feat: add support for VP8 and VP9 video codecs

Added support for VP8 and VP9 video codecs in MP4 containers: - Added vp08 and vp09 boxes to decoder map - Added VP Codec Configuration Box (vpcC) - Updated StsdBox to support VP8/VP9 codec configurations
Eyevinn · Jan 25, 2025 · 7978de0 · 7978de0
1 parent 9019d6c
commit 7978de0
Show file tree

Hide file tree

Showing 8 changed files with 402 additions and 211 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
diff --git a/mp4/box.go b/mp4/box.go
@@ -19,20 +19,24 @@ var decoders map[string]BoxDecoder
 
 func init() {
 	decoders = map[string]BoxDecoder{
+		"\xa9ART": DecodeGenericContainerBox,
+		"\xa9nam": DecodeGenericContainerBox,
+		"\xa9too": DecodeGenericContainerBox,
+		"\xa9cpy": DecodeGenericContainerBox,
 		"ac-3":    DecodeAudioSampleEntry,
 		"alou":    DecodeAlou,
 		"av01":    DecodeVisualSampleEntry,
+		"av1C":    DecodeAv1C,
 		"avc1":    DecodeVisualSampleEntry,
 		"avc3":    DecodeVisualSampleEntry,
 		"avcC":    DecodeAvcC,
-		"av1C":    DecodeAv1C,
 		"btrt":    DecodeBtrt,
 		"cdat":    DecodeCdat,
 		"cdsc":    DecodeTrefType,
 		"clap":    DecodeClap,
-		"cslg":    DecodeCslg,
 		"co64":    DecodeCo64,
 		"colr":    DecodeColr,
+		"cslg":    DecodeCslg,
 		"ctim":    DecodeCtim,
 		"ctts":    DecodeCtts,
 		"dac3":    DecodeDac3,
@@ -43,15 +47,15 @@ func init() {
 		"dpnd":    DecodeTrefType,
 		"dref":    DecodeDref,
 		"ec-3":    DecodeAudioSampleEntry,
-		"elng":    DecodeElng,
-		"esds":    DecodeEsds,
 		"edts":    DecodeEdts,
+		"elng":    DecodeElng,
 		"elst":    DecodeElst,
 		"emeb":    DecodeEmeb,
 		"emib":    DecodeEmib,
+		"emsg":    DecodeEmsg,
 		"enca":    DecodeAudioSampleEntry,
 		"encv":    DecodeVisualSampleEntry,
-		"emsg":    DecodeEmsg,
+		"esds":    DecodeEsds,
 		"evte":    DecodeEvte,
 		"font":    DecodeTrefType,
 		"free":    DecodeFree,
@@ -61,8 +65,8 @@ func init() {
 		"hev1":    DecodeVisualSampleEntry,
 		"hind":    DecodeTrefType,
 		"hint":    DecodeTrefType,
-		"hvcC":    DecodeHvcC,
 		"hvc1":    DecodeVisualSampleEntry,
+		"hvcC":    DecodeHvcC,
 		"iden":    DecodeIden,
 		"ilst":    DecodeIlst,
 		"iods":    DecodeUnknown,
@@ -82,10 +86,10 @@ func init() {
 		"minf":    DecodeMinf,
 		"moof":    DecodeMoof,
 		"moov":    DecodeMoov,
+		"mp4a":    DecodeAudioSampleEntry,
 		"mpod":    DecodeTrefType,
 		"mvex":    DecodeMvex,
 		"mvhd":    DecodeMvhd,
-		"mp4a":    DecodeAudioSampleEntry,
 		"nmhd":    DecodeNmhd,
 		"pasp":    DecodePasp,
 		"payl":    DecodePayl,
@@ -105,9 +109,9 @@ func init() {
 		"skip":    DecodeFree,
 		"smhd":    DecodeSmhd,
 		"ssix":    DecodeSsix,
-		"sthd":    DecodeSthd,
 		"stbl":    DecodeStbl,
 		"stco":    DecodeStco,
+		"sthd":    DecodeSthd,
 		"stpp":    DecodeStpp,
 		"stsc":    DecodeStsc,
 		"stsd":    DecodeStsd,
@@ -137,17 +141,16 @@ func init() {
 		"vdep":    DecodeTrefType,
 		"vlab":    DecodeVlab,
 		"vmhd":    DecodeVmhd,
+		"vp08":    DecodeVisualSampleEntry,
+		"vp09":    DecodeVisualSampleEntry,
+		"vpcC":    DecodeVppC,
 		"vplx":    DecodeTrefType,
 		"vsid":    DecodeVsid,
 		"vtta":    DecodeVtta,
 		"vttc":    DecodeVttc,
 		"vttC":    DecodeVttC,
 		"vtte":    DecodeVtte,
 		"wvtt":    DecodeWvtt,
-		"\xa9cpy": DecodeGenericContainerBox,
-		"\xa9nam": DecodeGenericContainerBox,
-		"\xa9too": DecodeGenericContainerBox,
-		"\xa9ART": DecodeGenericContainerBox,
 	}
 }
 

