Rework raw file http header logic (#20484)

- Always respect the user's configured mime type map
- Allow more types like image/pdf/video/audio to serve with correct content-type
- Shorten cache duration of raw files to 5 minutes, matching GitHub
- Don't set `content-disposition: attachment`, let the browser decide whether it wants to download or display a file directly
- Implement rfc5987 for filenames, remove previous hack. Confirmed it working in Safari.
- Make PDF attachment work in Safari by removing `sandbox` attribute.

This change will make a lot more file types open directly in browser now. Logic should generally be more readable than before with less `if` nesting and such.

Replaces: https://github.com/go-gitea/gitea/pull/20460
Replaces: https://github.com/go-gitea/gitea/pull/20455
Fixes: https://github.com/go-gitea/gitea/issues/20404

modules/typesniffer/typesniffer.go

@@ -70,6 +70,16 @@ func (ct SniffedType) IsRepresentableAsText() bool {
 	return ct.IsText() || ct.IsSvgImage()
 }
 
+// IsBrowsableType returns whether a non-text type can be displayed in a browser
+func (ct SniffedType) IsBrowsableBinaryType() bool {
+	return ct.IsImage() || ct.IsSvgImage() || ct.IsPDF() || ct.IsVideo() || ct.IsAudio()
+}
+
+// GetMimeType returns the mime type
+func (ct SniffedType) GetMimeType() string {
+	return strings.SplitN(ct.contentType, ";", 2)[0]
+}
+
 // DetectContentType extends http.DetectContentType with more content types. Defaults to text/unknown if input is empty.
 func DetectContentType(data []byte) SniffedType {
 	if len(data) == 0 {

routers/common/repo.go

@@ -7,12 +7,13 @@ package common
 import (
 	"fmt"
 	"io"
+	"net/url"
 	"path"
 	"path/filepath"
 	"strings"
 	"time"
 
-	"code.gitea.io/gitea/modules/charset"
+	charsetModule "code.gitea.io/gitea/modules/charset"
 	"code.gitea.io/gitea/modules/context"
 	"code.gitea.io/gitea/modules/git"
 	"code.gitea.io/gitea/modules/httpcache"
@@ -42,7 +43,7 @@ func ServeBlob(ctx *context.Context, blob *git.Blob, lastModified time.Time) err
 }
 
 // ServeData download file from io.Reader
-func ServeData(ctx *context.Context, name string, size int64, reader io.Reader) error {
+func ServeData(ctx *context.Context, filePath string, size int64, reader io.Reader) error {
 	buf := make([]byte, 1024)
 	n, err := util.ReadAtMost(reader, buf)
 	if err != nil {
@@ -52,56 +53,73 @@ func ServeData(ctx *context.Context, name string, size int64, reader io.Reader)
 		buf = buf[:n]
 	}
 
-	ctx.Resp.Header().Set("Cache-Control", "public,max-age=86400")
+	httpcache.AddCacheControlToHeader(ctx.Resp.Header(), 5*time.Minute)
 
 	if size >= 0 {
 		ctx.Resp.Header().Set("Content-Length", fmt.Sprintf("%d", size))
 	} else {
-		log.Error("ServeData called to serve data: %s with size < 0: %d", name, size)
+		log.Error("ServeData called to serve data: %s with size < 0: %d", filePath, size)
 	}
-	name = path.Base(name)
 
-	// Google Chrome dislike commas in filenames, so let's change it to a space
-	name = strings.ReplaceAll(name, ",", " ")
+	fileName := path.Base(filePath)
+	sniffedType := typesniffer.DetectContentType(buf)
+	isPlain := sniffedType.IsText() || ctx.FormBool("render")
+	mimeType := ""
+	charset := ""
 
