__get8(s); + // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused (useless?) + // image data + unsigned char *tga_data; + unsigned char *tga_palette = NULL; + int i, j; + unsigned char raw_data[4] = {0}; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + STBI_NOTUSED(ri); + STBI_NOTUSED(tga_x_origin); // @TODO + STBI_NOTUSED(tga_y_origin); // @TODO + + if (tga_height > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (tga_width > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + // do a tiny bit of precessing + if ( tga_image_type >= 8 ) + { + tga_image_type -= 8; + tga_is_RLE = 1; + } + tga_inverted = 1 - ((tga_inverted >> 5) & 1); + + // If I'm paletted, then I'll use the number of bits from the palette + if ( tga_indexed ) tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); + else tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), &tga_rgb16); + + if(!tga_comp) // shouldn't really happen, stbi__tga_test() should have ensured basic consistency + return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); + + // tga info + *x = tga_width; + *y = tga_height; + if (comp) *comp = tga_comp; + + if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) + return stbi__errpuc("too large", "Corrupt TGA"); + + tga_data = (unsigned char*)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); + if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); + + // skip to the data's starting position (offset usually = 0) + stbi__skip(s, tga_offset ); + + if ( !tga_indexed && !tga_is_RLE && !tga_rgb16 ) { + for (i=0; i < tga_height; ++i) { + int row = tga_inverted ? tga_height -i - 1 : i; + stbi_uc *tga_row = tga_data + row*tga_width*tga_comp; + stbi__getn(s, tga_row, tga_width * tga_comp); + } + } else { + // do I need to load a palette? + if ( tga_indexed) + { + if (tga_palette_len == 0) { /* you have to have at least one entry! */ + STBI_FREE(tga_data); + return stbi__errpuc("bad palette", "Corrupt TGA"); + } + + // any data to skip? (offset usually = 0) + stbi__skip(s, tga_palette_start ); + // load the palette + tga_palette = (unsigned char*)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); + if (!tga_palette) { + STBI_FREE(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); + } + if (tga_rgb16) { + stbi_uc *pal_entry = tga_palette; + STBI_ASSERT(tga_comp == STBI_rgb); + for (i=0; i < tga_palette_len; ++i) { + stbi__tga_read_rgb16(s, pal_entry); + pal_entry += tga_comp; + } + } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { + STBI_FREE(tga_data); + STBI_FREE(tga_palette); + return stbi__errpuc("bad palette", "Corrupt TGA"); + } + } + // load the data + for (i=0; i < tga_width * tga_height; ++i) + { + // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? + if ( tga_is_RLE ) + { + if ( RLE_count == 0 ) + { + // yep, get the next byte as a RLE command + int RLE_cmd = stbi__get8(s); + RLE_count = 1 + (RLE_cmd & 127); + RLE_repeating = RLE_cmd >> 7; + read_next_pixel = 1; + } else if ( !RLE_repeating ) + { + read_next_pixel = 1; + } + } else + { + read_next_pixel = 1; + } + // OK, if I need to read a pixel, do it now + if ( read_next_pixel ) + { + // load however much data we did have + if ( tga_indexed ) + { + // read in index, then perform the lookup + int pal_idx = (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); + if ( pal_idx >= tga_palette_len ) { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_comp; + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = tga_palette[pal_idx+j]; + } + } else if(tga_rgb16) { + STBI_ASSERT(tga_comp == STBI_rgb); + stbi__tga_read_rgb16(s, raw_data); + } else { + // read in the data raw + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = stbi__get8(s); + } + } + // clear the reading flag for the next pixel + read_next_pixel = 0; + } // end of reading a pixel + + // copy data + for (j = 0; j < tga_comp; ++j) + tga_data[i*tga_comp+j] = raw_data[j]; + + // in case we're in RLE mode, keep counting down + --RLE_count; + } + // do I need to invert the image? + if ( tga_inverted ) + { + for (j = 0; j*2 < tga_height; ++j) + { + int index1 = j * tga_width * tga_comp; + int index2 = (tga_height - 1 - j) * tga_width * tga_comp; + for (i = tga_width * tga_comp; i > 0; --i) + { + unsigned char temp = tga_data[index1]; + tga_data[index1] = tga_data[index2]; + tga_data[index2] = temp; + ++index1; + ++index2; + } + } + } + // clear my palette, if I had one + if ( tga_palette != NULL ) + { + STBI_FREE( tga_palette ); + } + } + + // swap RGB - if the source data was RGB16, it already is in the right order + if (tga_comp >= 3 && !tga_rgb16) + { + unsigned char* tga_pixel = tga_data; + for (i=0; i < tga_width * tga_height; ++i) + { + unsigned char temp = tga_pixel[0]; + tga_pixel[0] = tga_pixel[2]; + tga_pixel[2] = temp; + tga_pixel += tga_comp; + } + } + + // convert to target component count + if (req_comp && req_comp != tga_comp) + tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, tga_height); + + // the things I do to get rid of an error message, and yet keep + // Microsoft's C compilers happy... [8^( + tga_palette_start = tga_palette_len = tga_palette_bits = + tga_x_origin = tga_y_origin = 0; + STBI_NOTUSED(tga_palette_start); + // OK, done + return tga_data; +} +#endif + +// ************************************************************************************************* +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s) +{ + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; +} + +static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) +{ + int count, nleft, len; + + count = 0; + while ((nleft = pixelCount - count) > 0) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + if (len > nleft) return 0; // corrupt data + count += len; + while (len) { + *p = stbi__get8(s); + p += 4; + len--; + } + } else if (len > 128) { + stbi_uc val; + // Next -len+1 bytes in the dest are replicated from next source byte. + // (Interpret len as a negative 8-bit int.) + len = 257 - len; + if (len > nleft) return 0; // corrupt data + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } + + return 1; +} + +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) +{ + int pixelCount; + int channelCount, compression; + int channel, i; + int bitdepth; + int w,h; + stbi_uc *out; + STBI_NOTUSED(ri); + + // Check identifier + if (stbi__get32be(s) != 0x38425053) // "8BPS" + return stbi__errpuc("not PSD", "Corrupt PSD image"); + + // Check file type version. + if (stbi__get16be(s) != 1) + return stbi__errpuc("wrong version", "Unsupported version of PSD image"); + + // Skip 6 reserved bytes. + stbi__skip(s, 6 ); + + // Read the number of channels (R, G, B, A, etc). + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) + return stbi__errpuc("wrong channel count", "Unsupported number of channels in PSD image"); + + // Read the rows and columns of the image. + h = stbi__get32be(s); + w = stbi__get32be(s); + + if (h > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (w > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + // Make sure the depth is 8 bits. + bitdepth = stbi__get16be(s); + if (bitdepth != 8 && bitdepth != 16) + return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 or 16 bit"); + + // Make sure the color mode is RGB. + // Valid options are: + // 0: Bitmap + // 1: Grayscale + // 2: Indexed color + // 3: RGB color + // 4: CMYK color + // 7: Multichannel + // 8: Duotone + // 9: Lab color + if (stbi__get16be(s) != 3) + return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); + + // Skip the Mode Data. (It's the palette for indexed color; other info for other modes.) + stbi__skip(s,stbi__get32be(s) ); + + // Skip the image resources. (resolution, pen tool paths, etc) + stbi__skip(s, stbi__get32be(s) ); + + // Skip the reserved data. + stbi__skip(s, stbi__get32be(s) ); + + // Find out if the data is compressed. + // Known values: + // 0: no compression + // 1: RLE compressed + compression = stbi__get16be(s); + if (compression > 1) + return stbi__errpuc("bad compression", "PSD has an unknown compression format"); + + // Check size + if (!stbi__mad3sizes_valid(4, w, h, 0)) + return stbi__errpuc("too large", "Corrupt PSD"); + + // Create the destination image. + + if (!compression && bitdepth == 16 && bpc == 16) { + out = (stbi_uc *) stbi__malloc_mad3(8, w, h, 0); + ri->bits_per_channel = 16; + } else + out = (stbi_uc *) stbi__malloc(4 * w*h); + + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + pixelCount = w*h; + + // Initialize the data to zero. + //memset( out, 0, pixelCount * 4 ); + + // Finally, the image data. + if (compression) { + // RLE as used by .PSD and .TIFF + // Loop until you get the number of unpacked bytes you are expecting: + // Read the next source byte into n. + // If n is between 0 and 127 inclusive, copy the next n+1 bytes literally. + // Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times. + // Else if n is 128, noop. + // Endloop + + // The RLE-compressed data is preceded by a 2-byte data count for each row in the data, + // which we're going to just skip. + stbi__skip(s, h * channelCount * 2 ); + + // Read the RLE data by channel. + for (channel = 0; channel < 4; channel++) { + stbi_uc *p; + + p = out+channel; + if (channel >= channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++, p += 4) + *p = (channel == 3 ? 255 : 0); + } else { + // Read the RLE data. + if (!stbi__psd_decode_rle(s, p, pixelCount)) { + STBI_FREE(out); + return stbi__errpuc("corrupt", "bad RLE data"); + } + } + } + + } else { + // We're at the raw image data. It's each channel in order (Red, Green, Blue, Alpha, ...) + // where each channel consists of an 8-bit (or 16-bit) value for each pixel in the image. + + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + if (channel >= channelCount) { + // Fill this channel with default data. + if (bitdepth == 16 && bpc == 16) { + stbi__uint16 *q = ((stbi__uint16 *) out) + channel; + stbi__uint16 val = channel == 3 ? 65535 : 0; + for (i = 0; i < pixelCount; i++, q += 4) + *q = val; + } else { + stbi_uc *p = out+channel; + stbi_uc val = channel == 3 ? 255 : 0; + for (i = 0; i < pixelCount; i++, p += 4) + *p = val; + } + } else { + if (ri->bits_per_channel == 16) { // output bpc + stbi__uint16 *q = ((stbi__uint16 *) out) + channel; + for (i = 0; i < pixelCount; i++, q += 4) + *q = (stbi__uint16) stbi__get16be(s); + } else { + stbi_uc *p = out+channel; + if (bitdepth == 16) { // input bpc + for (i = 0; i < pixelCount; i++, p += 4) + *p = (stbi_uc) (stbi__get16be(s) >> 8); + } else { + for (i = 0; i < pixelCount; i++, p += 4) + *p = stbi__get8(s); + } + } + } + } + } + + // remove weird white matte from PSD + if (channelCount >= 4) { + if (ri->bits_per_channel == 16) { + for (i=0; i < w*h; ++i) { + stbi__uint16 *pixel = (stbi__uint16 *) out + 4*i; + if (pixel[3] != 0 && pixel[3] != 65535) { + float a = pixel[3] / 65535.0f; + float ra = 1.0f / a; + float inv_a = 65535.0f * (1 - ra); + pixel[0] = (stbi__uint16) (pixel[0]*ra + inv_a); + pixel[1] = (stbi__uint16) (pixel[1]*ra + inv_a); + pixel[2] = (stbi__uint16) (pixel[2]*ra + inv_a); + } + } + } else { + for (i=0; i < w*h; ++i) { + unsigned char *pixel = out + 4*i; + if (pixel[3] != 0 && pixel[3] != 255) { + float a = pixel[3] / 255.0f; + float ra = 1.0f / a; + float inv_a = 255.0f * (1 - ra); + pixel[0] = (unsigned char) (pixel[0]*ra + inv_a); + pixel[1] = (unsigned char) (pixel[1]*ra + inv_a); + pixel[2] = (unsigned char) (pixel[2]*ra + inv_a); + } + } + } + } + + // convert to desired output format + if (req_comp && req_comp != 4) { + if (ri->bits_per_channel == 16) + out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, 4, req_comp, w, h); + else + out = stbi__convert_format(out, 4, req_comp, w, h); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + if (comp) *comp = 4; + *y = h; + *x = w; + + return out; +} +#endif + +// ************************************************************************************************* +// Softimage PIC loader +// by Tom Seddon +// +// See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format +// See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/ + +#ifndef STBI_NO_PIC +static int stbi__pic_is4(stbi__context *s,const char *str) +{ + int i; + for (i=0; i<4; ++i) + if (stbi__get8(s) != (stbi_uc)str[i]) + return 0; + + return 1; +} + +static int stbi__pic_test_core(stbi__context *s) +{ + int i; + + if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) + return 0; + + for(i=0;i<84;++i) + stbi__get8(s); + + if (!stbi__pic_is4(s,"PICT")) + return 0; + + return 1; +} + +typedef struct +{ + stbi_uc size,type,channel; +} stbi__pic_packet; + +static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) +{ + int mask=0x80, i; + + for (i=0; i<4; ++i, mask>>=1) { + if (channel & mask) { + if (stbi__at_eof(s)) return stbi__errpuc("bad file","PIC file too short"); + dest[i]=stbi__get8(s); + } + } + + return dest; +} + +static void stbi__copyval(int channel,stbi_uc *dest,const stbi_uc *src) +{ + int mask=0x80,i; + + for (i=0;i<4; ++i, mask>>=1) + if (channel&mask) + dest[i]=src[i]; +} + +static stbi_uc *stbi__pic_load_core(stbi__context *s,int width,int height,int *comp, stbi_uc *result) +{ + int act_comp=0,num_packets=0,y,chained; + stbi__pic_packet packets[10]; + + // this will (should...) cater for even some bizarre stuff like having data + // for the same channel in multiple packets. + do { + stbi__pic_packet *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return stbi__errpuc("bad format","too many packets"); + + packet = &packets[num_packets++]; + + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + + act_comp |= packet->channel; + + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (reading packets)"); + if (packet->size != 8) return stbi__errpuc("bad format","packet isn't 8bpp"); + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? + + for(y=0; ytype) { + default: + return stbi__errpuc("bad format","packet has bad compression type"); + + case 0: {//uncompressed + int x; + + for(x=0;xchannel,dest)) + return 0; + break; + } + + case 1://Pure RLE + { + int left=width, i; + + while (left>0) { + stbi_uc count,value[4]; + + count=stbi__get8(s); + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pure read count)"); + + if (count > left) + count = (stbi_uc) left; + + if (!stbi__readval(s,packet->channel,value)) return 0; + + for(i=0; ichannel,dest,value); + left -= count; + } + } + break; + + case 2: {//Mixed RLE + int left=width; + while (left>0) { + int count = stbi__get8(s), i; + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (mixed read count)"); + + if (count >= 128) { // Repeated + stbi_uc value[4]; + + if (count==128) + count = stbi__get16be(s); + else + count -= 127; + if (count > left) + return stbi__errpuc("bad file","scanline overrun"); + + if (!stbi__readval(s,packet->channel,value)) + return 0; + + for(i=0;ichannel,dest,value); + } else { // Raw + ++count; + if (count>left) return stbi__errpuc("bad file","scanline overrun"); + + for(i=0;ichannel,dest)) + return 0; + } + left-=count; + } + break; + } + } + } + } + + return result; +} + +static void *stbi__pic_load(stbi__context *s,int *px,int *py,int *comp,int req_comp, stbi__result_info *ri) +{ + stbi_uc *result; + int i, x,y, internal_comp; + STBI_NOTUSED(ri); + + if (!comp) comp = &internal_comp; + + for (i=0; i<92; ++i) + stbi__get8(s); + + x = stbi__get16be(s); + y = stbi__get16be(s); + + if (y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pic header)"); + if (!stbi__mad3sizes_valid(x, y, 4, 0)) return stbi__errpuc("too large", "PIC image too large to decode"); + + stbi__get32be(s); //skip `ratio' + stbi__get16be(s); //skip `fields' + stbi__get16be(s); //skip `pad' + + // intermediate buffer is RGBA + result = (stbi_uc *) stbi__malloc_mad3(x, y, 4, 0); + if (!result) return stbi__errpuc("outofmem", "Out of memory"); + memset(result, 0xff, x*y*4); + + if (!stbi__pic_load_core(s,x,y,comp, result)) { + STBI_FREE(result); + result=0; + } + *px = x; + *py = y; + if (req_comp == 0) req_comp = *comp; + result=stbi__convert_format(result,4,req_comp,x,y); + + return result; +} + +static int stbi__pic_test(stbi__context *s) +{ + int r = stbi__pic_test_core(s); + stbi__rewind(s); + return r; +} +#endif + +// ************************************************************************************************* +// GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb + +#ifndef STBI_NO_GIF +typedef struct +{ + stbi__int16 prefix; + stbi_uc first; + stbi_uc suffix; +} stbi__gif_lzw; + +typedef struct +{ + int w,h; + stbi_uc *out; // output buffer (always 4 components) + stbi_uc *background; // The current "background" as far as a gif is concerned + stbi_uc *history; + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[8192]; + stbi_uc *color_table; + int parse, step; + int lflags; + int start_x, start_y; + int max_x, max_y; + int cur_x, cur_y; + int line_size; + int delay; +} stbi__gif; + +static int stbi__gif_test_raw(stbi__context *s) +{ + int sz; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') return 0; + sz = stbi__get8(s); + if (sz != '9' && sz != '7') return 0; + if (stbi__get8(s) != 'a') return 0; + return 1; +} + +static int stbi__gif_test(stbi__context *s) +{ + int r = stbi__gif_test_raw(s); + stbi__rewind(s); + return r; +} + +static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], int num_entries, int transp) +{ + int i; + for (i=0; i < num_entries; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + pal[i][3] = transp == i ? 0 : 255; + } +} + +static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, int is_info) +{ + stbi_uc version; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') + return stbi__err("not GIF", "Corrupt GIF"); + + version = stbi__get8(s); + if (version != '7' && version != '9') return stbi__err("not GIF", "Corrupt GIF"); + if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); + + stbi__g_failure_reason = ""; + g->w = stbi__get16le(s); + g->h = stbi__get16le(s); + g->flags = stbi__get8(s); + g->bgindex = stbi__get8(s); + g->ratio = stbi__get8(s); + g->transparent = -1; + + if (g->w > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (g->h > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + + if (comp != 0) *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the comments + + if (is_info) return 1; + + if (g->flags & 0x80) + stbi__gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1); + + return 1; +} + +static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__gif* g = (stbi__gif*) stbi__malloc(sizeof(stbi__gif)); + if (!g) return stbi__err("outofmem", "Out of memory"); + if (!stbi__gif_header(s, g, comp, 1)) { + STBI_FREE(g); + stbi__rewind( s ); + return 0; + } + if (x) *x = g->w; + if (y) *y = g->h; + STBI_FREE(g); + return 1; +} + +static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) +{ + stbi_uc *p, *c; + int idx; + + // recurse to decode the prefixes, since the linked-list is backwards, + // and working backwards through an interleaved image would be nasty + if (g->codes[code].prefix >= 0) + stbi__out_gif_code(g, g->codes[code].prefix); + + if (g->cur_y >= g->max_y) return; + + idx = g->cur_x + g->cur_y; + p = &g->out[idx]; + g->history[idx / 4] = 1; + + c = &g->color_table[g->codes[code].suffix * 4]; + if (c[3] > 128) { // don't render transparent pixels; + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } + g->cur_x += 4; + + if (g->cur_x >= g->max_x) { + g->cur_x = g->start_x; + g->cur_y += g->step; + + while (g->cur_y >= g->max_y && g->parse > 0) { + g->step = (1 << g->parse) * g->line_size; + g->cur_y = g->start_y + (g->step >> 1); + --g->parse; + } + } +} + +static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) +{ + stbi_uc lzw_cs; + stbi__int32 len, init_code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw *p; + + lzw_cs = stbi__get8(s); + if (lzw_cs > 12) return NULL; + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (init_code = 0; init_code < clear; init_code++) { + g->codes[init_code].prefix = -1; + g->codes[init_code].first = (stbi_uc) init_code; + g->codes[init_code].suffix = (stbi_uc) init_code; + } + + // support no starting clear code + avail = clear+2; + oldcode = -1; + + len = 0; + for(;;) { + if (valid_bits < codesize) { + if (len == 0) { + len = stbi__get8(s); // start new block + if (len == 0) + return g->out; + } + --len; + bits |= (stbi__int32) stbi__get8(s) << valid_bits; + valid_bits += 8; + } else { + stbi__int32 code = bits & codemask; + bits >>= codesize; + valid_bits -= codesize; + // @OPTIMIZE: is there some way we can accelerate the non-clear path? + if (code == clear) { // clear code + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + avail = clear + 2; + oldcode = -1; + first = 0; + } else if (code == clear + 1) { // end of stream code + stbi__skip(s, len); + while ((len = stbi__get8(s)) > 0) + stbi__skip(s,len); + return g->out; + } else if (code <= avail) { + if (first) { + return stbi__errpuc("no clear code", "Corrupt GIF"); + } + + if (oldcode >= 0) { + p = &g->codes[avail++]; + if (avail > 8192) { + return stbi__errpuc("too many codes", "Corrupt GIF"); + } + + p->prefix = (stbi__int16) oldcode; + p->first = g->codes[oldcode].first; + p->suffix = (code == avail) ? p->first : g->codes[code].first; + } else if (code == avail) + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + + stbi__out_gif_code(g, (stbi__uint16) code); + + if ((avail & codemask) == 0 && avail <= 0x0FFF) { + codesize++; + codemask = (1 << codesize) - 1; + } + + oldcode = code; + } else { + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + } + } + } +} + +// this function is designed to support animated gifs, although stb_image doesn't support it +// two back is the image from two frames ago, used for a very specific disposal format +static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, int req_comp, stbi_uc *two_back) +{ + int dispose; + int first_frame; + int pi; + int pcount; + STBI_NOTUSED(req_comp); + + // on first frame, any non-written pixels get the background colour (non-transparent) + first_frame = 0; + if (g->out == 0) { + if (!stbi__gif_header(s, g, comp,0)) return 0; // stbi__g_failure_reason set by stbi__gif_header + if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) + return stbi__errpuc("too large", "GIF image is too large"); + pcount = g->w * g->h; + g->out = (stbi_uc *) stbi__malloc(4 * pcount); + g->background = (stbi_uc *) stbi__malloc(4 * pcount); + g->history = (stbi_uc *) stbi__malloc(pcount); + if (!g->out || !g->background || !g->history) + return stbi__errpuc("outofmem", "Out of memory"); + + // image is treated as "transparent" at the start - ie, nothing overwrites the current background; + // background colour is only used for pixels that are not rendered first frame, after that "background" + // color refers to the color that was there the previous frame. + memset(g->out, 0x00, 4 * pcount); + memset(g->background, 0x00, 4 * pcount); // state of the background (starts transparent) + memset(g->history, 0x00, pcount); // pixels that were affected previous frame + first_frame = 1; + } else { + // second frame - how do we dispose of the previous one? + dispose = (g->eflags & 0x1C) >> 2; + pcount = g->w * g->h; + + if ((dispose == 3) && (two_back == 0)) { + dispose = 2; // if I don't have an image to revert back to, default to the old background + } + + if (dispose == 3) { // use previous graphic + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy( &g->out[pi * 4], &two_back[pi * 4], 4 ); + } + } + } else if (dispose == 2) { + // restore what was changed last frame to background before that frame; + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy( &g->out[pi * 4], &g->background[pi * 4], 4 ); + } + } + } else { + // This is a non-disposal case eithe way, so just + // leave the pixels as is, and they will become the new background + // 1: do not dispose + // 0: not specified. + } + + // background is what out is after the undoing of the previou frame; + memcpy( g->background, g->out, 4 * g->w * g->h ); + } + + // clear my history; + memset( g->history, 0x00, g->w * g->h ); // pixels that were affected previous frame + + for (;;) { + int tag = stbi__get8(s); + switch (tag) { + case 0x2C: /* Image Descriptor */ + { + stbi__int32 x, y, w, h; + stbi_uc *o; + + x = stbi__get16le(s); + y = stbi__get16le(s); + w = stbi__get16le(s); + h = stbi__get16le(s); + if (((x + w) > (g->w)) || ((y + h) > (g->h))) + return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); + + g->line_size = g->w * 4; + g->start_x = x * 4; + g->start_y = y * g->line_size; + g->max_x = g->start_x + w * 4; + g->max_y = g->start_y + h * g->line_size; + g->cur_x = g->start_x; + g->cur_y = g->start_y; + + // if the width of the specified rectangle is 0, that means + // we may not see *any* pixels or the image is malformed; + // to make sure this is caught, move the current y down to + // max_y (which is what out_gif_code checks). + if (w == 0) + g->cur_y = g->max_y; + + g->lflags = stbi__get8(s); + + if (g->lflags & 0x40) { + g->step = 8 * g->line_size; // first interlaced spacing + g->parse = 3; + } else { + g->step = g->line_size; + g->parse = 0; + } + + if (g->lflags & 0x80) { + stbi__gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1); + g->color_table = (stbi_uc *) g->lpal; + } else if (g->flags & 0x80) { + g->color_table = (stbi_uc *) g->pal; + } else + return stbi__errpuc("missing color table", "Corrupt GIF"); + + o = stbi__process_gif_raster(s, g); + if (!o) return NULL; + + // if this was the first frame, + pcount = g->w * g->h; + if (first_frame && (g->bgindex > 0)) { + // if first frame, any pixel not drawn to gets the background color + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi] == 0) { + g->pal[g->bgindex][3] = 255; // just in case it was made transparent, undo that; It will be reset next frame if need be; + memcpy( &g->out[pi * 4], &g->pal[g->bgindex], 4 ); + } + } + } + + return o; + } + + case 0x21: // Comment Extension. + { + int len; + int ext = stbi__get8(s); + if (ext == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + g->delay = 10 * stbi__get16le(s); // delay - 1/100th of a second, saving as 1/1000ths. + + // unset old transparent + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 255; + } + if (g->eflags & 0x01) { + g->transparent = stbi__get8(s); + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 0; + } + } else { + // don't need transparent + stbi__skip(s, 1); + g->transparent = -1; + } + } else { + stbi__skip(s, len); + break; + } + } + while ((len = stbi__get8(s)) != 0) { + stbi__skip(s, len); + } + break; + } + + case 0x3B: // gif stream termination code + return (stbi_uc *) s; // using '1' causes warning on some compilers + + default: + return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } +} + +static void *stbi__load_gif_main_outofmem(stbi__gif *g, stbi_uc *out, int **delays) +{ + STBI_FREE(g->out); + STBI_FREE(g->history); + STBI_FREE(g->background); + + if (out) STBI_FREE(out); + if (delays && *delays) STBI_FREE(*delays); + return stbi__errpuc("outofmem", "Out of memory"); +} + +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp) +{ + if (stbi__gif_test(s)) { + int layers = 0; + stbi_uc *u = 0; + stbi_uc *out = 0; + stbi_uc *two_back = 0; + stbi__gif g; + int stride; + int out_size = 0; + int delays_size = 0; + + STBI_NOTUSED(out_size); + STBI_NOTUSED(delays_size); + + memset(&g, 0, sizeof(g)); + if (delays) { + *delays = 0; + } + + do { + u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); + if (u == (stbi_uc *) s) u = 0; // end of animated gif marker + + if (u) { + *x = g.w; + *y = g.h; + ++layers; + stride = g.w * g.h * 4; + + if (out) { + void *tmp = (stbi_uc*) STBI_REALLOC_SIZED( out, out_size, layers * stride ); + if (!tmp) + return stbi__load_gif_main_outofmem(&g, out, delays); + else { + out = (stbi_uc*) tmp; + out_size = layers * stride; + } + + if (delays) { + int *new_delays = (int*) STBI_REALLOC_SIZED( *delays, delays_size, sizeof(int) * layers ); + if (!new_delays) + return stbi__load_gif_main_outofmem(&g, out, delays); + *delays = new_delays; + delays_size = layers * sizeof(int); + } + } else { + out = (stbi_uc*)stbi__malloc( layers * stride ); + if (!out) + return stbi__load_gif_main_outofmem(&g, out, delays); + out_size = layers * stride; + if (delays) { + *delays = (int*) stbi__malloc( layers * sizeof(int) ); + if (!