Add OS Lyrics extension (#2656)

* draft commit

* time to fight pipeline

* round 2 changes

* remove unnecessary line

* fight taglib. again

* make taglib work again???

* add id3 tags

* taglib 1.12 vs 1.13

* use int instead for windows

* store as json now

* add migration, more tests

* support repeated line, multiline

* fix ms and support .m, .mm, .mmm

* address some concerns, make cpp a bit safer

* separate responses from model

* remove [:]

* Add trace log

* Try to unblock pipeline

* Fix merge errors

* Fix SIGSEGV error (proper handling of empty frames)

* Add fallback artist/title to structured lyrics

* Rename conflicting named vars

* Fix tests

* Do we still need ffmpeg in the pipeline?

* Revert "Do we still need ffmpeg in the pipeline?"

Yes we do.

This reverts commit 87df7f6df79bccee83f48c4b7a8118a7636a5e66.

* Does this passes now, with a newer ffmpeg version?

* Revert "Does this passes now, with a newer ffmpeg version?"

No, it does not :(

This reverts commit 372eb4b0ae05d9ffe98078e9bc4e56a9b2921f32.

* My OCD made me do it :P

---------

Co-authored-by: Deluan Quintão <deluan@navidrome.org>

scanner/metadata/metadata_test.go

@@ -1,15 +1,64 @@
 package metadata_test
 
 import (
+	"encoding/json"
+	"strings"
+
 	"github.com/navidrome/navidrome/conf"
+	"github.com/navidrome/navidrome/conf/configtest"
+	"github.com/navidrome/navidrome/core/ffmpeg"
+	"github.com/navidrome/navidrome/model"
 	"github.com/navidrome/navidrome/scanner/metadata"
 	_ "github.com/navidrome/navidrome/scanner/metadata/ffmpeg"
 	_ "github.com/navidrome/navidrome/scanner/metadata/taglib"
 	. "github.com/onsi/ginkgo/v2"
 	. "github.com/onsi/gomega"
+	"golang.org/x/exp/slices"
 )
 
 var _ = Describe("Tags", func() {
+	var zero int64 = 0
+	var secondTs int64 = 2500
+
+	makeLyrics := func(synced bool, lang, secondLine string) model.Lyrics {
+		lines := []model.Line{
+			{Value: "This is"},
+			{Value: secondLine},
+		}
+
+		if synced {
+			lines[0].Start = &zero
+			lines[1].Start = &secondTs
+		}
+
+		lyrics := model.Lyrics{
+			Lang:   lang,
+			Line:   lines,
+			Synced: synced,
+		}
+
+		return lyrics
+	}
+
+	sortLyrics := func(lines model.LyricList) model.LyricList {
+		slices.SortFunc(lines, func(a, b model.Lyrics) bool {
+			langDiff := strings.Compare(a.Lang, b.Lang)
+			if langDiff == 0 {
+				return strings.Compare(a.Line[1].Value, b.Line[1].Value) < 0
+			} else {
+				return langDiff < 0
+			}
+		})
+
+		return lines
+	}
+
+	compareLyrics := func(m metadata.Tags, expected model.LyricList) {
+		lyrics := model.LyricList{}
+		Expect(json.Unmarshal([]byte(m.Lyrics()), &lyrics)).To(BeNil())
+		Expect(sortLyrics(lyrics)).To(Equal(sortLyrics(expected)))
+	}
+
 	Context("Extract", func() {
 		BeforeEach(func() {
 			conf.Server.Scanner.Extractor = "taglib"
@@ -61,10 +110,10 @@ var _ = Describe("Tags", func() {
 			Expect(m.Duration()).To(BeNumerically("~", 1.04, 0.01))
 			Expect(m.Suffix()).To(Equal("ogg"))
 			Expect(m.FilePath()).To(Equal("tests/fixtures/test.ogg"))
-			Expect(m.Size()).To(Equal(int64(6333)))
+			Expect(m.Size()).To(Equal(int64(5534)))
 			// TabLib 1.12 returns 18, previous versions return 39.
 			// See https://github.com/taglib/taglib/commit/2f238921824741b2cfe6fbfbfc9701d9827ab06b
-			Expect(m.BitRate()).To(BeElementOf(18, 39, 40, 49))
+			Expect(m.BitRate()).To(BeElementOf(18, 39, 40, 43, 49))
 
