fix: extract stream level metadata

scanner/metadata_ffmpeg.go

@@ -143,21 +143,21 @@ func isAudioFile(extension string) bool {
 
 var (
 	tagsRx = map[*regexp.Regexp]string{
-		regexp.MustCompile(`(?i)^\s{4}compilation\s+:(.*)`):    "compilation",
+		regexp.MustCompile(`(?i)^\s{4,6}compilation\s+:(.*)`):         "compilation",
-		regexp.MustCompile(`(?i)^\s{4}genre\s+:\s(.*)`):        "genre",
+		regexp.MustCompile(`(?i)^\s{4,6}genre\s+:\s(.*)`):             "genre",
-		regexp.MustCompile(`(?i)^\s{4}title\s+:\s(.*)`):        "title",
+		regexp.MustCompile(`(?i)^\s{4,6}title\s+:\s(.*)`):             "title",
-		regexp.MustCompile(`(?i)^\s{4}comment\s+:\s(.*)`):      "comment",
+		regexp.MustCompile(`(?i)^\s{4,6}comment\s+:\s(.*)`):           "comment",
-		regexp.MustCompile(`(?i)^\s{4}artist\s+:\s(.*)`):       "artist",
+		regexp.MustCompile(`(?i)^\s{4,6}artist\s+:\s(.*)`):            "artist",
-		regexp.MustCompile(`(?i)^\s{4}album_artist\s+:\s(.*)`): "album_artist",
+		regexp.MustCompile(`(?i)^\s{4,6}album_artist\s+:\s(.*)`):      "album_artist",
-		regexp.MustCompile(`(?i)^\s{4}TCM\s+:\s(.*)`):          "composer",
+		regexp.MustCompile(`(?i)^\s{4,6}TCM\s+:\s(.*)`):               "composer",
-		regexp.MustCompile(`(?i)^\s{4}album\s+:\s(.*)`):        "album",
+		regexp.MustCompile(`(?i)^\s{4,6}album\s+:\s(.*)`):             "album",
-		regexp.MustCompile(`(?i)^\s{4}track\s+:\s(.*)`):        "trackNum",
+		regexp.MustCompile(`(?i)^\s{4,6}track\s+:\s(.*)`):             "trackNum",
-		regexp.MustCompile(`(?i)^\s{4}tracktotal\s+:\s(.*)`):   "trackTotal",
+		regexp.MustCompile(`(?i)^\s{4,6}tracktotal\s+:\s(.*)`):        "trackTotal",
-		regexp.MustCompile(`(?i)^\s{4}disc\s+:\s(.*)`):         "discNum",
+		regexp.MustCompile(`(?i)^\s{4,6}disc\s+:\s(.*)`):              "discNum",
-		regexp.MustCompile(`(?i)^\s{4}disctotal\s+:\s(.*)`):    "discTotal",
+		regexp.MustCompile(`(?i)^\s{4,6}disctotal\s+:\s(.*)`):         "discTotal",
-		regexp.MustCompile(`(?i)^\s{4}TPA\s+:\s(.*)`):          "discNum",
+		regexp.MustCompile(`(?i)^\s{4,6}TPA\s+:\s(.*)`):               "discNum",
-		regexp.MustCompile(`(?i)^\s{4}date\s+:\s(.*)`):         "year",
+		regexp.MustCompile(`(?i)^\s{4,6}date\s+:\s(.*)`):              "year",
-		regexp.MustCompile(`^\s{4}Stream #\d+:\d+: (.+):\s`):   "hasPicture",
+		regexp.MustCompile(`^\s{4}Stream #\d+:\d+: (Video):.*pic\)$`): "hasPicture",
 	}
 
 	durationRx = regexp.MustCompile(`^\s\sDuration: ([\d.:]+).*bitrate: (\d+)`)
@@ -168,7 +168,14 @@ func (m *Metadata) parseInfo(info string) {
 	scanner := bufio.NewScanner(reader)
 	for scanner.Scan() {
 		line := scanner.Text()
+		if len(line) == 0 {
+			continue
+		}
 		for rx, tag := range tagsRx {
+			// Skip when the tag was previously found
+			if _, ok := m.tags[tag]; ok {
+				continue
+			}
 			match := rx.FindStringSubmatch(line)
 			if len(match) > 0 {
 				m.tags[tag] = match[1]

scanner/metadata_test.go

@@ -67,31 +67,55 @@ var _ = Describe("Metadata", func() {
 	})
 
