nune/gitea
nune/gitea
0
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Add support for linguist-detectable and linguist-documentation (#29267)

Browse source 
Add support for `linguist-detectable` and `linguist-documentation`
Add tests for the attributes


https://github.com/github-linguist/linguist/blob/master/docs/overrides.md#detectable

https://github.com/github-linguist/linguist/blob/master/docs/overrides.md#documentation
This commit is contained in:
KN4CK3R 2024-02-23 18:24:27 +01:00 • committed by GitHub
parent 7d0903bf90
commit 2a278b996f
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 365 additions and 71 deletions
Download patch file Download diff file Expand all files Collapse all files

77
modules/git/repo_language_stats_gogit.go
View file

@ -11,6 +11,7 @@ import (
"strings"
"code.gitea.io/gitea/modules/analyze"
"code.gitea.io/gitea/modules/optional"
"github.com/go-enry/go-enry/v2"
"github.com/go-git/go-git/v5"
@ -57,25 +58,47 @@ func (repo *Repository) GetLanguageStats(commitID string) (map[string]int64, err
return nil
}
notVendored := false
notGenerated := false
isVendored := optional.None[bool]()
isGenerated := optional.None[bool]()
isDocumentation := optional.None[bool]()
isDetectable := optional.None[bool]()
if checker != nil {
attrs, err := checker.CheckPath(f.Name)
if err == nil {
if vendored, has := attrs["linguist-vendored"]; has {
if vendored == "set" || vendored == "true" {
return nil
}
notVendored = vendored == "false"
isVendored = attributeToBool(attrs, "linguist-vendored")
if isVendored.ValueOrDefault(false) {
return nil
}
if generated, has := attrs["linguist-generated"]; has {
if generated == "set" || generated == "true" {
return nil
}
notGenerated = generated == "false"
isGenerated = attributeToBool(attrs, "linguist-generated")
if isGenerated.ValueOrDefault(false) {
return nil
}
if language, has := attrs["linguist-language"]; has && language != "unspecified" && language != "" {
isDocumentation = attributeToBool(attrs, "linguist-documentation")
if isDocumentation.ValueOrDefault(false) {
return nil
}
isDetectable = attributeToBool(attrs, "linguist-detectable")
if !isDetectable.ValueOrDefault(true) {
return nil
}
hasLanguage := attributeToString(attrs, "linguist-language")
if hasLanguage.Value() == "" {
hasLanguage = attributeToString(attrs, "gitlab-language")
if hasLanguage.Has() {
language := hasLanguage.Value()
if idx := strings.IndexByte(language, '?'); idx >= 0 {
hasLanguage = optional.Some(language[:idx])
}
}
}
if hasLanguage.Value() != "" {
language := hasLanguage.Value()
// group languages, such as Pug -> HTML; SCSS -> CSS
group := enry.GetLanguageGroup(language)
if len(group) != 0 {
@ -85,28 +108,14 @@ func (repo *Repository) GetLanguageStats(commitID string) (map[string]int64, err
// this language will always be added to the size
sizes[language] += f.Size
return nil
} else if language, has := attrs["gitlab-language"]; has && language != "unspecified" && language != "" {
// strip off a ? if present
if idx := strings.IndexByte(language, '?'); idx >= 0 {
language = language[:idx]
}
if len(language) != 0 {
// group languages, such as Pug -> HTML; SCSS -> CSS
group := enry.GetLanguageGroup(language)
if len(group) != 0 {
language = group
}
// this language will always be added to the size
sizes[language] += f.Size
return nil
}
}
}
}
if (!notVendored && analyze.IsVendor(f.Name)) || enry.IsDotFile(f.Name) ||
enry.IsDocumentation(f.Name) || enry.IsConfiguration(f.Name) {
if (!isVendored.Has() && analyze.IsVendor(f.Name)) ||
enry.IsDotFile(f.Name) ||
(!isDocumentation.Has() && enry.IsDocumentation(f.Name)) ||
enry.IsConfiguration(f.Name) {
return nil
}
@ -115,12 +124,10 @@ func (repo *Repository) GetLanguageStats(commitID string) (map[string]int64, err
if f.Size <= bigFileSize {
content, _ = readFile(f, fileSizeLimit)
}
if !notGenerated && enry.IsGenerated(f.Name, content) {
if !isGenerated.Has() && enry.IsGenerated(f.Name, content) {
return nil
}
// TODO: Use .gitattributes file for linguist overrides
language := analyze.GetCodeLanguage(f.Name, content)
if language == enry.OtherLanguage || language == "" {
return nil
@ -138,7 +145,7 @@ func (repo *Repository) GetLanguageStats(commitID string) (map[string]int64, err
included = langtype == enry.Programming || langtype == enry.Markup
includedLanguage[language] = included
}
if included {
if included || isDetectable.ValueOrDefault(false) {
sizes[language] += f.Size
} else if len(sizes) == 0 && (firstExcludedLanguage == "" || firstExcludedLanguage == language) {
firstExcludedLanguage = language