*delays) + return stbi__load_gif_main_outofmem(&g, out, delays); + delays_size = layers * sizeof(int); + } + } + memcpy( out + ((layers - 1) * stride), u, stride ); + if (layers >= 2) { + two_back = out - 2 * stride; + } + + if (delays) { + (*delays)[layers - 1U] = g.delay; + } + } + } while (u != 0); + + // free temp buffer; + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + + // do the final conversion after loading everything; + if (req_comp && req_comp != 4) + out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); + + *z = layers; + return out; + } else { + return stbi__errpuc("not GIF", "Image was not as a gif type."); + } +} + +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + STBI_NOTUSED(ri); + + u = stbi__gif_load_next(s, &g, comp, req_comp, 0); + if (u == (stbi_uc *) s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + + // moved conversion to after successful load so that the same + // can be done for multiple frames. + if (req_comp && req_comp != 4) + u = stbi__convert_format(u, 4, req_comp, g.w, g.h); + } else if (g.out) { + // if there was an error and we allocated an image buffer, free it! + STBI_FREE(g.out); + } + + // free buffers needed for multiple frame loading; + STBI_FREE(g.history); + STBI_FREE(g.background); + + return u; +} + +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) +{ + return stbi__gif_info_raw(s,x,y,comp); +} +#endif + +// ************************************************************************************************* +// Radiance RGBE HDR loader +// originally by Nicolas Schulz +#ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context *s, const char *signature) +{ + int i; + for (i=0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) + return 0; + stbi__rewind(s); + return 1; +} + +static int stbi__hdr_test(stbi__context* s) +{ + int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); + stbi__rewind(s); + if(!r) { + r = stbi__hdr_test_core(s, "#?RGBE\n"); + stbi__rewind(s); + } + return r; +} + +#define STBI__HDR_BUFLEN 1024 +static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) +{ + int len=0; + char c = '\0'; + + c = (char) stbi__get8(z); + + while (!stbi__at_eof(z) && c != '\n') { + buffer[len++] = c; + if (len == STBI__HDR_BUFLEN-1) { + // flush to end of line + while (!stbi__at_eof(z) && stbi__get8(z) != '\n') + ; + break; + } + c = (char) stbi__get8(z); + } + + buffer[len] = 0; + return buffer; +} + +static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) +{ + if ( input[3] != 0 ) { + float f1; + // Exponent + f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8)); + if (req_comp <= 2) + output[0] = (input[0] + input[1] + input[2]) * f1 / 3; + else { + output[0] = input[0] * f1; + output[1] = input[1] * f1; + output[2] = input[2] * f1; + } + if (req_comp == 2) output[1] = 1; + if (req_comp == 4) output[3] = 1; + } else { + switch (req_comp) { + case 4: output[3] = 1; /* fallthrough */ + case 3: output[0] = output[1] = output[2] = 0; + break; + case 2: output[1] = 1; /* fallthrough */ + case 1: output[0] = 0; + break; + } + } +} + +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int width, height; + stbi_uc *scanline; + float *hdr_data; + int len; + unsigned char count, value; + int i, j, k, c1,c2, z; + const char *headerToken; + STBI_NOTUSED(ri); + + // Check identifier + headerToken = stbi__hdr_gettoken(s,buffer); + if (strcmp(headerToken, "#?RADIANCE") != 0 && strcmp(headerToken, "#?RGBE") != 0) + return stbi__errpf("not HDR", "Corrupt HDR image"); + + // Parse header + for(;;) { + token = stbi__hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) return stbi__errpf("unsupported format", "Unsupported HDR format"); + + // Parse width and height + // can't use sscanf() if we're not using stdio! + token = stbi__hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + height = (int) strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + width = (int) strtol(token, NULL, 10); + + if (height > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); + if (width > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); + + *x = width; + *y = height; + + if (comp) *comp = 3; + if (req_comp == 0) req_comp = 3; + + if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) + return stbi__errpf("too large", "HDR image is too large"); + + // Read data + hdr_data = (float *) stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); + if (!hdr_data) + return stbi__errpf("outofmem", "Out of memory"); + + // Load image data + // image data is stored as some number of sca + if ( width < 8 || width >= 32768) { + // Read flat data + for (j=0; j < height; ++j) { + for (i=0; i < width; ++i) { + stbi_uc rgbe[4]; + main_decode_loop: + stbi__getn(s, rgbe, 4); + stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp); + } + } + } else { + // Read RLE-encoded data + scanline = NULL; + + for (j = 0; j < height; ++j) { + c1 = stbi__get8(s); + c2 = stbi__get8(s); + len = stbi__get8(s); + if (c1 != 2 || c2 != 2 || (len & 0x80)) { + // not run-length encoded, so we have to actually use THIS data as a decoded + // pixel (note this can't be a valid pixel--one of RGB must be >= 128) + stbi_uc rgbe[4]; + rgbe[0] = (stbi_uc) c1; + rgbe[1] = (stbi_uc) c2; + rgbe[2] = (stbi_uc) len; + rgbe[3] = (stbi_uc) stbi__get8(s); + stbi__hdr_convert(hdr_data, rgbe, req_comp); + i = 1; + j = 0; + STBI_FREE(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } + if (scanline == NULL) { + scanline = (stbi_uc *) stbi__malloc_mad2(width, 4, 0); + if (!scanline) { + STBI_FREE(hdr_data); + return stbi__errpf("outofmem", "Out of memory"); + } + } + + for (k = 0; k < 4; ++k) { + int nleft; + i = 0; + while ((nleft = width - i) > 0) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + if ((count == 0) || (count > nleft)) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } else { + // Dump + if ((count == 0) || (count > nleft)) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = stbi__get8(s); + } + } + } + for (i=0; i < width; ++i) + stbi__hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp); + } + if (scanline) + STBI_FREE(scanline); + } + + return hdr_data; +} + +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) +{ + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int dummy; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (stbi__hdr_test(s) == 0) { + stbi__rewind( s ); + return 0; + } + + for(;;) { + token = stbi__hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) { + stbi__rewind( s ); + return 0; + } + token = stbi__hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) { + stbi__rewind( s ); + return 0; + } + token += 3; + *y = (int) strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) { + stbi__rewind( s ); + return 0; + } + token += 3; + *x = (int) strtol(token, NULL, 10); + *comp = 3; + return 1; +} +#endif // STBI_NO_HDR + +#ifndef STBI_NO_BMP +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) +{ + void *p; + stbi__bmp_data info; + + info.all_a = 255; + p = stbi__bmp_parse_header(s, &info); + if (p == NULL) { + stbi__rewind( s ); + return 0; + } + if (x) *x = s->img_x; + if (y) *y = s->img_y; + if (comp) { + if (info.bpp == 24 && info.ma == 0xff000000) + *comp = 3; + else + *comp = info.ma ? 4 : 3; + } + return 1; +} +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) +{ + int channelCount, dummy, depth; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind( s ); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind( s ); + return 0; + } + *y = stbi__get32be(s); + *x = stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 8 && depth != 16) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind( s ); + return 0; + } + *comp = 4; + return 1; +} + +static int stbi__psd_is16(stbi__context *s) +{ + int channelCount, depth; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind( s ); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind( s ); + return 0; + } + STBI_NOTUSED(stbi__get32be(s)); + STBI_NOTUSED(stbi__get32be(s)); + depth = stbi__get16be(s); + if (depth != 16) { + stbi__rewind( s ); + return 0; + } + return 1; +} +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) +{ + int act_comp=0,num_packets=0,chained,dummy; + stbi__pic_packet packets[10]; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) { + stbi__rewind(s); + return 0; + } + + stbi__skip(s, 88); + + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) { + stbi__rewind( s); + return 0; + } + if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) { + stbi__rewind( s ); + return 0; + } + + stbi__skip(s, 8); + + do { + stbi__pic_packet *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return 0; + + packet = &packets[num_packets++]; + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + act_comp |= packet->channel; + + if (stbi__at_eof(s)) { + stbi__rewind( s ); + return 0; + } + if (packet->size != 8) { + stbi__rewind( s ); + return 0; + } + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); + + return 1; +} +#endif + +// ************************************************************************************************* +// Portable Gray Map and Portable Pixel Map loader +// by Ken Miller +// +// PGM: http://netpbm.sourceforge.net/doc/pgm.html +// PPM: http://netpbm.sourceforge.net/doc/ppm.html +// +// Known limitations: +// Does not support comments in the header section +// Does not support ASCII image data (formats P2 and P3) + +#ifndef STBI_NO_PNM + +static int stbi__pnm_test(stbi__context *s) +{ + char p, t; + p = (char) stbi__get8(s); + t = (char) stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind( s ); + return 0; + } + return 1; +} + +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *out; + STBI_NOTUSED(ri); + + ri->bits_per_channel = stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n); + if (ri->bits_per_channel == 0) + return 0; + + if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + + if (!stbi__mad4sizes_valid(s->img_n, s->img_x, s->img_y, ri->bits_per_channel / 8, 0)) + return stbi__errpuc("too large", "PNM too large"); + + out = (stbi_uc *) stbi__malloc_mad4(s->img_n, s->img_x, s->img_y, ri->bits_per_channel / 8, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (!stbi__getn(s, out, s->img_n * s->img_x * s->img_y * (ri->bits_per_channel / 8))) { + STBI_FREE(out); + return stbi__errpuc("bad PNM", "PNM file truncated"); + } + + if (req_comp && req_comp != s->img_n) { + if (ri->bits_per_channel == 16) { + out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, s->img_n, req_comp, s->img_x, s->img_y); + } else { + out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); + } + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + return out; +} + +static int stbi__pnm_isspace(char c) +{ + return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r'; +} + +static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) +{ + for (;;) { + while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) + *c = (char) stbi__get8(s); + + if (stbi__at_eof(s) || *c != '#') + break; + + while (!stbi__at_eof(s) && *c != '\n' && *c != '\r' ) + *c = (char) stbi__get8(s); + } +} + +static int stbi__pnm_isdigit(char c) +{ + return c >= '0' && c <= '9'; +} + +static int stbi__pnm_getinteger(stbi__context *s, char *c) +{ + int value = 0; + + while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { + value = value*10 + (*c - '0'); + *c = (char) stbi__get8(s); + if((value > 214748364) || (value == 214748364 && *c > '7')) + return stbi__err("integer parse overflow", "Parsing an integer in the PPM header overflowed a 32-bit int"); + } + + return value; +} + +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) +{ + int maxv, dummy; + char c, p, t; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + stbi__rewind(s); + + // Get identifier + p = (char) stbi__get8(s); + t = (char) stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + + *comp = (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm + + c = (char) stbi__get8(s); + stbi__pnm_skip_whitespace(s, &c); + + *x = stbi__pnm_getinteger(s, &c); // read width + if(*x == 0) + return stbi__err("invalid width", "PPM image header had zero or overflowing width"); + stbi__pnm_skip_whitespace(s, &c); + + *y = stbi__pnm_getinteger(s, &c); // read height + if (*y == 0) + return stbi__err("invalid width", "PPM image header had zero or overflowing width"); + stbi__pnm_skip_whitespace(s, &c); + + maxv = stbi__pnm_getinteger(s, &c); // read max value + if (maxv > 65535) + return stbi__err("max value > 65535", "PPM image supports only 8-bit and 16-bit images"); + else if (maxv > 255) + return 16; + else + return 8; +} + +static int stbi__pnm_is16(stbi__context *s) +{ + if (stbi__pnm_info(s, NULL, NULL, NULL) == 16) + return 1; + return 0; +} +#endif + +static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) +{ + #ifndef STBI_NO_JPEG + if (stbi__jpeg_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PNG + if (stbi__png_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_GIF + if (stbi__gif_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_BMP + if (stbi__bmp_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PSD + if (stbi__psd_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PIC + if (stbi__pic_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PNM + if (stbi__pnm_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) return 1; + #endif + + // test tga last because it's a crappy test! + #ifndef STBI_NO_TGA + if (stbi__tga_info(s, x, y, comp)) + return 1; + #endif + return stbi__err("unknown image type", "Image not of any known type, or corrupt"); +} + +static int stbi__is_16_main(stbi__context *s) +{ + #ifndef STBI_NO_PNG + if (stbi__png_is16(s)) return 1; + #endif + + #ifndef STBI_NO_PSD + if (stbi__psd_is16(s)) return 1; + #endif + + #ifndef STBI_NO_PNM + if (stbi__pnm_is16(s)) return 1; + #endif + return 0; +} + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_info_from_file(f, x, y, comp); + fclose(f); + return result; +} + +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__info_main(&s,x,y,comp); + fseek(f,pos,SEEK_SET); + return r; +} + +STBIDEF int stbi_is_16_bit(char const *filename) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_is_16_bit_from_file(f); + fclose(f); + return result; +} + +STBIDEF int stbi_is_16_bit_from_file(FILE *f) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__is_16_main(&s); + fseek(f,pos,SEEK_SET); + return r; +} +#endif // !STBI_NO_STDIO + +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__info_main(&s,x,y,comp); +} + +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); + return stbi__info_main(&s,x,y,comp); +} + +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__is_16_main(&s); +} + +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); + return stbi__is_16_main(&s); +} + +#endif // STB_IMAGE_IMPLEMENTATION + +/* + revision history: + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug + 1-bit BMP + *_is_16_bit api + avoid warnings + 2.16 (2017-07-23) all functions have 16-bit variants; + STBI_NO_STDIO works again; + compilation fixes; + fix rounding in