 	Context("extractMetadata", func() {
+		It("detects embedded cover art correctly", func() {
+			const output = `
+Input #0, mp3, from '/Users/deluan/Music/iTunes/iTunes Media/Music/Compilations/Putumayo Presents Blues Lounge/09 Pablo's Blues.mp3':
+  Metadata:
+    compilation     : 1
+  Duration: 00:00:01.02, start: 0.000000, bitrate: 477 kb/s
+    Stream #0:0: Audio: mp3, 44100 Hz, stereo, fltp, 192 kb/s
+    Stream #0:1: Video: mjpeg (Baseline), gray(bt470bg/unknown/unknown), 600x600 [SAR 72:72 DAR 1:1], 90k tbr, 90k tbn, 90k tbc (attached pic)`
+			md, _ := extractMetadata("tests/fixtures/test.mp3", output)
+			Expect(md.HasPicture()).To(BeTrue())
+		})
+
 		It("parses correctly the compilation tag", func() {
-			const outputWithOverlappingTitleTag = `
+			const output = `
 Input #0, mp3, from '/Users/deluan/Music/iTunes/iTunes Media/Music/Compilations/Putumayo Presents Blues Lounge/09 Pablo's Blues.mp3':
   Metadata:
     compilation     : 1
   Duration: 00:05:02.63, start: 0.000000, bitrate: 140 kb/s`
-			md, _ := extractMetadata("tests/fixtures/test.mp3", outputWithOverlappingTitleTag)
+			md, _ := extractMetadata("tests/fixtures/test.mp3", output)
 			Expect(md.Compilation()).To(BeTrue())
 		})
 
-		It("parses correct the title without overlapping with the stream tag", func() {
+		It("parses stream level tags", func() {
-			const outputWithOverlappingTitleTag = `
+			const output = `
+Input #0, ogg, from './01-02 Drive (Teku).opus':
+  Metadata:
+    ALBUM           : Hot Wheels Acceleracers Soundtrack
+  Duration: 00:03:37.37, start: 0.007500, bitrate: 135 kb/s
+    Stream #0:0(eng): Audio: opus, 48000 Hz, stereo, fltp
+    Metadata:
+      TITLE           : Drive (Teku)`
+			md, _ := extractMetadata("tests/fixtures/test.mp3", output)
+			Expect(md.Title()).To(Equal("Drive (Teku)"))
+		})
+
+		It("does not overlap top level tags with the stream level tags", func() {
+			const output = `
 Input #0, mp3, from 'groovin.mp3':
   Metadata:
     title           : Groovin' (feat. Daniel Sneijers, Susanne Alt)
   Duration: 00:03:34.28, start: 0.025056, bitrate: 323 kb/s
     Metadata:
-      title           : cover
+      title           : garbage`
-At least one output file must be specified`
+			md, _ := extractMetadata("tests/fixtures/test.mp3", output)
-			md, _ := extractMetadata("tests/fixtures/test.mp3", outputWithOverlappingTitleTag)
 			Expect(md.Title()).To(Equal("Groovin' (feat. Daniel Sneijers, Susanne Alt)"))
 		})
 
 		It("ignores case in the tag name", func() {
-			const outputWithOverlappingTitleTag = `
+			const output = `
 Input #0, flac, from '/Users/deluan/Downloads/06. Back In Black.flac':
   Metadata:
     ALBUM           : Back In Black
@@ -103,7 +127,7 @@ Input #0, flac, from '/Users/deluan/Downloads/06. Back In Black.flac':
     TRACKTOTAL      : 10
     track           : 6
   Duration: 00:04:16.00, start: 0.000000, bitrate: 995 kb/s`
-			md, _ := extractMetadata("tests/fixtures/test.mp3", outputWithOverlappingTitleTag)
+			md, _ := extractMetadata("tests/fixtures/test.mp3", output)
 			Expect(md.Title()).To(Equal("Back In Black"))
 			Expect(md.Album()).To(Equal("Back In Black"))
 			Expect(md.Genre()).To(Equal("Hard Rock"))