-	st := typesniffer.DetectContentType(buf)
-
-	mappedMimeType := ""
 	if setting.MimeTypeMap.Enabled {
-		fileExtension := strings.ToLower(filepath.Ext(name))
-		mappedMimeType = setting.MimeTypeMap.Map[fileExtension]
+		fileExtension := strings.ToLower(filepath.Ext(fileName))
+		mimeType = setting.MimeTypeMap.Map[fileExtension]
 	}
-	if st.IsText() || ctx.FormBool("render") {
-		cs, err := charset.DetectEncoding(buf)
-		if err != nil {
-			log.Error("Detect raw file %s charset failed: %v, using by default utf-8", name, err)
-			cs = "utf-8"
+
+	if mimeType == "" {
+		if sniffedType.IsBrowsableBinaryType() {
+			mimeType = sniffedType.GetMimeType()
+		} else if isPlain {
+			mimeType = "text/plain"
+		} else {
+			mimeType = typesniffer.ApplicationOctetStream
 		}
-		if mappedMimeType == "" {
-			mappedMimeType = "text/plain"
+	}
+
+	if isPlain {
+		charset, err = charsetModule.DetectEncoding(buf)
+		if err != nil {
+			log.Error("Detect raw file %s charset failed: %v, using by default utf-8", filePath, err)
+			charset = "utf-8"
 		}
-		ctx.Resp.Header().Set("Content-Type", mappedMimeType+"; charset="+strings.ToLower(cs))
+	}
+
+	if charset != "" {
+		ctx.Resp.Header().Set("Content-Type", mimeType+"; charset="+strings.ToLower(charset))
 	} else {
-		ctx.Resp.Header().Set("Access-Control-Expose-Headers", "Content-Disposition")
-		if mappedMimeType != "" {
-			ctx.Resp.Header().Set("Content-Type", mappedMimeType)
-		}
-		if (st.IsImage() || st.IsPDF()) && (setting.UI.SVG.Enabled || !st.IsSvgImage()) {
-			ctx.Resp.Header().Set("Content-Disposition", fmt.Sprintf(`inline; filename="%s"`, name))
-			if st.IsSvgImage() || st.IsPDF() {
-				ctx.Resp.Header().Set("Content-Security-Policy", "default-src 'none'; style-src 'unsafe-inline'; sandbox")
-				ctx.Resp.Header().Set("X-Content-Type-Options", "nosniff")
-				if st.IsSvgImage() {
-					ctx.Resp.Header().Set("Content-Type", typesniffer.SvgMimeType)
-				} else {
-					ctx.Resp.Header().Set("Content-Type", typesniffer.ApplicationOctetStream)
-				}
-			}
-		} else {
-			ctx.Resp.Header().Set("Content-Disposition", fmt.Sprintf(`attachment; filename="%s"`, name))
-		}
+		ctx.Resp.Header().Set("Content-Type", mimeType)
+	}
+	ctx.Resp.Header().Set("X-Content-Type-Options", "nosniff")
+
+	isSVG := sniffedType.IsSvgImage()
+
+	// serve types that can present a security risk with CSP
+	if isSVG {
+		ctx.Resp.Header().Set("Content-Security-Policy", "default-src 'none'; style-src 'unsafe-inline'; sandbox")
+	} else if sniffedType.IsPDF() {
+		// no sandbox attribute for pdf as it breaks rendering in at least safari. this
+		// should generally be safe as scripts inside PDF can not escape the PDF document
+		// see https://bugs.chromium.org/p/chromium/issues/detail?id=413851 for more discussion
+		ctx.Resp.Header().Set("Content-Security-Policy", "default-src 'none'; style-src 'unsafe-inline'")
 	}
 
+	disposition := "inline"
+	if isSVG && !setting.UI.SVG.Enabled {
+		disposition = "attachment"
+	}
+
+	// encode filename per https://datatracker.ietf.org/doc/html/rfc5987
+	encodedFileName := `filename*=UTF-8''` + url.PathEscape(fileName)
+
+	ctx.Resp.Header().Set("Content-Disposition", disposition+"; "+encodedFileName)
+	ctx.Resp.Header().Set("Access-Control-Expose-Headers", "Content-Disposition")
+
 	_, err = ctx.Resp.Write(buf)
 	if err != nil {
 		return err