unpremultiply; + optimize vertical flip; + disable raw_len validation; + documentation fixes + 2.15 (2017-03-18) fix png-1,2,4 bug; now all Imagenet JPGs decode; + warning fixes; disable run-time SSE detection on gcc; + uniform handling of optional "return" values; + thread-safe initialization of zlib tables + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) allocate large structures on the stack + remove white matting for transparent PSD + fix reported channel count for PNG & BMP + re-enable SSE2 in non-gcc 64-bit + support RGB-formatted JPEG + read 16-bit PNGs (only as 8-bit) + 2.10 (2016-01-22) avoid warning introduced in 2.09 by STBI_REALLOC_SIZED + 2.09 (2016-01-16) allow comments in PNM files + 16-bit-per-pixel TGA (not bit-per-component) + info() for TGA could break due to .hdr handling + info() for BMP to shares code instead of sloppy parse + can use STBI_REALLOC_SIZED if allocator doesn't support realloc + code cleanup + 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD as RGBA + 2.07 (2015-09-13) fix compiler warnings + partial animated GIF support + limited 16-bpc PSD support + #ifdef unused functions + bug with < 92 byte PIC,PNM,HDR,TGA + 2.06 (2015-04-19) fix bug where PSD returns wrong '*comp' value + 2.05 (2015-04-19) fix bug in progressive JPEG handling, fix warning + 2.04 (2015-04-15) try to re-enable SIMD on MinGW 64-bit + 2.03 (2015-04-12) extra corruption checking (mmozeiko) + stbi_set_flip_vertically_on_load (nguillemot) + fix NEON support; fix mingw support + 2.02 (2015-01-19) fix incorrect assert, fix warning + 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit without -msse2 + 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG + 2.00 (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) + progressive JPEG (stb) + PGM/PPM support (Ken Miller) + STBI_MALLOC,STBI_REALLOC,STBI_FREE + GIF bugfix -- seemingly never worked + STBI_NO_*, STBI_ONLY_* + 1.48 (2014-12-14) fix incorrectly-named assert() + 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar Cornut & stb) + optimize PNG (ryg) + fix bug in interlaced PNG with user-specified channel count (stb) + 1.46 (2014-08-26) + fix broken tRNS chunk (colorkey-style transparency) in non-paletted PNG + 1.45 (2014-08-16) + fix MSVC-ARM internal compiler error by wrapping malloc + 1.44 (2014-08-07) + various warning fixes from Ronny Chevalier + 1.43 (2014-07-15) + fix MSVC-only compiler problem in code changed in 1.42 + 1.42 (2014-07-09) + don't define _CRT_SECURE_NO_WARNINGS (affects user code) + fixes to stbi__cleanup_jpeg path + added STBI_ASSERT to avoid requiring assert.h + 1.41 (2014-06-25) + fix search&replace from 1.36 that messed up comments/error messages + 1.40 (2014-06-22) + fix gcc struct-initialization warning + 1.39 (2014-06-15) + fix to TGA optimization when req_comp != number of components in TGA; + fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my test suite) + add support for BMP version 5 (more ignored fields) + 1.38 (2014-06-06) + suppress MSVC warnings on integer casts truncating values + fix accidental rename of 'skip' field of I/O + 1.37 (2014-06-04) + remove duplicate typedef + 1.36 (2014-06-03) + convert to header file single-file library + if de-iphone isn't set, load iphone images color-swapped instead of returning NULL + 1.35 (2014-05-27) + various warnings + fix broken STBI_SIMD path + fix bug where stbi_load_from_file no longer left file pointer in correct place + fix broken non-easy path for 32-bit BMP (possibly never used) + TGA optimization by Arseny Kapoulkine + 1.34 (unknown) + use STBI_NOTUSED in stbi__resample_row_generic(), fix one more leak in tga failure case + 1.33 (2011-07-14) + make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements + 1.32 (2011-07-13) + support for "info" function for all supported filetypes (SpartanJ) + 1.31 (2011-06-20) + a few more leak fixes, bug in PNG handling (SpartanJ) + 1.30 (2011-06-11) + added ability to load files via callbacks to accomidate custom input streams (Ben Wenger) + removed deprecated format-specific test/load functions + removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway + error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha) + fix inefficiency in decoding 32-bit BMP (David Woo) + 1.29 (2010-08-16) + various warning fixes from Aurelien Pocheville + 1.28 (2010-08-01) + fix bug in GIF palette transparency (SpartanJ) + 1.27 (2010-08-01) + cast-to-stbi_uc to fix warnings + 1.26 (2010-07-24) + fix bug in file buffering for PNG reported by SpartanJ + 1.25 (2010-07-17) + refix trans_data warning (Won Chun) + 1.24 (2010-07-12) + perf improvements reading from files on platforms with lock-heavy fgetc() + minor perf improvements for jpeg + deprecated type-specific functions so we'll get feedback if they're needed + attempt to fix trans_data warning (Won Chun) + 1.23 fixed bug in iPhone support + 1.22 (2010-07-10) + removed image *writing* support + stbi_info support from Jetro Lauha + GIF support from Jean-Marc Lienher + iPhone PNG-extensions from James Brown + warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. Janez (U+017D)emva) + 1.21 fix use of 'stbi_uc' in header (reported by jon blow) + 1.20 added support for Softimage PIC, by Tom Seddon + 1.19 bug in interlaced PNG corruption check (found by ryg) + 1.18 (2008-08-02) + fix a threading bug (local mutable static) + 1.17 support interlaced PNG + 1.16 major bugfix - stbi__convert_format converted one too many pixels + 1.15 initialize some fields for thread safety + 1.14 fix threadsafe conversion bug + header-file-only version (#define STBI_HEADER_FILE_ONLY before including) + 1.13 threadsafe + 1.12 const qualifiers in the API + 1.11 Support installable IDCT, colorspace conversion routines + 1.10 Fixes for 64-bit (don't use "unsigned long") + optimized upsampling by Fabian "ryg" Giesen + 1.09 Fix format-conversion for PSD code (bad global variables!) + 1.08 Thatcher Ulrich's PSD code integrated by Nicolas Schulz + 1.07 attempt to fix C++ warning/errors again + 1.06 attempt to fix C++ warning/errors again + 1.05 fix TGA loading to return correct *comp and use good luminance calc + 1.04 default float alpha is 1, not 255; use 'void *' for stbi_image_free + 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR + 1.02 support for (subset of) HDR files, float interface for preferred access to them + 1.01 fix bug: possible bug in handling right-side up bmps... not sure + fix bug: the stbi__bmp_load() and stbi__tga_load() functions didn't work at all + 1.00 interface to zlib that skips zlib header + 0.99 correct handling of alpha in palette + 0.98 TGA loader by lonesock; dynamically add loaders (untested) + 0.97 jpeg errors on too large a file; also catch another malloc failure + 0.96 fix detection of invalid v value - particleman@mollyrocket forum + 0.95 during header scan, seek to markers in case of padding + 0.94 STBI_NO_STDIO to disable stdio usage; rename all #defines the same + 0.93 handle jpegtran output; verbose errors + 0.92 read 4,8,16,24,32-bit BMP files of several formats + 0.91 output 24-bit Windows 3.0 BMP files + 0.90 fix a few more warnings; bump version number to approach 1.0 + 0.61 bugfixes due to Marc LeBlanc, Christopher Lloyd + 0.60 fix compiling as c++ + 0.59 fix warnings: merge Dave Moore's -Wall fixes + 0.58 fix bug: zlib uncompressed mode len/nlen was wrong endian + 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available + 0.56 fix bug: zlib uncompressed mode len vs. nlen + 0.55 fix bug: restart_interval not initialized to 0 + 0.54 allow NULL for 'int *comp' + 0.53 fix bug in png 3->4; speedup png decoding + 0.52 png handles req_comp=3,4 directly; minor cleanup; jpeg comments + 0.51 obey req_comp requests, 1-component jpegs return as 1-component, + on 'test' only check type, not whether we support this variant + 0.50 (2006-11-19) + first released version +*/ + + +/* +------------------------------------------------------------------------------ +This software is available under 2 licenses -- choose whichever you prefer. +------------------------------------------------------------------------------ +ALTERNATIVE A - MIT License +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +------------------------------------------------------------------------------ +ALTERNATIVE B - Public Domain (www.unlicense.org) +This is free and unencumbered software released into the public domain. +Anyone is free to copy, modify, publish, use, compile, sell, or distribute this +software, either in source code form or as a compiled binary, for any purpose, +commercial or non-commercial, and by any means. +In jurisdictions that recognize copyright laws, the author or authors of this +software dedicate any and all copyright interest in the software to the public +domain. We make this dedication for the benefit of the public at large and to +the detriment of our heirs and successors. We intend this dedication to be an +overt act of relinquishment in perpetuity of all present and future rights to +this software under copyright law. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN +ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +------------------------------------------------------------------------------ +*/ diff --git a/trig.c b/trig.c new file mode 100644 index 0000000..e0ecad9 --- /dev/null +++ b/trig.c @@ -0,0 +1,73 @@ +#include "allocator.h" +#include "kitty.h" +#include "matrix.h" +#include "object.h" +#include "types.h" + +#include + +struct Trig { + struct Kitty *kitty; + struct mat3x3 model; +}; + +#define TRIG_VERTEX_COUNT 6 +struct Vertex trig_vertices[TRIG_VERTEX_COUNT] = { + (struct Vertex){(struct Vec2){.x = 0.0f, .y = 0.0f}, + (struct Vec3){1.0f, 0.0f, 1.0f}}, + (struct Vertex){(struct Vec2){.x = 1.0f, .y = 0.0f}, + (struct Vec3){1.0f, 0.0f, 0.2f}}, + (struct Vertex){(struct Vec2){.x = 1.0f, .y = 1.0f}, + (struct Vec3){0.2f, 0.0f, 1.0f}}, + (struct Vertex){(struct Vec2){.x = 0.0f, .y = 0.0f}, + (struct Vec3){1.0f, 0.0f, 1.0f}}, + (struct Vertex){(struct Vec2){.x = 1.0f, .y = 1.0f}, + (struct Vec3){1.0f, 0.0f, 0.2f}}, + (struct Vertex){(struct Vec2){.x = 0.0f, .y = 1.0f}, + (struct Vec3){0.2f, 0.0f, 1.0f}}, +}; +struct TrigUBO { + struct mat3x3 model; + struct mat3x3 view; + struct mat3x3 proj; +}; + +struct Trig *trig_make_args(struct Scene *scene) { + struct Trig *trig = mem_malloc(scene->mem, sizeof(struct Trig)); + + trig->kitty = kitty_make(); + kitty_set_vertex_shader(trig->kitty, "./Shaders/vert.spv"); + kitty_set_fragment_shader(trig->kitty, "./Shaders/frag.spv"); + kitty_set_vertex_buffer(trig->kitty, trig_vertices, TRIG_VERTEX_COUNT, + sizeof(struct Vertex)); + kitty_add_vertex_buffer_format(trig->kitty, VK_FORMAT_R32G32_SFLOAT); + kitty_add_vertex_buffer_format(trig->kitty, VK_FORMAT_R32G32B32_SFLOAT); + kitty_attatch_ubo(trig->kitty, sizeof(struct TrigUBO)); + kitty_attatch_image(trig->kitty, "./image.png"); + kitty_finalise(scene->vk, trig->kitty); + + return trig; +} + +struct Trig *make_trig(struct Scene *scene, char *_data, int _len) { + return trig_make_args(scene); +} + +void free_trig(struct Scene *scene, struct Trig *trig) { + free_kitty(scene->vk, trig->kitty); + mem_free(scene->mem, trig); +} + +void trig_tick(struct Scene *scene, struct Trig *trig) { + struct TrigUBO ubo = {0}; + ubo.model = multiply3x3( + translate3x3( + (struct Vec2){sinf((double)(scene->msecs) * 0.000001f), 0.f}), + multiply3x3(rotate3x3(PI / 2.f * (double)(scene->msecs) * 0.000001f), + scale3x3((struct Vec2){0.5f, 0.5f}))); + ubo.view = IDENT3x3; + ubo.proj = scale3x3( + (struct Vec2){1.f, (float)scene->vk->width / (float)scene->vk->heigh}); + + kitty_set_next_ubo(scene->vk, trig->kitty, 0, &ubo); +} diff --git a/types.c b/types.c new file mode 100644 index 0000000..8e71fd1 --- /dev/null +++ b/types.c @@ -0,0 +1 @@ +#include "types.h" diff --git a/types.h b/types.h new file mode 100644 index 0000000..da30a67 --- /dev/null +++ b/types.h @@ -0,0 +1,44 @@ +#ifndef INCLUDE_WAYLANDCLIENT_TYPES_H_ +#define INCLUDE_WAYLANDCLIENT_TYPES_H_ + +#define PI 3.141592653589793238462643f + +// struct vec2 { +// union { +// struct { +// float x, y; +// }; +// float arr[2]; +// }; +// }; +struct Vec2 { + float x, y; +}; +struct IVec2 { + int x, y; +}; + +// struct vec3 { +// union { +// struct { +// float x, y, z; +// }; +// struct { +// float r, g, b; +// }; +// float arr[3]; +// }; +// }; +struct Vec3 { + float x, y, z; +}; +struct Vec4 { + float x, y, z, w; +}; + +struct Vertex { + struct Vec2 pos; + struct Vec3 col; +}; + +#endif // INCLUDE_WAYLANDCLIENT_TYPES_H_ diff --git a/vulkan.c b/vulkan.c new file mode 100644 index 0000000..e0fd713 --- /dev/null +++ b/vulkan.c @@ -0,0 +1,955 @@ +#include "vulkan.h" +#include "kitty.h" +#include "log.h" +#include "matrix.h" +#include "types.h" + +#include "image.h" + +#include +#include +#include +#include +#include + +#define VERTEX_ATTRIBUTE_COUNT 2 + +#define QUAD_VERTEX_COUNT 3 +struct Vertex quad_vertices[QUAD_VERTEX_COUNT] = { + (struct Vertex){(struct Vec2){.x = 0.0f, .y = -0.5f}, + (struct Vec3){1.0f, 0.0f, 1.0f}}, + (struct Vertex){(struct Vec2){.x = 0.5f, .y = 0.5f}, + (struct Vec3){1.0f, 0.0f, 0.2f}}, + (struct Vertex){(struct Vec2){.x = -0.5f, .y = 0.5f}, + (struct Vec3){0.2f, 0.0f, 1.0f}}, +}; + +struct UBO { + struct mat3x3 model; + struct mat3x3 view; + struct mat3x3 proj; +}; + +VkVertexInputBindingDescription get_vertex_binding_description() { + VkVertexInputBindingDescription description = {0}; + description.binding = 0; + description.stride = sizeof(struct Vertex); + description.inputRate = VK_VERTEX_INPUT_RATE_VERTEX; + return description; +} + +// return value owned by caller +VkVertexInputAttributeDescription *get_vertex_attribute_description() { + VkVertexInputAttributeDescription *description = malloc( + sizeof(VkVertexInputAttributeDescription) * VERTEX_ATTRIBUTE_COUNT); + description[0].binding = 0; + description[0].location = 0; + description[0].format = VK_FORMAT_R32G32_SFLOAT; + description[0].offset = offsetof(struct Vertex, pos); + + description[1].binding = 0; + description[1].location = 1; + description[1].format = VK_FORMAT_R32G32B32_SFLOAT; + description[1].offset = offsetof(struct Vertex, col); + return description; +} + +static const char *const extensions[] = { + "VK_EXT_debug_utils", "VK_KHR_surface", "VK_KHR_wayland_surface", + /* "VK_KHR_xcb_surface", */ +}; +static const char *const layers[] = {"VK_LAYER_KHRONOS_validation"}; +static const char *const