 			m = mds["tests/fixtures/test.wma"]
 			Expect(err).To(BeNil())
@@ -74,8 +123,86 @@ var _ = Describe("Tags", func() {
 			Expect(m.Duration()).To(BeNumerically("~", 1.02, 0.01))
 			Expect(m.Suffix()).To(Equal("wma"))
 			Expect(m.FilePath()).To(Equal("tests/fixtures/test.wma"))
-			Expect(m.Size()).To(Equal(int64(21431)))
+			Expect(m.Size()).To(Equal(int64(21581)))
 			Expect(m.BitRate()).To(BeElementOf(128))
 		})
+
+		DescribeTable("Lyrics test",
+			func(file string, langEncoded bool) {
+				path := "tests/fixtures/" + file
+				mds, err := metadata.Extract(path)
+				Expect(err).ToNot(HaveOccurred())
+				Expect(mds).To(HaveLen(1))
+
+				m := mds[path]
+				lyrics := model.LyricList{
+					makeLyrics(true, "xxx", "English"),
+					makeLyrics(true, "xxx", "unspecified"),
+				}
+				if langEncoded {
+					lyrics[0].Lang = "eng"
+				}
+				compareLyrics(m, lyrics)
+			},
+
+			Entry("Parses AIFF file", "test.aiff", true),
+			Entry("Parses FLAC files", "test.flac", false),
+			Entry("Parses M4A files", "01 Invisible (RED) Edit Version.m4a", false),
+			Entry("Parses OGG Vorbis files", "test.ogg", false),
+			Entry("Parses WAV files", "test.wav", true),
+			Entry("Parses WMA files", "test.wma", false),
+			Entry("Parses WV files", "test.wv", false),
+		)
+
+		It("Should parse mp3 with USLT and SYLT", func() {
+			path := "tests/fixtures/test.mp3"
+			mds, err := metadata.Extract(path)
+			Expect(err).ToNot(HaveOccurred())
+			Expect(mds).To(HaveLen(1))
+
+			m := mds[path]
+			compareLyrics(m, model.LyricList{
+				makeLyrics(true, "eng", "English SYLT"),
+				makeLyrics(true, "eng", "English"),
+				makeLyrics(true, "xxx", "unspecified SYLT"),
+				makeLyrics(true, "xxx", "unspecified"),
+			})
+		})
+	})
+
+	// Only run these tests if FFmpeg is available
+	FFmpegContext := XContext
+	if ffmpeg.New().IsAvailable() {
+		FFmpegContext = Context
+	}
+	FFmpegContext("Extract with FFmpeg", func() {
+		BeforeEach(func() {
+			DeferCleanup(configtest.SetupConfig())
+			conf.Server.Scanner.Extractor = "ffmpeg"
+		})
+
+		DescribeTable("Lyrics test",
+			func(file string) {
+				path := "tests/fixtures/" + file
+				mds, err := metadata.Extract(path)
+				Expect(err).ToNot(HaveOccurred())
+				Expect(mds).To(HaveLen(1))
+
+				m := mds[path]
+				compareLyrics(m, model.LyricList{
+					makeLyrics(true, "eng", "English"),
+					makeLyrics(true, "xxx", "unspecified"),
+				})
+			},
+
+			Entry("Parses AIFF file", "test.aiff"),
+			Entry("Parses MP3 files", "test.mp3"),
+			// Disabled, because it fails in pipeline
+			// Entry("Parses WAV files", "test.wav"),
+
+			// FFMPEG behaves very weirdly for multivalued tags for non-ID3
+			// Specifically, they are separated by ";, which is indistinguishable
+			// from other fields
+		)
 	})
 })