Add .gitattribute assisted language detection to blame, diff and render (#17590)

Use check attribute code to check the assigned language of a file and send that in to
chroma as a hint for the language of the file.

Signed-off-by: Andrew Thornton <art27@cantab.net>

2 files changed, 54 insertions, 2 deletions

diff --git a/routers/web/repo/blame.go b/routers/web/repo/blame.go
index 110ec037e1..2fd72d81ab 100644
--- a/routers/web/repo/blame.go
+++ b/routers/web/repo/blame.go
@@ -16,6 +16,7 @@ import (
 	"code.gitea.io/gitea/modules/context"
 	"code.gitea.io/gitea/modules/git"
 	"code.gitea.io/gitea/modules/highlight"
+	"code.gitea.io/gitea/modules/log"
 	"code.gitea.io/gitea/modules/templates"
 	"code.gitea.io/gitea/modules/timeutil"
 	"code.gitea.io/gitea/modules/util"
@@ -204,6 +205,31 @@ func processBlameParts(ctx *context.Context, blameParts []git.BlamePart) (map[st
 func renderBlame(ctx *context.Context, blameParts []git.BlamePart, commitNames map[string]*models.UserCommit, previousCommits map[string]string) {
 	repoLink := ctx.Repo.RepoLink
 
+	language := ""
+
+	indexFilename, worktree, deleteTemporaryFile, err := ctx.Repo.GitRepo.ReadTreeToTemporaryIndex(ctx.Repo.CommitID)
+	if err == nil {
+		defer deleteTemporaryFile()
+
+		filename2attribute2info, err := ctx.Repo.GitRepo.CheckAttribute(git.CheckAttributeOpts{
+			CachedOnly: true,
+			Attributes: []string{"linguist-language", "gitlab-language"},
+			Filenames:  []string{ctx.Repo.TreePath},
+			IndexFile:  indexFilename,
+			WorkTree:   worktree,
+		})
+		if err != nil {
+			log.Error("Unable to load attributes for %-v:%s. Error: %v", ctx.Repo.Repository, ctx.Repo.TreePath, err)
+		}
+
+		language = filename2attribute2info[ctx.Repo.TreePath]["linguist-language"]
+		if language == "" || language == "unspecified" {
+			language = filename2attribute2info[ctx.Repo.TreePath]["gitlab-language"]
+		}
+		if language == "unspecified" {
+			language = ""
+		}
+	}
 	var lines = make([]string, 0)
 	rows := make([]*blameRow, 0)
 
@@ -248,7 +274,7 @@ func renderBlame(ctx *context.Context, blameParts []git.BlamePart, commitNames m
 				line += "\n"
 			}
 			fileName := fmt.Sprintf("%v", ctx.Data["FileName"])
-			line = highlight.Code(fileName, line)
+			line = highlight.Code(fileName, language, line)
 
 			br.Code = gotemplate.HTML(line)
 			rows = append(rows, br)
diff --git a/routers/web/repo/view.go b/routers/web/repo/view.go
index 12b3aef505..938292a37a 100644
--- a/routers/web/repo/view.go
+++ b/routers/web/repo/view.go
@@ -502,7 +502,33 @@ func renderFile(ctx *context.Context, entry *git.TreeEntry, treeLink, rawLink st
 			lineNums := linesBytesCount(buf)
 			ctx.Data["NumLines"] = strconv.Itoa(lineNums)
 			ctx.Data["NumLinesSet"] = true
-			ctx.Data["FileContent"] = highlight.File(lineNums, blob.Name(), buf)
+
+			language := ""
+
+			indexFilename, worktree, deleteTemporaryFile, err := ctx.Repo.GitRepo.ReadTreeToTemporaryIndex(ctx.Repo.CommitID)
+			if err == nil {
+				defer deleteTemporaryFile()
+
+				filename2attribute2info, err := ctx.Repo.GitRepo.CheckAttribute(git.CheckAttributeOpts{
+					CachedOnly: true,
+					Attributes: []string{"linguist-language", "gitlab-language"},
+					Filenames:  []string{ctx.Repo.TreePath},
+					IndexFile:  indexFilename,
+					WorkTree:   worktree,
+				})
+				if err != nil {
+					log.Error("Unable to load attributes for %-v:%s. Error: %v", ctx.Repo.Repository, ctx.Repo.TreePath, err)
+				}
+
+				language = filename2attribute2info[ctx.Repo.TreePath]["linguist-language"]
+				if language == "" || language == "unspecified" {
+					language = filename2attribute2info[ctx.Repo.TreePath]["gitlab-language"]
+				}
+				if language == "unspecified" {
+					language = ""
+				}
+			}
+			ctx.Data["FileContent"] = highlight.File(lineNums, blob.Name(), language, buf)
 		}
 		if !isLFSFile {
 			if ctx.Repo.CanEnableEditor() {