diff --git a/mp4/boxsr.go b/mp4/boxsr.go
@@ -10,20 +10,24 @@ var decodersSR map[string]BoxDecoderSR
 
 func init() {
 	decodersSR = map[string]BoxDecoderSR{
+		"\xa9ART": DecodeGenericContainerBoxSR,
+		"\xa9cpy": DecodeGenericContainerBoxSR,
+		"\xa9nam": DecodeGenericContainerBoxSR,
+		"\xa9too": DecodeGenericContainerBoxSR,
 		"ac-3":    DecodeAudioSampleEntrySR,
+		"alou":    DecodeAlouBoxSR,
 		"av01":    DecodeVisualSampleEntrySR,
+		"av1C":    DecodeAv1CSR,
 		"avc1":    DecodeVisualSampleEntrySR,
 		"avc3":    DecodeVisualSampleEntrySR,
-		"alou":    DecodeAlouBoxSR,
 		"avcC":    DecodeAvcCSR,
-		"av1C":    DecodeAv1CSR,
 		"btrt":    DecodeBtrtSR,
 		"cdat":    DecodeCdatSR,
 		"cdsc":    DecodeTrefTypeSR,
 		"clap":    DecodeClapSR,
-		"cslg":    DecodeCslgSR,
 		"co64":    DecodeCo64SR,
 		"colr":    DecodeColrSR,
+		"cslg":    DecodeCslgSR,
 		"ctim":    DecodeCtimSR,
 		"ctts":    DecodeCttsSR,
 		"dac3":    DecodeDac3SR,
@@ -34,15 +38,15 @@ func init() {
 		"dpnd":    DecodeTrefTypeSR,
 		"dref":    DecodeDrefSR,
 		"ec-3":    DecodeAudioSampleEntrySR,
+		"edts":    DecodeEdtsSR,
 		"elng":    DecodeElngSR,
+		"elst":    DecodeElstSR,
 		"emeb":    DecodeEmebSR,
 		"emib":    DecodeEmibSR,
-		"esds":    DecodeEsdsSR,
-		"edts":    DecodeEdtsSR,
-		"elst":    DecodeElstSR,
+		"emsg":    DecodeEmsgSR,
 		"enca":    DecodeAudioSampleEntrySR,
 		"encv":    DecodeVisualSampleEntrySR,
-		"emsg":    DecodeEmsgSR,
+		"esds":    DecodeEsdsSR,
 		"evte":    DecodeEvteSR,
 		"font":    DecodeTrefTypeSR,
 		"free":    DecodeFreeSR,
@@ -52,8 +56,8 @@ func init() {
 		"hev1":    DecodeVisualSampleEntrySR,
 		"hind":    DecodeTrefTypeSR,
 		"hint":    DecodeTrefTypeSR,
-		"hvcC":    DecodeHvcCSR,
 		"hvc1":    DecodeVisualSampleEntrySR,
+		"hvcC":    DecodeHvcCSR,
 		"iden":    DecodeIdenSR,
 		"ilst":    DecodeIlstSR,
 		"iods":    DecodeUnknownSR,
@@ -73,10 +77,10 @@ func init() {
 		"minf":    DecodeMinfSR,
 		"moof":    DecodeMoofSR,
 		"moov":    DecodeMoovSR,
+		"mp4a":    DecodeAudioSampleEntrySR,
 		"mpod":    DecodeTrefTypeSR,
 		"mvex":    DecodeMvexSR,
 		"mvhd":    DecodeMvhdSR,
-		"mp4a":    DecodeAudioSampleEntrySR,
 		"nmhd":    DecodeNmhdSR,
 		"pasp":    DecodePaspSR,
 		"payl":    DecodePaylSR,
@@ -96,9 +100,9 @@ func init() {
 		"skip":    DecodeFreeSR,
 		"smhd":    DecodeSmhdSR,
 		"ssix":    DecodeSsixSR,
-		"sthd":    DecodeSthdSR,
 		"stbl":    DecodeStblSR,
 		"stco":    DecodeStcoSR,
+		"sthd":    DecodeSthdSR,
 		"stpp":    DecodeStppSR,
 		"stsc":    DecodeStscSR,
 		"stsd":    DecodeStsdSR,
@@ -128,17 +132,16 @@ func init() {
 		"vdep":    DecodeTrefTypeSR,
 		"vlab":    DecodeVlabSR,
 		"vmhd":    DecodeVmhdSR,
+		"vp08":    DecodeVisualSampleEntrySR,
+		"vp09":    DecodeVisualSampleEntrySR,
+		"vpcC":    DecodeVppCSR,
 		"vplx":    DecodeTrefTypeSR,
 		"vsid":    DecodeVsidSR,
 		"vtta":    DecodeVttaSR,
 		"vttc":    DecodeVttcSR,
 		"vttC":    DecodeVttCSR,
 		"vtte":    DecodeVtteSR,
 		"wvtt":    DecodeWvttSR,
-		"\xa9cpy": DecodeGenericContainerBoxSR,
-		"\xa9nam": DecodeGenericContainerBoxSR,
-		"\xa9too": DecodeGenericContainerBoxSR,
-		"\xa9ART": DecodeGenericContainerBoxSR,
 	}
 }
 