device_extensions[] = {"VK_KHR_swapchain"}; + +static VkBool32 +handle_vulkan_error(VkDebugUtilsMessageSeverityFlagBitsEXT severity, + VkDebugUtilsMessageTypeFlagsEXT type, + const VkDebugUtilsMessengerCallbackDataEXT *callbackData, + void *userData) { + bool backtrace = false; + printf("Vulkan "); + switch (type) { + case VK_DEBUG_UTILS_MESSAGE_TYPE_GENERAL_BIT_EXT: + printf("general "); + break; + case VK_DEBUG_UTILS_MESSAGE_TYPE_VALIDATION_BIT_EXT: + printf("validation "); + break; + case VK_DEBUG_UTILS_MESSAGE_TYPE_PERFORMANCE_BIT_EXT: + printf("performance "); + break; + } + switch (severity) { + case VK_DEBUG_UTILS_MESSAGE_SEVERITY_VERBOSE_BIT_EXT: + printf("(verbose): "); + break; + default: + case VK_DEBUG_UTILS_MESSAGE_SEVERITY_INFO_BIT_EXT: + printf("(info): "); + break; + case VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT: + printf("(warning): "); + break; + case VK_DEBUG_UTILS_MESSAGE_SEVERITY_ERROR_BIT_EXT: + printf("(error): "); + backtrace = true; + break; + } + printf("%s\n", callbackData->pMessage); + if (backtrace) { + print_backtrace(); + } + return 0; +} + +void create_swapchain(struct Vk *state) { + VkSurfaceCapabilitiesKHR capabilities; + CHECK_VK_RESULT(vkGetPhysicalDeviceSurfaceCapabilitiesKHR( + state->phys_device, state->vulkan_surface, &capabilities)); + uint32_t format_count; + CHECK_VK_RESULT(vkGetPhysicalDeviceSurfaceFormatsKHR( + state->phys_device, state->vulkan_surface, &format_count, NULL)); + VkSurfaceFormatKHR formats[format_count]; + CHECK_VK_RESULT(vkGetPhysicalDeviceSurfaceFormatsKHR( + state->phys_device, state->vulkan_surface, &format_count, formats)); + + VkSurfaceFormatKHR chosen_format = formats[0]; + for (uint32_t i = 0; i < format_count; i++) { + if (formats[i].format == VK_FORMAT_R8G8B8A8_SRGB) { + chosen_format = formats[i]; + break; + } + } + + state->format = chosen_format.format; + state->image_count = capabilities.minImageCount + 1; + if (capabilities.maxImageCount > 0 && + capabilities.maxImageCount < state->image_count) { + state->image_count = capabilities.maxImageCount; + } + + VkSwapchainCreateInfoKHR swapchain_info = {0}; + swapchain_info.sType = VK_STRUCTURE_TYPE_SWAPCHAIN_CREATE_INFO_KHR; + swapchain_info.surface = state->vulkan_surface; + swapchain_info.minImageCount = state->image_count; + swapchain_info.imageFormat = chosen_format.format; + swapchain_info.imageColorSpace = chosen_format.colorSpace; + swapchain_info.imageExtent.width = state->width; + swapchain_info.imageExtent.height = state->heigh; + swapchain_info.imageArrayLayers = 1; + swapchain_info.imageUsage = VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT; + if (state->queue_family_index_present != state->queue_family_index_graphics) { + swapchain_info.imageSharingMode = VK_SHARING_MODE_CONCURRENT; + swapchain_info.queueFamilyIndexCount = 2; + uint32_t indices[] = {state->queue_family_index_graphics, + state->queue_family_index_present}; + swapchain_info.pQueueFamilyIndices = indices; + } else { + swapchain_info.imageSharingMode = VK_SHARING_MODE_EXCLUSIVE; + } + swapchain_info.preTransform = capabilities.currentTransform; + swapchain_info.compositeAlpha = VK_COMPOSITE_ALPHA_OPAQUE_BIT_KHR; + swapchain_info.presentMode = VK_PRESENT_MODE_MAILBOX_KHR; + swapchain_info.clipped = VK_TRUE; + swapchain_info.oldSwapchain = VK_NULL_HANDLE; + CHECK_VK_RESULT(vkCreateSwapchainKHR(state->device, &swapchain_info, NULL, + &state->swapchain)); + + CHECK_VK_RESULT(vkGetSwapchainImagesKHR(state->device, state->swapchain, + &state->image_count, NULL)); + + VkImage images[state->image_count]; + CHECK_VK_RESULT(vkGetSwapchainImagesKHR(state->device, state->swapchain, + &state->image_count, images)); + state->elements = malloc(state->image_count * sizeof(struct SwapchainElm)); + + for (int i = 0; i < state->image_count; i++) { + state->elements[i].image = images[i]; + + VkImageViewCreateInfo image_view_info = {0}; + image_view_info.sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO; + image_view_info.viewType = VK_IMAGE_VIEW_TYPE_2D; + image_view_info.components.r = VK_COMPONENT_SWIZZLE_IDENTITY; + image_view_info.components.g = VK_COMPONENT_SWIZZLE_IDENTITY; + image_view_info.components.b = VK_COMPONENT_SWIZZLE_IDENTITY; + image_view_info.components.a = VK_COMPONENT_SWIZZLE_IDENTITY; + image_view_info.subresourceRange.baseMipLevel = 0; + image_view_info.subresourceRange.levelCount = 1; + image_view_info.subresourceRange.baseArrayLayer = 0; + image_view_info.subresourceRange.layerCount = 1; + image_view_info.subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT; + image_view_info.image = state->elements[i].image; + image_view_info.format = state->format; + CHECK_VK_RESULT(vkCreateImageView(state->device, &image_view_info, NULL, + &state->elements[i].image_view)); + } +} + +VkCommandBuffer begin_single_time_commands(struct Vk *state) { + VkCommandBufferAllocateInfo alloc_info = {0}; + alloc_info.sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO; + alloc_info.level = VK_COMMAND_BUFFER_LEVEL_PRIMARY; + alloc_info.commandPool = state->command_pool; + alloc_info.commandBufferCount = 1; + + VkCommandBuffer command_buffer; + vkAllocateCommandBuffers(state->device, &alloc_info, &command_buffer); + + VkCommandBufferBeginInfo begin_info = {0}; + begin_info.sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO; + begin_info.flags = VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT; + + vkBeginCommandBuffer(command_buffer, &begin_info); + + return command_buffer; +} + +void end_single_time_commands(struct Vk *state, + VkCommandBuffer command_buffer) { + vkEndCommandBuffer(command_buffer); + + VkSubmitInfo submit_info = {0}; + submit_info.sType = VK_STRUCTURE_TYPE_SUBMIT_INFO; + submit_info.commandBufferCount = 1; + submit_info.pCommandBuffers = &command_buffer; + + vkQueueSubmit(state->queue_graphics, 1, &submit_info, VK_NULL_HANDLE); + vkQueueWaitIdle(state->queue_graphics); + + vkFreeCommandBuffers(state->device, state->command_pool, 1, &command_buffer); +} + +void setup_framebuffers(struct Vk *state) { + for (int i = 0; i < state->image_count; i++) { + VkFramebufferCreateInfo framebuffer_info = {0}; + framebuffer_info.sType = VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO; + framebuffer_info.renderPass = state->render_pass; + framebuffer_info.attachmentCount = 1; + framebuffer_info.pAttachments = &state->elements[i].image_view; + framebuffer_info.width = state->width; + framebuffer_info.height = state->heigh; + framebuffer_info.layers = 1; + CHECK_VK_RESULT(vkCreateFramebuffer(state->device, &framebuffer_info, NULL, + &state->elements[i].framebuffer)); + } +} + +void destroy_swapchain(struct Vk *state) { + for (int i = 0; i < state->image_count; i++) { + vkDestroyFramebuffer(state->device, state->elements[i].framebuffer, NULL); + vkDestroyImageView(state->device, state->elements[i].image_view, NULL); + } + free(state->elements); + vkDestroySwapchainKHR(state->device, state->swapchain, NULL); +} + +void recreate_swapchain(struct Vk *state) { + destroy_swapchain(state); + create_swapchain(state); + setup_framebuffers(state); +} + +VkShaderModule create_shader_module(struct Vk *state, uint32_t *data, + int size) { + VkShaderModuleCreateInfo shader_module_info = {0}; + shader_module_info.sType = VK_STRUCTURE_TYPE_SHADER_MODULE_CREATE_INFO; + shader_module_info.codeSize = size; + shader_module_info.pCode = data; + + VkShaderModule shader_module; + CHECK_VK_RESULT(vkCreateShaderModule(state->device, &shader_module_info, NULL, + &shader_module)); + return shader_module; +} + +VkInstance create_instance() { + VkApplicationInfo app_info = {0}; + app_info.sType = VK_STRUCTURE_TYPE_APPLICATION_INFO; + app_info.pApplicationName = "meooow"; + app_info.applicationVersion = VK_MAKE_VERSION(0, 0, 1); + app_info.pEngineName = "Cat Engine"; + app_info.engineVersion = VK_MAKE_VERSION(0, 0, 1); + app_info.apiVersion = VK_API_VERSION_1_3; + + VkInstanceCreateInfo instance_info = {0}; + instance_info.sType = VK_STRUCTURE_TYPE_INSTANCE_CREATE_INFO; + instance_info.pApplicationInfo = &app_info; + instance_info.enabledExtensionCount = + sizeof(extensions) / sizeof(const char *); + instance_info.ppEnabledExtensionNames = extensions; + + size_t found_layers = 0; + + uint32_t instance_layer_count; + CHECK_VK_RESULT( + vkEnumerateInstanceLayerProperties(&instance_layer_count, NULL)); + + VkLayerProperties *instance_layer_properties = + malloc(instance_layer_count * sizeof(VkLayerProperties)); + CHECK_VK_RESULT(vkEnumerateInstanceLayerProperties( + &instance_layer_count, instance_layer_properties)); + for (uint32_t i = 0; i < instance_layer_count; i++) { + for (int j = 0; j < sizeof(layers) / sizeof(const char *); j++) { + if (strcmp(instance_layer_properties[i].layerName, layers[j]) == 0) { + found_layers++; + } + } + } + free(instance_layer_properties); + if (found_layers >= sizeof(layers) / sizeof(const char *)) { + instance_info.enabledLayerCount = sizeof(layers) / sizeof(const char *); + instance_info.ppEnabledLayerNames = layers; + } + + VkInstance instance; + CHECK_VK_RESULT(vkCreateInstance(&instance_info, NULL, &instance)); + return instance; +} + +void setup_debug_messenger(struct Vk *state) { + VkDebugUtilsMessengerCreateInfoEXT debug_utils_info = {0}; + debug_utils_info.sType = + VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CREATE_INFO_EXT; + debug_utils_info.messageSeverity = + VK_DEBUG_UTILS_MESSAGE_SEVERITY_INFO_BIT_EXT | + VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT | + VK_DEBUG_UTILS_MESSAGE_SEVERITY_ERROR_BIT_EXT; + debug_utils_info.messageType = + VK_DEBUG_UTILS_MESSAGE_TYPE_GENERAL_BIT_EXT | + VK_DEBUG_UTILS_MESSAGE_TYPE_VALIDATION_BIT_EXT | + VK_DEBUG_UTILS_MESSAGE_TYPE_PERFORMANCE_BIT_EXT; + debug_utils_info.pfnUserCallback = handle_vulkan_error; + + CHECK_VK_RESULT( + GET_EXTENSION_FUNCTION(state->instance, vkCreateDebugUtilsMessengerEXT)( + state->instance, &debug_utils_info, NULL, &state->debug_messenger)); +} + +VkPhysicalDevice create_physical_device(struct Vk *state) { + uint32_t phys_device_count; + vkEnumeratePhysicalDevices(state->instance, &phys_device_count, NULL); + VkPhysicalDevice phys_devices[phys_device_count]; + vkEnumeratePhysicalDevices(state->instance, &phys_device_count, phys_devices); + + VkPhysicalDevice device; + int best_score = 0; + for (int i = 0; i < phys_device_count; i++) { + VkPhysicalDeviceProperties properties; + vkGetPhysicalDeviceProperties(phys_devices[i], &properties); + + int score; + switch (properties.deviceType) { + case VK_PHYSICAL_DEVICE_TYPE_OTHER: + score = 1; + break; + case VK_PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU: + score = 4; + break; + case VK_PHYSICAL_DEVICE_TYPE_DISCRETE_GPU: + score = 5; + break; + case VK_PHYSICAL_DEVICE_TYPE_VIRTUAL_GPU: + score = 3; + break; + case VK_PHYSICAL_DEVICE_TYPE_CPU: + score = 2; + default: + break; + } + if (score > best_score) { + best_score = score; + device = phys_devices[i]; + } + } + return device; +} + +void setup_device_and_queues(struct Vk *state) { + uint32_t queue_family_count; + vkGetPhysicalDeviceQueueFamilyProperties(state->phys_device, + &queue_family_count, NULL); + VkQueueFamilyProperties queue_families[queue_family_count]; + vkGetPhysicalDeviceQueueFamilyProperties(state->phys_device, + &queue_family_count, queue_families); + for (int i = 0; i < queue_family_count; i++) { + VkBool32 present = 0; + CHECK_VK_RESULT(vkGetPhysicalDeviceSurfaceSupportKHR( + state->phys_device, i, state->vulkan_surface, &present)); + + if (queue_families[i].queueFlags & VK_QUEUE_GRAPHICS_BIT) { + state->queue_family_index_graphics = i; + } + if (present) { + state->queue_family_index_present = i; + } + } + uint32_t *indices; + int index_count; + if (state->queue_family_index_present != state->queue_family_index_graphics) { + index_count = 2; + indices = alloca(sizeof(uint32_t) * index_count); + indices[0] = state->queue_family_index_graphics; + indices[1] = state->queue_family_index_present; + } else { + index_count = 1; + indices = alloca(sizeof(uint32_t) * index_count); + indices[0] = state->queue_family_index_graphics; + } + VkDeviceQueueCreateInfo queue_infos[index_count]; + memset(queue_infos, 0, sizeof(VkDeviceQueueCreateInfo) * index_count); + + state->queue_priority = 1.0f; + for (int i = 0; i < index_count; i++) { + queue_infos[i].sType = VK_STRUCTURE_TYPE_DEVICE_QUEUE_CREATE_INFO; + queue_infos[i].queueFamilyIndex = indices[i]; + queue_infos[i].queueCount = 1; + queue_infos[i].pQueuePriorities = &state->queue_priority; + } + + VkPhysicalDeviceFeatures device_features = {0}; + + VkDeviceCreateInfo device_info = {0}; + device_info.sType = VK_STRUCTURE_TYPE_DEVICE_CREATE_INFO; + device_info.queueCreateInfoCount = index_count; + device_info.pQueueCreateInfos = queue_infos; + device_info.enabledExtensionCount = + sizeof(device_extensions) / sizeof(const char *); + device_info.ppEnabledExtensionNames = device_extensions; + device_info.pEnabledFeatures = &device_features; + + CHECK_VK_RESULT( + vkCreateDevice(state->phys_device, &device_info, NULL, &state->device)); + vkGetDeviceQueue(state->device, state->queue_family_index_present, 0, + &state->queue_present); + vkGetDeviceQueue(state->device, state->queue_family_index_graphics, 0, + &state->queue_graphics); +} + +VkRenderPass create_render_pass(struct Vk *state) { + VkAttachmentDescription color_attachment = {0}; + color_attachment.format = state->format; + color_attachment.samples = VK_SAMPLE_COUNT_1_BIT; + color_attachment.loadOp = VK_ATTACHMENT_LOAD_OP_CLEAR; + color_attachment.storeOp = VK_ATTACHMENT_STORE_OP_STORE; + color_attachment.stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE; + color_attachment.stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE; + color_attachment.initialLayout = VK_IMAGE_LAYOUT_UNDEFINED; + color_attachment.finalLayout = VK_IMAGE_LAYOUT_PRESENT_SRC_KHR; + + VkAttachmentReference color_attachment_ref = {0}; + color_attachment_ref.attachment = 0; + color_attachment_ref.layout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL; + + VkSubpassDescription subpass = {0}; + subpass.pipelineBindPoint = VK_PIPELINE_BIND_POINT_GRAPHICS; + subpass.colorAttachmentCount = 1; + subpass.pColorAttachments = &color_attachment_ref; + + VkSubpassDependency dependency = {0}; + dependency.srcSubpass = VK_SUBPASS_EXTERNAL; + dependency.dstSubpass = 0; + dependency.srcStageMask = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT; + dependency.srcAccessMask = 0; + dependency.dstStageMask = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT; + dependency.dstAccessMask = VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT; + + VkRenderPassCreateInfo render_pass_info = {0}; + render_pass_info.sType = VK_STRUCTURE_TYPE_RENDER_PASS_CREATE_INFO; + render_pass_info.attachmentCount = 1; + render_pass_info.pAttachments = &color_attachment; + render_pass_info.subpassCount = 1; + render_pass_info.pSubpasses = &subpass; + render_pass_info.dependencyCount = 1; + render_pass_info.pDependencies = &dependency; + + VkRenderPass render_pass; + CHECK_VK_RESULT( + vkCreateRenderPass(state->device, &render_pass_info, NULL, &render_pass)); + return render_pass; +} + +void setup_command_pool(struct Vk *state) { + VkCommandPoolCreateInfo command_pool_info = {0}; + command_pool_info.sType = VK_STRUCTURE_TYPE_COMMAND_POOL_CREATE_INFO; + command_pool_info.queueFamilyIndex = state->queue_family_index_graphics; + command_pool_info.flags = VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT; + CHECK_VK_RESULT(vkCreateCommandPool(state->device, &command_pool_info, NULL, + &state->command_pool)); +} + +void setup_command_buffers(struct Vk *state) { + VkCommandBufferAllocateInfo command_buffer_alloc_info = {0}; + command_buffer_alloc_info.sType = + VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO; + command_buffer_alloc_info.commandPool = state->command_pool; + command_buffer_alloc_info.level = VK_COMMAND_BUFFER_LEVEL_PRIMARY; + command_buffer_alloc_info.commandBufferCount = 1; + command_buffer_alloc_info.commandBufferCount = MAX_FRAMES_IN_FLIGHT; + VkCommandBuffer buffers[MAX_FRAMES_IN_FLIGHT]; + CHECK_VK_RESULT(vkAllocateCommandBuffers( + state->device, &command_buffer_alloc_info, buffers)); + for (int i = 0; i < MAX_FRAMES_IN_FLIGHT; i++) { + state->flights[i].command_buffer = buffers[i]; + } +} + +void setup_sync_objects(struct Vk *state) { + VkSemaphoreCreateInfo semaphore_info = {0}; + semaphore_info.sType = VK_STRUCTURE_TYPE_SEMAPHORE_CREATE_INFO; + + VkFenceCreateInfo fence_info = {0}; + fence_info.sType = VK_STRUCTURE_TYPE_FENCE_CREATE_INFO; + fence_info.flags = VK_FENCE_CREATE_SIGNALED_BIT; + + for (int i = 0; i < MAX_FRAMES_IN_FLIGHT; i++) { + VkSemaphore ias, rfs; + VkFence iff; + CHECK_VK_RESULT( + vkCreateSemaphore(state->device, &semaphore_info, NULL, &ias)); + CHECK_VK_RESULT( + vkCreateSemaphore(state->device, &semaphore_info, NULL, &rfs)); + CHECK_VK_RESULT(vkCreateFence(state->device, &fence_info, NULL, &iff)); + state->flights[i].image_available_semaphore = ias; + state->flights[i].render_finished_semaphore = rfs; + state->flights[i].in_flight_fence = iff; + } +} + +uint32_t find_memory_type(struct Vk *state, uint32_t memory_type_bits, + uint32_t properties) { + VkPhysicalDeviceMemoryProperties phys_mem_props; + vkGetPhysicalDeviceMemoryProperties(state->phys_device, &phys_mem_props); + + uint32_t memory_type_index = -1; + for (uint32_t i = 0; i < phys_mem_props.memoryTypeCount; i++) { + if (memory_type_bits & (1 << i) && + (phys_mem_props.memoryTypes[i].propertyFlags & properties) == + properties) { + memory_type_index = i; + break; + } + } + if (memory_type_index == -1) { + meow("failed to find memory type"); + } + return memory_type_index; +} + +void create_buffer(struct Vk *state, VkDeviceSize size, + VkBufferUsageFlags usage, VkMemoryPropertyFlags properties, + VkBuffer *buffer, VkDeviceMemory *buffer_memory) { + + VkBufferCreateInfo buffer_info = {0}; + buffer_info.sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO; + buffer_info.size = size; + buffer_info.usage = usage; + buffer_info.sharingMode = VK_SHARING_MODE_EXCLUSIVE; + + CHECK_VK_RESULT(vkCreateBuffer(state->device, &buffer_info, NULL, buffer)); + + VkMemoryRequirements mem_reqs; + vkGetBufferMemoryRequirements(state->device, *buffer, &mem_reqs); + + VkMemoryAllocateInfo alloc_info = {0}; + alloc_info.sType = VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO; + alloc_info.allocationSize = mem_reqs.size; + alloc_info.memoryTypeIndex = + find_memory_type(state, mem_reqs.memoryTypeBits, properties); + + CHECK_VK_RESULT( + vkAllocateMemory(state->device, &alloc_info, NULL, buffer_memory)); + vkBindBufferMemory(state->device, *buffer, *buffer_memory, 0); +} + +void copy_buffer(struct Vk *state, VkBuffer src_buffer, VkBuffer dst_buffer, + VkDeviceSize size) { + VkCommandBuffer command_buffer = begin_single_time_commands(state); + + VkBufferCopy copy_region = { + .srcOffset = 0, + .dstOffset = 0, + .size = size, + }; + vkCmdCopyBuffer(command_buffer, src_buffer, dst_buffer, 1, ©_region); + + end_single_time_commands(state, command_buffer); +} + +void transition_image_layout(struct Vk *state, VkImage image, VkFormat format, + VkImageLayout old_layout, + VkImageLayout new_layout) { + VkCommandBuffer command_buffer = begin_single_time_commands(state); + + VkImageMemoryBarrier barrier = {0}; + barrier.sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER; + barrier.oldLayout = old_layout; + barrier.newLayout = new_layout; + barrier.srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED; + barrier.dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED; + barrier.image = image; + barrier.subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT; + barrier.subresourceRange.baseMipLevel = 0; + barrier.subresourceRange.levelCount = 1; + barrier.subresourceRange.baseArrayLayer = 0; + barrier.subresourceRange.layerCount = 1; + + VkPipelineStageFlags source_stage; + VkPipelineStageFlags dest_stage; + + if (old_layout == VK_IMAGE_LAYOUT_UNDEFINED && + new_layout == VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL) { + barrier.srcAccessMask = 0; + barrier.dstAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + + source_stage = VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT; + dest_stage = VK_PIPELINE_STAGE_TRANSFER_BIT; + } else if (old_layout == VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL && + new_layout == VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL) { + barrier.srcAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + barrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT; + + source_stage = VK_PIPELINE_STAGE_TRANSFER_BIT; + dest_stage = VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT; + } else { + meow("invalid transition OwO rudeeee"); + } + vkCmdPipelineBarrier(command_buffer, source_stage, dest_stage, 0, 0, NULL, 0, + NULL, 1, &barrier); + + end_single_time_commands(state, command_buffer); +} + +void copy_buffer_to_image(struct Vk *state, VkBuffer buffer, VkImage image, + struct IVec2 size) { + VkCommandBuffer command_buffer = begin_single_time_commands(state); + + VkBufferImageCopy region = {0}; + region.bufferOffset = 0; + region.bufferRowLength = 0; + region.bufferImageHeight = 0; + + region.imageSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT; + region.imageSubresource.mipLevel = 0; + region.imageSubresource.baseArrayLayer = 0; + region.imageSubresource.layerCount = 1; + + region.imageOffset = (VkOffset3D){0, 0, 0}; + region.imageExtent = (VkExtent3D){size.x, size.y, 1}; + + vkCmdCopyBufferToImage(command_buffer, buffer, image, + VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, 1, ®ion); + + end_single_time_commands(state, command_buffer); +} + +struct ImageBuffer { + VkImage image; + VkDeviceMemory memory; + struct IVec2 dims; + VkDeviceSize size; + VkImageView view; + VkSampler sampler; +}; + +void create_image(struct Vk *state, struct IVec2 dims, VkFormat format, + VkImageTiling tiling, VkImageUsageFlags usage, + VkMemoryPropertyFlags properties, VkImage *image, + VkDeviceMemory *memory) { + VkImageCreateInfo image_info = {0}; + image_info.sType = VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO; + image_info.imageType = VK_IMAGE_TYPE_2D; + image_info.extent.width = dims.x; + image_info.extent.height = dims.y; + image_info.extent.depth = 1; + image_info.mipLevels = 1; + image_info.arrayLayers = 1; + image_info.format = format; + image_info.tiling = tiling; + image_info.initialLayout = VK_IMAGE_LAYOUT_UNDEFINED; + image_info.usage = usage; + image_info.sharingMode = VK_SHARING_MODE_EXCLUSIVE; + image_info.samples = VK_SAMPLE_COUNT_1_BIT; + + CHECK_VK_RESULT(vkCreateImage(state->device, &image_info, NULL, image)); + + VkMemoryRequirements mem_reqs; + vkGetImageMemoryRequirements(state->device, *image, &mem_reqs); + + VkMemoryAllocateInfo alloc_info = {0}; + alloc_info.sType = VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO; + alloc_info.allocationSize = mem_reqs.size; + alloc_info.memoryTypeIndex = + find_memory_type(state, mem_reqs.memoryTypeBits, properties); + + CHECK_VK_RESULT(vkAllocateMemory(state->device, &alloc_info, NULL, memory)); + + vkBindImageMemory(state->device, *image, *memory, 0); +} + +struct ImageBuffer create_image_buffer(struct Vk *state) { + struct ImageBuffer buffer = {0}; + + uchar *image_data = load_image("image.png", &buffer.dims); + + buffer.size = buffer.dims.x * buffer.dims.y * 4; + + VkBuffer staging_buffer; + VkDeviceMemory staging_buffer_memory; + create_buffer(state, buffer.size, VK_BUFFER_USAGE_TRANSFER_SRC_BIT, + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | + VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, + &staging_buffer, &staging_buffer_memory); + + void *data; + vkMapMemory(state->device, staging_buffer_memory, 0, buffer.size, 0, &data); + memcpy(data, image_data, buffer.size); + vkUnmapMemory(state->device, staging_buffer_memory); + + free(image_data); + + create_image( + state, buffer.dims, VK_FORMAT_R8G8B8A8_SRGB, VK_IMAGE_TILING_OPTIMAL, + VK_IMAGE_USAGE_TRANSFER_DST_BIT | VK_IMAGE_USAGE_SAMPLED_BIT, + VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT, &buffer.image, &buffer.memory); + + transition_image_layout(state, buffer.image, VK_FORMAT_R8G8B8A8_SRGB, + VK_IMAGE_LAYOUT_UNDEFINED, + VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL); + copy_buffer_to_image(state, staging_buffer, buffer.image, buffer.dims); + transition_image_layout(state, buffer.image, VK_FORMAT_R8G8B8A8_SRGB, + VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, + VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL); + + vkDestroyBuffer(state->device, staging_buffer, NULL); + vkFreeMemory(state->device, staging_buffer_memory, NULL); + + VkImageViewCreateInfo view_info = {0}; + view_info.sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO; + view_info.image = buffer.image; + view_info.viewType = VK_IMAGE_VIEW_TYPE_2D; + view_info.format = VK_FORMAT_R8G8B8A8_SRGB; + view_info.subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT; + view_info.subresourceRange.baseMipLevel = 0; + view_info.subresourceRange.levelCount = 1; + view_info.subresourceRange.baseArrayLayer = 0; + view_info.subresourceRange.layerCount = 1; + CHECK_VK_RESULT( + vkCreateImageView(state->device, &view_info, NULL, &buffer.view)); + + VkSamplerCreateInfo sampler_info = {0}; + sampler_info.sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO; + sampler_info.magFilter = VK_FILTER_NEAREST; + sampler_info.minFilter = VK_FILTER_NEAREST; + sampler_info.addressModeU = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + sampler_info.addressModeV = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + sampler_info.addressModeW = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + sampler_info.anisotropyEnable = VK_FALSE; + sampler_info.maxAnisotropy = 1.0f; + sampler_info.borderColor = VK_BORDER_COLOR_INT_TRANSPARENT_BLACK; + sampler_info.unnormalizedCoordinates = VK_FALSE; + sampler_info.compareEnable = VK_FALSE; + sampler_info.compareOp = VK_COMPARE_OP_ALWAYS; + sampler_info.mipmapMode = VK_SAMPLER_MIPMAP_MODE_NEAREST; + sampler_info.mipLodBias = 0.0f; + sampler_info.minLod = 0.0; + sampler_info.maxLod = 0.0; + CHECK_VK_RESULT( + vkCreateSampler(state->device, &sampler_info, NULL, &buffer.sampler)); + + return buffer; +} + +struct VertexBuffer create_vertex_buffer(struct Vk *state, int count, + struct Vertex *vertex_data) { + VkDeviceSize size = count * sizeof(struct Vertex); + struct VertexBuffer buffer = {0}; + buffer.count = count; + + VkBuffer staging_buffer; + VkDeviceMemory staging_buffer_memory; + + create_buffer(state, size, VK_BUFFER_USAGE_TRANSFER_SRC_BIT, + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | + VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, + &staging_buffer, &staging_buffer_memory); + + void *data; + vkMapMemory(state->device, staging_buffer_memory, 0, size, 0, &data); + memcpy(data, vertex_data, size); + vkUnmapMemory(state->device, staging_buffer_memory); + + create_buffer( + state, size, + VK_BUFFER_USAGE_TRANSFER_DST_BIT | VK_BUFFER_USAGE_VERTEX_BUFFER_BIT, + VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT, &buffer.buffer, &buffer.memory); + + copy_buffer(state, staging_buffer, buffer.buffer, size); + + vkDestroyBuffer(state->device, staging_buffer, NULL); + vkFreeMemory(state->device, staging_buffer_memory, NULL); + + return buffer; +} + +struct Vk *init_vk(void *data, int width, int heigh, + VkSurfaceKHR (*surface)(void *, VkInstance)) { + struct Vk *state = malloc(sizeof(struct Vk)); + memset(state, 0, sizeof(struct Vk)); + memset(state->flights, 0, sizeof(state->flights)); + state->width = width; + state->heigh = heigh; + + state->instance = create_instance(); + + setup_debug_messenger(state); + + state->vulkan_surface = surface(data, state->instance); + meow("got a surface at %p", state->vulkan_surface); + + state->phys_device = create_physical_device(state); + setup_device_and_queues(state); + create_swapchain(state); + state->render_pass = create_render_pass(state); + setup_command_pool(state); + setup_framebuffers(state); + setup_command_buffers(state); + setup_sync_objects(state); + + dyn_array_create_inplace(&state->kitties); + + return state; +} + +void uninit_vk(struct Vk *state) { + CHECK_VK_RESULT(vkDeviceWaitIdle(state->device)); + + dyn_array_destroy(&state->kitties); + + destroy_swapchain(state); + + vkDestroyRenderPass(state->device, state->render_pass, NULL); + + for (int i = 0; i < MAX_FRAMES_IN_FLIGHT; i++) { + vkDestroySemaphore(state->device, + state->flights[i].render_finished_semaphore, NULL); + vkDestroySemaphore(state->device, + state->flights[i].image_available_semaphore, NULL); + vkDestroyFence(state->device, state->flights[i].in_flight_fence, NULL); + } + + vkDestroyCommandPool(state->device, state->command_pool, NULL); + + vkDestroyDevice(state->device, NULL); + + GET_EXTENSION_FUNCTION(state->instance, vkDestroyDebugUtilsMessengerEXT)( + state->instance, state->debug_messenger, NULL); + + vkDestroySurfaceKHR(state->instance, state->vulkan_surface, NULL); + vkDestroyInstance(state->instance, NULL); + + memset(state, 