diff --git a/mp4/stsd.go b/mp4/stsd.go
@@ -24,6 +24,8 @@ type StsdBox struct {
 	Av01 *VisualSampleEntryBox
 	// Encv is a pointer to a box with name encv
 	Encv *VisualSampleEntryBox
+	// VppX is a pointer to a box with name vp08 or vp09 (VP8 or VP9 video)
+	VppX *VisualSampleEntryBox
 	// Mp4a is a pointer to a box with name mp4a
 	Mp4a *AudioSampleEntryBox
 	// AC3 is a pointer to a box with name ac-3
@@ -57,6 +59,8 @@ func (s *StsdBox) AddChild(box Box) {
 		s.Encv = box.(*VisualSampleEntryBox)
 	case "av01":
 		s.Av01 = box.(*VisualSampleEntryBox)
+	case "vp08", "vp09":
+		s.VppX = box.(*VisualSampleEntryBox)
 	case "mp4a":
 		s.Mp4a = box.(*AudioSampleEntryBox)
 	case "ac-3":

diff --git a/mp4/stsd_test.go b/mp4/stsd_test.go
@@ -2,6 +2,7 @@ package mp4
 
 import (
 	"bytes"
+	"encoding/hex"
 	"testing"
 
 	"github.com/Eyevinn/mp4ff/aac"
@@ -72,3 +73,20 @@ func TestStsdEncodeDecode(t *testing.T) {
 		t.Errorf("Expected nil, got %v", btrt)
 	}
 }
+
+func TestStsdVP9(t *testing.T) {
+	hexData := "" +
+		"000000a87374736400000000000000010000009876703039000000000000" +
+		"000100000000000000000000000000000000050002d00048000000480000" +
+		"000000000001184c61766336312e31392e313030206c69627670782d7670" +
+		"39000000000000000018ffff000000147670634301000000001f80020202" +
+		"00000000000a6669656c0100000000107061737000000001000000010000" +
+		"001462747274000000000010152200101522"
+
+	binData, err := hex.DecodeString(hexData)
+	if err != nil {
+		t.Error(err)
+	}
+
+	cmpAfterDecodeEncodeBox(t, binData)
+}
diff --git a/mp4/visualsampleentry.go b/mp4/visualsampleentry.go
@@ -21,6 +21,7 @@ type VisualSampleEntryBox struct {
 	AvcC               *AvcCBox
 	HvcC               *HvcCBox
 	Av1C               *Av1CBox
+	VppC               *VppCBox
 	Btrt               *BtrtBox
 	Clap               *ClapBox
 	Pasp               *PaspBox
@@ -63,6 +64,8 @@ func (b *VisualSampleEntryBox) AddChild(child Box) {
 		b.HvcC = box
 	case *Av1CBox:
 		b.Av1C = box
+	case *VppCBox:
+		b.VppC = box
 	case *BtrtBox:
 		b.Btrt = box
 	case *ClapBox:
@@ -72,7 +75,6 @@ func (b *VisualSampleEntryBox) AddChild(child Box) {
 	case *SinfBox:
 		b.Sinf = box
 	}
-
 	b.Children = append(b.Children, child)
 }
 
@@ -112,7 +114,7 @@ func DecodeVisualSampleEntrySR(hdr BoxHeader, startPos uint64, sr bits.SliceRead
 	}
 	b.CompressorName = sr.ReadFixedLengthString(int(compressorNameLength))
 	sr.SkipBytes(int(31 - compressorNameLength))
-	sr.ReadUint16() // depth == 0x0018
+	sr.SkipBytes(2) // Skip depth
 	sr.ReadUint16() // pre_defined == -1
 
 	// Now there may be clap and pasp boxes
@@ -179,7 +181,7 @@ func (b *VisualSampleEntryBox) Encode(w io.Writer) error {
 	sw.WriteUint8(compressorNameLength)
 	sw.WriteString(b.CompressorName, false)
 	sw.WriteZeroBytes(int(31 - compressorNameLength))
-	sw.WriteUint16(0x0018) // depth == 0x0018
+	sw.WriteUint16(0x0018) // depth
 	sw.WriteUint16(0xffff) // pre_defined == -1  //86 bytes
 
 	_, err = w.Write(buf[:sw.Offset()]) // Only write  written bytes
@@ -218,7 +220,7 @@ func (b *VisualSampleEntryBox) EncodeSW(sw bits.SliceWriter) error {
 	sw.WriteUint8(compressorNameLength)
 	sw.WriteString(b.CompressorName, false)
 	sw.WriteZeroBytes(int(31 - compressorNameLength))
-	sw.WriteUint16(0x0018) // depth == 0x0018
+	sw.WriteUint16(0x0018) // depth
 	sw.WriteUint16(0xffff) // pre_defined == -1  //86 bytes
 
 	// Next output child boxes in order