0, sizeof(struct Vk)); + free(state); + meow("unmade a vk, nini :3"); +} + +void vk_draw(struct Vk *state) { + CHECK_VK_RESULT(vkWaitForFences( + state->device, 1, &state->flights[state->current_frame].in_flight_fence, + VK_TRUE, UINT64_MAX)); + CHECK_VK_RESULT(vkResetFences( + state->device, 1, &state->flights[state->current_frame].in_flight_fence)); + + uint32_t image_index; + VkResult result = vkAcquireNextImageKHR( + state->device, state->swapchain, UINT64_MAX, + state->flights[state->current_frame].image_available_semaphore, NULL, + &image_index); + + struct SwapchainElm *element = &state->elements[image_index]; + struct InFlightObjects flight = state->flights[state->current_frame]; + + if (result == VK_ERROR_OUT_OF_DATE_KHR || result == VK_SUBOPTIMAL_KHR) { + meow("recreating swapchain"); + CHECK_VK_RESULT(vkDeviceWaitIdle(state->device)); + recreate_swapchain(state); + return; + } else if (result < 0) { + CHECK_VK_RESULT(result); + } + + vkResetCommandBuffer(flight.command_buffer, 0); + + VkCommandBufferBeginInfo command_buffer_begin_info = {0}; + command_buffer_begin_info.sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO; + CHECK_VK_RESULT( + vkBeginCommandBuffer(flight.command_buffer, &command_buffer_begin_info)); + + VkClearValue clear_value = {{0.0f, 0.0f, 0.0f, 1.0f}}; + + VkRenderPassBeginInfo begin_info = {0}; + begin_info.sType = VK_STRUCTURE_TYPE_RENDER_PASS_BEGIN_INFO; + begin_info.renderPass = state->render_pass; + begin_info.framebuffer = element->framebuffer; + begin_info.renderArea.offset = (VkOffset2D){0, 0}; + begin_info.renderArea.extent = (VkExtent2D){state->width, state->heigh}; + begin_info.clearValueCount = 1; + begin_info.pClearValues = &clear_value; + + vkCmdBeginRenderPass(flight.command_buffer, &begin_info, + VK_SUBPASS_CONTENTS_INLINE); + + for (int i = 0; i < state->kitties.count; i++) { + kitty_draw(state, image_index, state->kitties.items[i]); + } + + vkCmdEndRenderPass(flight.command_buffer); + + CHECK_VK_RESULT(vkEndCommandBuffer(flight.command_buffer)); + + VkPipelineStageFlags wait_stage[] = { + VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT}; + VkSemaphore wait_semaphore[] = {flight.image_available_semaphore}; + VkSemaphore signal_semaphore[] = {flight.render_finished_semaphore}; + VkSubmitInfo submit_info = {0}; + submit_info.sType = VK_STRUCTURE_TYPE_SUBMIT_INFO; + submit_info.waitSemaphoreCount = 1; + submit_info.pWaitSemaphores = wait_semaphore; + submit_info.pWaitDstStageMask = wait_stage; + submit_info.commandBufferCount = 1; + submit_info.pCommandBuffers = &flight.command_buffer; + submit_info.signalSemaphoreCount = 1; + submit_info.pSignalSemaphores = signal_semaphore; + CHECK_VK_RESULT(vkQueueSubmit(state->queue_graphics, 1, &submit_info, + flight.in_flight_fence)); + + VkSwapchainKHR swapchain[] = {state->swapchain}; + VkPresentInfoKHR present_info = {0}; + present_info.sType = VK_STRUCTURE_TYPE_PRESENT_INFO_KHR; + present_info.waitSemaphoreCount = 1; + present_info.pWaitSemaphores = signal_semaphore; + present_info.swapchainCount = 1; + present_info.pSwapchains = swapchain; + present_info.pImageIndices = &image_index; + result = vkQueuePresentKHR(state->queue_present, &present_info); + + if (result == VK_ERROR_OUT_OF_DATE_KHR || result == VK_SUBOPTIMAL_KHR) { + CHECK_VK_RESULT(vkDeviceWaitIdle(state->device)); + recreate_swapchain(state); + meow("recreated swapchain"); + } else if (result < 0) { + CHECK_VK_RESULT(result); + } + + state->current_frame = (state->current_frame + 1) % MAX_FRAMES_IN_FLIGHT; +} + +void vk_resize(struct Vk *state, int width, int heigh) { + state->width = width; + state->heigh = heigh; + + CHECK_VK_RESULT(vkDeviceWaitIdle(state->device)); + + recreate_swapchain(state); + + state->current_frame = 0; +} diff --git a/vulkan.h b/vulkan.h new file mode 100644 index 0000000..7cd6de1 --- /dev/null +++ b/vulkan.h @@ -0,0 +1,100 @@ +#ifndef INCLUDE_WAYLANDCLIENT_CAT_VULKAN_H_ +#define INCLUDE_WAYLANDCLIENT_CAT_VULKAN_H_ + +#include + +#include "dynarray.h" +#include "types.h" + +#define CHECK_VK_RESULT(_expr) \ + do { \ + VkResult result = _expr; \ + if (result != VK_SUCCESS) { \ + meow("Error executing %s: %i\n", #_expr, result); \ + } \ + } while (0); + +#define GET_EXTENSION_FUNCTION($instance, _id) \ + ((PFN_##_id)(vkGetInstanceProcAddr($instance, #_id))) + +#define MAX_FRAMES_IN_FLIGHT 2 + +dyn_array_define(da_VK_FORMAT, enum VkFormat); +dyn_array_define(da_Kitty, struct Kitty *); + +struct VertexBuffer { + void *data; + VkBuffer buffer; + VkDeviceMemory memory; + uint32_t count; + int vertex_size; + int binding; + struct da_VK_FORMAT format; +}; + +struct InFlightObjects { + VkSemaphore image_available_semaphore; + VkSemaphore render_finished_semaphore; + VkFence in_flight_fence; + VkCommandBuffer command_buffer; +}; + +struct SwapchainElm { + VkImage image; + VkImageView image_view; + VkFramebuffer framebuffer; +}; + +struct Vk { + VkInstance instance; + + VkPhysicalDevice phys_device; + VkDevice device; + + VkDebugUtilsMessengerEXT debug_messenger; + + VkSurfaceKHR vulkan_surface; + + uint32_t queue_family_index_present; + VkQueue queue_present; + uint32_t queue_family_index_graphics; + VkQueue queue_graphics; + float queue_priority; + + VkCommandPool command_pool; + VkFormat format; + uint32_t image_count; + VkSwapchainKHR swapchain; + struct SwapchainElm *elements; + struct InFlightObjects flights[MAX_FRAMES_IN_FLIGHT]; + int current_frame; + uint32_t width; + uint32_t heigh; + VkRenderPass render_pass; + + struct da_Kitty kitties; +}; + +struct Vk *init_vk(void *data, int width, int heigh, + VkSurfaceKHR (*surface)(void *, VkInstance)); +void uninit_vk(struct Vk *state); +void vk_draw(struct Vk *state); +void vk_resize(struct Vk *state, int width, int heigh); + +void create_buffer(struct Vk *state, VkDeviceSize size, + VkBufferUsageFlags usage, VkMemoryPropertyFlags properties, + VkBuffer *buffer, VkDeviceMemory *buffer_memory); +void copy_buffer(struct Vk *state, VkBuffer src_buffer, VkBuffer dst_buffer, + VkDeviceSize size); +void create_image(struct Vk *state, struct IVec2 dims, VkFormat format, + VkImageTiling tiling, VkImageUsageFlags usage, + VkMemoryPropertyFlags properties, VkImage *image, + VkDeviceMemory *memory); +void transition_image_layout(struct Vk *state, VkImage image, VkFormat format, + VkImageLayout old_layout, + VkImageLayout new_layout); +void copy_buffer_to_image(struct Vk *state, VkBuffer buffer, VkImage image, + struct IVec2 size); +VkShaderModule create_shader_module(struct Vk *state, uint32_t *data, int size); + +#endif // INCLUDE_WAYLANDCLIENT_CAT_VULKAN_H_ diff --git a/vulkan_helpers.c b/vulkan_helpers.c new file mode 100644 index 0000000..9c7a76a --- /dev/null +++ b/vulkan_helpers.c @@ -0,0 +1,265 @@ +#include "log.h" + +#include + +int get_size_of_format(enum VkFormat format) { + switch (format) { + case VK_FORMAT_R32_UINT: + case VK_FORMAT_R32_SINT: + case VK_FORMAT_R32_SFLOAT: + return 4; + case VK_FORMAT_R32G32_UINT: + case VK_FORMAT_R32G32_SINT: + case VK_FORMAT_R32G32_SFLOAT: + return 8; + case VK_FORMAT_R32G32B32_UINT: + case VK_FORMAT_R32G32B32_SINT: + case VK_FORMAT_R32G32B32_SFLOAT: + return 12; + case VK_FORMAT_R32G32B32A32_UINT: + case VK_FORMAT_R32G32B32A32_SINT: + case VK_FORMAT_R32G32B32A32_SFLOAT: + return 16; + case VK_FORMAT_UNDEFINED: + case VK_FORMAT_R4G4_UNORM_PACK8: + case VK_FORMAT_R4G4B4A4_UNORM_PACK16: + case VK_FORMAT_B4G4R4A4_UNORM_PACK16: + case VK_FORMAT_R5G6B5_UNORM_PACK16: + case VK_FORMAT_B5G6R5_UNORM_PACK16: + case VK_FORMAT_R5G5B5A1_UNORM_PACK16: + case VK_FORMAT_B5G5R5A1_UNORM_PACK16: + case VK_FORMAT_A1R5G5B5_UNORM_PACK16: + case VK_FORMAT_R8_UNORM: + case VK_FORMAT_R8_SNORM: + case VK_FORMAT_R8_USCALED: + case VK_FORMAT_R8_SSCALED: + case VK_FORMAT_R8_UINT: + case VK_FORMAT_R8_SINT: + case VK_FORMAT_R8_SRGB: + case VK_FORMAT_R8G8_UNORM: + case VK_FORMAT_R8G8_SNORM: + case VK_FORMAT_R8G8_USCALED: + case VK_FORMAT_R8G8_SSCALED: + case VK_FORMAT_R8G8_UINT: + case VK_FORMAT_R8G8_SINT: + case VK_FORMAT_R8G8_SRGB: + case VK_FORMAT_R8G8B8_UNORM: + case VK_FORMAT_R8G8B8_SNORM: + case VK_FORMAT_R8G8B8_USCALED: + case VK_FORMAT_R8G8B8_SSCALED: + case VK_FORMAT_R8G8B8_UINT: + case VK_FORMAT_R8G8B8_SINT: + case VK_FORMAT_R8G8B8_SRGB: + case VK_FORMAT_B8G8R8_UNORM: + case VK_FORMAT_B8G8R8_SNORM: + case VK_FORMAT_B8G8R8_USCALED: + case VK_FORMAT_B8G8R8_SSCALED: + case VK_FORMAT_B8G8R8_UINT: + case VK_FORMAT_B8G8R8_SINT: + case VK_FORMAT_B8G8R8_SRGB: + case VK_FORMAT_R8G8B8A8_UNORM: + case VK_FORMAT_R8G8B8A8_SNORM: + case VK_FORMAT_R8G8B8A8_USCALED: + case VK_FORMAT_R8G8B8A8_SSCALED: + case VK_FORMAT_R8G8B8A8_UINT: + case VK_FORMAT_R8G8B8A8_SINT: + case VK_FORMAT_R8G8B8A8_SRGB: + case VK_FORMAT_B8G8R8A8_UNORM: + case VK_FORMAT_B8G8R8A8_SNORM: + case VK_FORMAT_B8G8R8A8_USCALED: + case VK_FORMAT_B8G8R8A8_SSCALED: + case VK_FORMAT_B8G8R8A8_UINT: + case VK_FORMAT_B8G8R8A8_SINT: + case VK_FORMAT_B8G8R8A8_SRGB: + case VK_FORMAT_A8B8G8R8_UNORM_PACK32: + case VK_FORMAT_A8B8G8R8_SNORM_PACK32: + case VK_FORMAT_A8B8G8R8_USCALED_PACK32: + case VK_FORMAT_A8B8G8R8_SSCALED_PACK32: + case VK_FORMAT_A8B8G8R8_UINT_PACK32: + case VK_FORMAT_A8B8G8R8_SINT_PACK32: + case VK_FORMAT_A8B8G8R8_SRGB_PACK32: + case VK_FORMAT_A2R10G10B10_UNORM_PACK32: + case VK_FORMAT_A2R10G10B10_SNORM_PACK32: + case VK_FORMAT_A2R10G10B10_USCALED_PACK32: + case VK_FORMAT_A2R10G10B10_SSCALED_PACK32: + case VK_FORMAT_A2R10G10B10_UINT_PACK32: + case VK_FORMAT_A2R10G10B10_SINT_PACK32: + case VK_FORMAT_A2B10G10R10_UNORM_PACK32: + case VK_FORMAT_A2B10G10R10_SNORM_PACK32: + case VK_FORMAT_A2B10G10R10_USCALED_PACK32: + case VK_FORMAT_A2B10G10R10_SSCALED_PACK32: + case VK_FORMAT_A2B10G10R10_UINT_PACK32: + case VK_FORMAT_A2B10G10R10_SINT_PACK32: + case VK_FORMAT_R16_UNORM: + case VK_FORMAT_R16_SNORM: + case VK_FORMAT_R16_USCALED: + case VK_FORMAT_R16_SSCALED: + case VK_FORMAT_R16_UINT: + case VK_FORMAT_R16_SINT: + case VK_FORMAT_R16_SFLOAT: + case VK_FORMAT_R16G16_UNORM: + case VK_FORMAT_R16G16_SNORM: + case VK_FORMAT_R16G16_USCALED: + case VK_FORMAT_R16G16_SSCALED: + case VK_FORMAT_R16G16_UINT: + case VK_FORMAT_R16G16_SINT: + case VK_FORMAT_R16G16_SFLOAT: + case VK_FORMAT_R16G16B16_UNORM: + case VK_FORMAT_R16G16B16_SNORM: + case VK_FORMAT_R16G16B16_USCALED: + case VK_FORMAT_R16G16B16_SSCALED: + case VK_FORMAT_R16G16B16_UINT: + case VK_FORMAT_R16G16B16_SINT: + case VK_FORMAT_R16G16B16_SFLOAT: + case VK_FORMAT_R16G16B16A16_UNORM: + case VK_FORMAT_R16G16B16A16_SNORM: + case VK_FORMAT_R16G16B16A16_USCALED: + case VK_FORMAT_R16G16B16A16_SSCALED: + case VK_FORMAT_R16G16B16A16_UINT: + case VK_FORMAT_R16G16B16A16_SINT: + case VK_FORMAT_R16G16B16A16_SFLOAT: + case VK_FORMAT_R64_UINT: + case VK_FORMAT_R64_SINT: + case VK_FORMAT_R64_SFLOAT: + case VK_FORMAT_R64G64_UINT: + case VK_FORMAT_R64G64_SINT: + case VK_FORMAT_R64G64_SFLOAT: + case VK_FORMAT_R64G64B64_UINT: + case VK_FORMAT_R64G64B64_SINT: + case VK_FORMAT_R64G64B64_SFLOAT: + case VK_FORMAT_R64G64B64A64_UINT: + case VK_FORMAT_R64G64B64A64_SINT: + case VK_FORMAT_R64G64B64A64_SFLOAT: + case VK_FORMAT_B10G11R11_UFLOAT_PACK32: + case VK_FORMAT_E5B9G9R9_UFLOAT_PACK32: + case VK_FORMAT_D16_UNORM: + case VK_FORMAT_X8_D24_UNORM_PACK32: + case VK_FORMAT_D32_SFLOAT: + case VK_FORMAT_S8_UINT: + case VK_FORMAT_D16_UNORM_S8_UINT: + case VK_FORMAT_D24_UNORM_S8_UINT: + case VK_FORMAT_D32_SFLOAT_S8_UINT: + case VK_FORMAT_BC1_RGB_UNORM_BLOCK: + case VK_FORMAT_BC1_RGB_SRGB_BLOCK: + case VK_FORMAT_BC1_RGBA_UNORM_BLOCK: + case VK_FORMAT_BC1_RGBA_SRGB_BLOCK: + case VK_FORMAT_BC2_UNORM_BLOCK: + case VK_FORMAT_BC2_SRGB_BLOCK: + case VK_FORMAT_BC3_UNORM_BLOCK: + case VK_FORMAT_BC3_SRGB_BLOCK: + case VK_FORMAT_BC4_UNORM_BLOCK: + case VK_FORMAT_BC4_SNORM_BLOCK: + case VK_FORMAT_BC5_UNORM_BLOCK: + case VK_FORMAT_BC5_SNORM_BLOCK: + case VK_FORMAT_BC6H_UFLOAT_BLOCK: + case VK_FORMAT_BC6H_SFLOAT_BLOCK: + case VK_FORMAT_BC7_UNORM_BLOCK: + case VK_FORMAT_BC7_SRGB_BLOCK: + case VK_FORMAT_ETC2_R8G8B8_UNORM_BLOCK: + case VK_FORMAT_ETC2_R8G8B8_SRGB_BLOCK: + case VK_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK: + case VK_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK: + case VK_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK: + case VK_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK: + case VK_FORMAT_EAC_R11_UNORM_BLOCK: + case VK_FORMAT_EAC_R11_SNORM_BLOCK: + case VK_FORMAT_EAC_R11G11_UNORM_BLOCK: + case VK_FORMAT_EAC_R11G11_SNORM_BLOCK: + case VK_FORMAT_ASTC_4x4_UNORM_BLOCK: + case VK_FORMAT_ASTC_4x4_SRGB_BLOCK: + case VK_FORMAT_ASTC_5x4_UNORM_BLOCK: + case VK_FORMAT_ASTC_5x4_SRGB_BLOCK: + case VK_FORMAT_ASTC_5x5_UNORM_BLOCK: + case VK_FORMAT_ASTC_5x5_SRGB_BLOCK: + case VK_FORMAT_ASTC_6x5_UNORM_BLOCK: + case VK_FORMAT_ASTC_6x5_SRGB_BLOCK: + case VK_FORMAT_ASTC_6x6_UNORM_BLOCK: + case VK_FORMAT_ASTC_6x6_SRGB_BLOCK: + case VK_FORMAT_ASTC_8x5_UNORM_BLOCK: + case VK_FORMAT_ASTC_8x5_SRGB_BLOCK: + case VK_FORMAT_ASTC_8x6_UNORM_BLOCK: + case VK_FORMAT_ASTC_8x6_SRGB_BLOCK: + case VK_FORMAT_ASTC_8x8_UNORM_BLOCK: + case VK_FORMAT_ASTC_8x8_SRGB_BLOCK: + case VK_FORMAT_ASTC_10x5_UNORM_BLOCK: + case VK_FORMAT_ASTC_10x5_SRGB_BLOCK: + case VK_FORMAT_ASTC_10x6_UNORM_BLOCK: + case VK_FORMAT_ASTC_10x6_SRGB_BLOCK: + case VK_FORMAT_ASTC_10x8_UNORM_BLOCK: + case VK_FORMAT_ASTC_10x8_SRGB_BLOCK: + case VK_FORMAT_ASTC_10x10_UNORM_BLOCK: + case VK_FORMAT_ASTC_10x10_SRGB_BLOCK: + case VK_FORMAT_ASTC_12x10_UNORM_BLOCK: + case VK_FORMAT_ASTC_12x10_SRGB_BLOCK: + case VK_FORMAT_ASTC_12x12_UNORM_BLOCK: + case VK_FORMAT_ASTC_12x12_SRGB_BLOCK: + case VK_FORMAT_G8B8G8R8_422_UNORM: + case VK_FORMAT_B8G8R8G8_422_UNORM: + case VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM: + case VK_FORMAT_G8_B8R8_2PLANE_420_UNORM: + case VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM: + case VK_FORMAT_G8_B8R8_2PLANE_422_UNORM: + case VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM: + case VK_FORMAT_R10X6_UNORM_PACK16: + case VK_FORMAT_R10X6G10X6_UNORM_2PACK16: + case VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16: + case VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16: + case VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16: + case VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16: + case VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16: + case VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16: + case VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16: + case VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16: + case VK_FORMAT_R12X4_UNORM_PACK16: + case VK_FORMAT_R12X4G12X4_UNORM_2PACK16: + case VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16: + case VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16: + case VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16: + case VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16: + case VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16: + case VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16: + case VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16: + case VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16: + case VK_FORMAT_G16B16G16R16_422_UNORM: + case VK_FORMAT_B16G16R16G16_422_UNORM: + case VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM: + case VK_FORMAT_G16_B16R16_2PLANE_420_UNORM: + case VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM: + case VK_FORMAT_G16_B16R16_2PLANE_422_UNORM: + case VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM: + case VK_FORMAT_G8_B8R8_2PLANE_444_UNORM: + case VK_FORMAT_G10X6_B10X6R10X6_2PLANE_444_UNORM_3PACK16: + case VK_FORMAT_G12X4_B12X4R12X4_2PLANE_444_UNORM_3PACK16: + case VK_FORMAT_G16_B16R16_2PLANE_444_UNORM: + case VK_FORMAT_A4R4G4B4_UNORM_PACK16: + case VK_FORMAT_A4B4G4R4_UNORM_PACK16: + case VK_FORMAT_ASTC_4x4_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_5x4_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_5x5_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_6x5_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_6x6_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_8x5_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_8x6_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_8x8_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_10x5_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_10x6_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_10x8_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_10x10_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_12x10_SFLOAT_BLOCK: + case VK_FORMAT_ASTC_12x12_SFLOAT_BLOCK: + case VK_FORMAT_PVRTC1_2BPP_UNORM_BLOCK_IMG: + case VK_FORMAT_PVRTC1_4BPP_UNORM_BLOCK_IMG: + case VK_FORMAT_PVRTC2_2BPP_UNORM_BLOCK_IMG: + case VK_FORMAT_PVRTC2_4BPP_UNORM_BLOCK_IMG: + case VK_FORMAT_PVRTC1_2BPP_SRGB_BLOCK_IMG: + case VK_FORMAT_PVRTC1_4BPP_SRGB_BLOCK_IMG: + case VK_FORMAT_PVRTC2_2BPP_SRGB_BLOCK_IMG: + case VK_FORMAT_PVRTC2_4BPP_SRGB_BLOCK_IMG: + case VK_FORMAT_R16G16_SFIXED5_NV: + case VK_FORMAT_A1B5G5R5_UNORM_PACK16_KHR: + case VK_FORMAT_A8_UNORM_KHR: + case VK_FORMAT_MAX_ENUM: + crash("unsuported format %d", format); + break; + } +} diff --git a/wayland.c b/wayland.c new file mode 100644 index 0000000..0ef7879 --- /dev/null +++ b/wayland.c @@ -0,0 +1,361 @@ +#include "Wayland/xdg-shell-client-protocol.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "dynarray.h" +#include "hashmap.h" +#include "log.h" + +#include "vulkan.h" +#include + +struct cat_Wl { + struct wl_display *display; + struct wl_registry *registry; + struct wl_compositor *compositor; + struct xdg_wm_base *wm_base; + struct wl_seat *seat; + + struct wl_surface *wl_surface; + struct xdg_surface *xdg_surface; + struct xdg_toplevel *xdg_toplevel; + struct wl_keyboard *keyboard; + struct wl_pointer *pointer; + + struct xkb_context *xkb_context; + struct xkb_keymap *keymap; + struct xkb_state *xkb_state; + + int width, heigh; + int new_width, new_heigh; + bool resize_coming; + bool resize_ready; + bool should_close; + struct Vk *vk; + struct Hashmap_ui32 *keys; + + struct da_uint32_t pressed; + struct da_uint32_t released; +}; + +static void wl_buffer_release(void *data, struct wl_buffer *wl_buffer) { + /* Sent by the compositor when it's no longer using this buffer */ + wl_buffer_destroy(wl_buffer); +} + +static const struct wl_buffer_listener wl_buffer_listener = { + .release = wl_buffer_release, +}; + +static void xdg_surface_configure(void *data, struct xdg_surface *xdg_surface, + uint32_t serial) { + struct cat_Wl *state = data; + xdg_surface_ack_configure(xdg_surface, serial); + + if (state->resize_coming) { + state->resize_ready = true; + } +} + +static const struct xdg_surface_listener xdg_surface_listener = { + .configure = xdg_surface_configure, +}; + +static void xdg_wm_base_ping(void *data, struct xdg_wm_base *xdg_wm_base, + uint32_t serial) { + xdg_wm_base_pong(xdg_wm_base, serial); +} + +static const struct xdg_wm_base_listener wm_base_listener = { + .ping = xdg_wm_base_ping, +}; + +static void keyboard_event_keymap(void *data, struct wl_keyboard *wl_keyboard, + uint32_t format, int32_t fd, uint32_t size) { + struct cat_Wl *state = data; + if (format == WL_KEYBOARD_KEYMAP_FORMAT_XKB_V1) { + meow("there is xkb"); + void *data = mmap(NULL, size, PROT_READ, MAP_PRIVATE, fd, 0); + const char *keymap = data; + state->keymap = xkb_keymap_new_from_string(state->xkb_context, keymap, + XKB_KEYMAP_FORMAT_TEXT_V1, + XKB_KEYMAP_COMPILE_NO_FLAGS); + munmap(data, size); + close(fd); + state->xkb_state = xkb_state_new(state->keymap); + } else if (format == WL_KEYBOARD_KEYMAP_FORMAT_NO_KEYMAP) { + meow("keymap is none"); + } +} +static void keyboard_event_enter(void *data, struct wl_keyboard *wl_keyboard, + uint32_t serial, struct wl_surface *surface, + struct wl_array *keys) {} +static void keyboard_event_leave(void *data, struct wl_keyboard *wl_keyboard, + uint32_t serial, struct wl_surface *surface) {} +static void keyboard_event_key(void *data, struct wl_keyboard *wl_keyboard, + uint32_t serial, uint32_t time, uint32_t key, + uint32_t key_state) { + key += 8; + struct cat_Wl *state = data; + if (state->xkb_state == NULL) { + meow("there is no xkb"); + return; + } + xkb_keysym_t sym = xkb_state_key_get_one_sym(state->xkb_state, key); + char buffer[128]; + xkb_state_key_get_utf8(state->xkb_state, key, buffer, sizeof(buffer)); + if (key_state == 1) { + insert_hashmap_ui32(state->keys, sym); + dyn_array_append(&state->pressed, sym); + } else if (key_state == 0) { + remove_hashmap_ui32(state->keys, sym); + dyn_array_append(&state->released, sym); + } else { + meow("bad wayland, a key can either go up or down, not %d", key_state); + } +} +static void keyboard_event_modifiers(void *data, + struct wl_keyboard *wl_keyboard, + uint32_t serial, uint32_t mods_depressed, + uint32_t mods_latched, + uint32_t mods_locked, uint32_t group) { + struct cat_Wl *client_state = data; + xkb_state_update_mask(client_state->xkb_state, mods_depressed, mods_latched, + mods_locked, 0, 0, group); +} +static void keyboard_event_repeat_info(void *data, + struct wl_keyboard *wl_keyboard, + int32_t rate, int32_t delay) {} + +static const struct wl_keyboard_listener keyboard_listener = { + .keymap = keyboard_event_keymap, + .enter = keyboard_event_enter, + .leave = keyboard_event_leave, + .key = keyboard_event_key, + .modifiers = keyboard_event_modifiers, + .repeat_info = keyboard_event_repeat_info, +}; + +static void seat_capabilities_listener(void *data, struct wl_seat *wl_seat, + uint32_t capabilities) { + struct cat_Wl *state = data; + if (capabilities & WL_SEAT_CAPABILITY_KEYBOARD) { + state->keyboard = wl_seat_get_keyboard(wl_seat); + wl_keyboard_add_listener(state->keyboard, &keyboard_listener, state); + } else { + meow("uuuh there is no keyboard"); + if (state->keyboard != NULL) { + wl_keyboard_destroy(state->keyboard); + } + } + if (capabilities & WL_SEAT_CAPABILITY_POINTER) { + state->pointer = wl_seat_get_pointer(wl_seat); + } else { + meow("uuuh there is no mouse"); + if (state->pointer != NULL) { + wl_pointer_destroy(state->pointer); + } + } +} + +static void seat_name_listener(void *data, struct wl_seat *wl_seat, + const char *name) { + meow("the seat is named %s", name); +} + +static const struct wl_seat_listener seat_listener = { + .capabilities = seat_capabilities_listener, + .name = seat_name_listener, +}; + +void xdg_toplevel_configure(void *data, struct xdg_toplevel *xdg_toplevel, + int32_t width, int32_t height, + struct wl_array *states) { + struct cat_Wl *state = data; + if (width != 0 && height != 0) { + state->resize_coming = true; + state->new_width = width; + state->new_heigh = height; + } +} + +void xdg_toplevel_close(void *data, struct xdg_toplevel *xdg_toplevel) { + struct cat_Wl *state = data; + state->should_close = true; +} + +static const struct xdg_toplevel_listener toplevel_listener = { + .configure = xdg_toplevel_configure, + .close = xdg_toplevel_close, +}; + +static void registry_global(void *data, struct wl_registry *registry, + uint32_t name, const char *interface, + uint32_t version) { + struct cat_Wl *state = data; + if (strcmp(interface, wl_compositor_interface.name) == 0) { + state->compositor = + wl_registry_bind(registry, name, &wl_compositor_interface, 6); + } else if (strcmp(interface, xdg_wm_base_interface.name) == 0) { + state->wm_base = + wl_registry_bind(registry, name, &xdg_wm_base_interface, 3); + xdg_wm_base_add_listener(state->wm_base, &wm_base_listener, state); + } else if (strcmp(interface, wl_seat_interface.name) == 0) { + state->seat = wl_registry_bind(registry, name, &wl_seat_interface, 8); + wl_seat_add_listener(state->seat, &seat_listener, state); + } + meow("interface %s: version %d", interface, version); +} + +static void registry_global_remove(void *data, struct wl_registry *wl_registry, + uint32_t name) { + /* This space deliberately left blank */ +} + +static const struct wl_registry_listener registry_listener = { + .global = registry_global, + .global_remove = registry_global_remove, +}; + +VkSurfaceKHR create_wayland_surface(void *data, VkInstance instance) { + struct cat_Wl *state = data; + VkSurfaceKHR surface; + VkWaylandSurfaceCreateInfoKHR surface_info = {0}; + surface_info.sType = VK_STRUCTURE_TYPE_WAYLAND_SURFACE_CREATE_INFO_KHR; + surface_info.display = state->display; + surface_info.surface = state->wl_surface; + vkCreateWaylandSurfaceKHR(instance, &surface_info, NULL, &surface); + return surface; +} + +struct cat_Wl *cat_init_wl(const char *name, int width, int heigh, + struct Vk **vk) { + struct cat_Wl *state = malloc(sizeof(struct cat_Wl)); + memset(state, 0, sizeof(struct cat_Wl)); + state->keys = create_hashmap_ui32(); + dyn_array_create_inplace(&state->pressed); + dyn_array_create_inplace(&state->released); + + state->should_close = state->resize_ready = state->resize_coming = false; + + state->width = width; + state->heigh = heigh; + + state->xkb_context = xkb_context_new(XKB_CONTEXT_NO_FLAGS); + + state->display = wl_display_connect(NULL); + + state->registry = wl_display_get_registry(state->display); + wl_registry_add_listener(state->registry, ®istry_listener, state); + wl_display_roundtrip(state->display); + + state->wl_surface = wl_compositor_create_surface(state->compositor); + state->xdg_surface = + xdg_wm_base_get_xdg_surface(state->wm_base, state->wl_surface); + xdg_surface_add_listener(state->xdg_surface, &xdg_surface_listener, state); + state->xdg_toplevel = xdg_surface_get_toplevel(state->xdg_surface); + xdg_toplevel_add_listener(state->xdg_toplevel, &toplevel_listener, state); + + xdg_toplevel_set_title(state->xdg_toplevel, name); + xdg_toplevel_set_app_id(state->xdg_toplevel, name); + + wl_surface_commit(state->wl_surface); + wl_display_roundtrip(state->display); + wl_surface_commit(state->wl_surface); + + state->vk = init_vk(state, width, heigh, create_wayland_surface); + *vk = state->vk; + return state; +} + +void cat_wl_draw(struct cat_Wl *state) { + dyn_array_reset(&state->pressed); + dyn_array_reset(&state->released); + + if (state->resize_ready && state->resize_coming) { + meow("resizing to %d x %d", state->new_width, state->new_heigh); + state->resize_coming = false; + state->resize_ready = false; + state->width = state->new_width; + state->heigh = state->new_heigh; + + vk_resize(state->vk, state->width, state->heigh); + + wl_surface_commit(state->wl_surface); + } + + vk_draw(state->vk); + + wl_surface_damage_buffer(state->wl_surface, 0, 0, INT32_MAX, INT32_MAX); + wl_surface_commit(state->wl_surface); + + wl_display_roundtrip(state->display); +} + +void cat_uninit_wl(struct cat_Wl *state) { + uninit_vk(state->vk); + + destroy_hashmap_ui32(state->keys); + dyn_array_destroy(&state->pressed); + dyn_array_destroy(&state->released); + + if (state->xkb_state != NULL) { + xkb_state_unref(state->xkb_state); + xkb_keymap_unref(state->keymap); + xkb_context_unref(state->xkb_context); + } + + if (state->keyboard != NULL) { + wl_keyboard_release(state->keyboard); + } + if (state->pointer != NULL) { + wl_pointer_release(state->pointer); + } + if (state->seat != NULL) { + wl_seat_release(state->seat); + } + xdg_toplevel_destroy(state->xdg_toplevel); + xdg_surface_destroy(state->xdg_surface); + wl_surface_destroy(state->wl_surface); + + xdg_wm_base_destroy(state->wm_base); + wl_compositor_destroy(state->compositor); + wl_registry_destroy(state->registry); + wl_display_disconnect(state->display); + + memset(state, 0, sizeof(struct cat_Wl)); + free(state); + + meow("unmade a wl, nini :3"); +} + +bool cat_wl_should_close(struct cat_Wl *state) { return state->should_close; } +bool cat_wl_is_key_pressed(struct cat_Wl *state, uint32_t key) { + return get_hashmap_ui32(state->keys, key); +} +bool cat_wl_was_key_just_pressed(struct cat_Wl *state, uint32_t key) { + for (int i = 0; i < state->pressed.count; i++) { + if (state->pressed.items[i] == key) { + return true; + } + } + return false; +} +bool cat_wl_was_key_just_released(struct cat_Wl *state, uint32_t key) { + for (int i = 0; i < state->released.count; i++) { + if (state->released.items[i] == key) { + return true; + } + } + return false; +} diff --git a/wayland.h b/wayland.h new file mode 100644 index 0000000..c0dca01 --- /dev/null +++ b/wayland.h @@ -0,0 +1,17 @@ +#ifndef INCLUDE_WAYLANDCLIENT_CAT_WAYLAND_H_ +#define INCLUDE_WAYLANDCLIENT_CAT_WAYLAND_H_ + +#include +#include + +struct cat_Wl; +struct cat_Wl *cat_init_wl(const char *name, int width, int heigh, + struct Vk **vk); +void cat_wl_draw(struct cat_Wl *state); +void cat_uninit_wl(struct cat_Wl *state); +bool cat_wl_should_close(struct cat_Wl *state); +bool cat_wl_is_key_pressed(struct cat_Wl *state, uint32_t key); +bool cat_wl_was_key_just_pressed(struct cat_Wl *state, uint32_t key); +bool cat_wl_was_key_just_released(struct cat_Wl *state, uint32_t key); + +#endif // INCLUDE_WAYLANDCLIENT_CAT_WAYLAND_H_