LibMedia: Let the Matroska LanguageBCP47 element override Language

The Matroska spec indicates that this field should always override the
Language element when it is present.
This commit is contained in:
Zaggy1024 2025-09-30 17:00:11 -05:00 committed by Jelle Raaijmakers
parent e11da1f85f
commit 29ab9c5fd5
3 changed files with 15 additions and 1 deletions

View File

@ -120,6 +120,8 @@ public:
void set_name(String const& name) { m_name = name; }
String language() const { return m_language; }
void set_language(String const& language) { m_language = language; }
Optional<String> const& language_bcp_47() const { return m_language_bcp_47; }
void set_language_bcp_47(String const& language_bcp_47) { m_language_bcp_47 = language_bcp_47; }
String codec_id() const { return m_codec_id; }
void set_codec_id(String const& codec_id) { m_codec_id = codec_id; }
ReadonlyBytes codec_private_data() const LIFETIME_BOUND { return m_codec_private_data.span(); }
@ -155,6 +157,7 @@ private:
TrackType m_track_type { Invalid };
String m_name;
String m_language = "eng"_string;
Optional<String> m_language_bcp_47;
String m_codec_id;
FixedArray<u8> m_codec_private_data;
double m_timestamp_scale { 1 };

View File

@ -68,7 +68,13 @@ static TrackType track_type_from_matroska_track_type(TrackEntry::TrackType type)
static Track track_from_track_entry(TrackEntry const& track_entry)
{
auto name = Utf16String::from_utf8(track_entry.name());
auto language = Utf16String::from_utf8(track_entry.language());
auto language = [&] {
// LanguageBCP47 - The language of the track, in the BCP47 form; see basics on language codes. If this Element is used,
// then any Language Elements used in the same TrackEntry MUST be ignored.
if (track_entry.language_bcp_47().has_value())
return Utf16String::from_utf8(track_entry.language_bcp_47().value());
return Utf16String::from_utf8(track_entry.language());
}();
Track track(track_type_from_matroska_track_type(track_entry.track_type()), track_entry.track_number(), name, language);
if (track.type() == TrackType::Video) {

View File

@ -52,6 +52,7 @@ constexpr u32 TRACK_UID_ID = 0x73C5;
constexpr u32 TRACK_TYPE_ID = 0x83;
constexpr u32 TRACK_NAME_ID = 0x536E;
constexpr u32 TRACK_LANGUAGE_ID = 0x22B59C;
constexpr u32 TRACK_LANGUAGE_BCP_47_ID = 0x22B59D;
constexpr u32 TRACK_CODEC_ID = 0x86;
constexpr u32 TRACK_CODEC_PRIVATE_ID = 0x63A2;
constexpr u32 TRACK_CODEC_DELAY_ID = 0x56AA;
@ -485,6 +486,10 @@ static DecoderErrorOr<NonnullRefPtr<TrackEntry>> parse_track_entry(Streamer& str
track_entry->set_language(DECODER_TRY_ALLOC(String::from_byte_string(TRY_READ(streamer.read_string()))));
dbgln_if(MATROSKA_TRACE_DEBUG, "Read Track's Language attribute: {}", track_entry->language());
break;
case TRACK_LANGUAGE_BCP_47_ID:
track_entry->set_language_bcp_47(DECODER_TRY_ALLOC(String::from_byte_string(TRY_READ(streamer.read_string()))));
dbgln_if(MATROSKA_TRACE_DEBUG, "Read Track's LanguageBCP47 attribute: {}", track_entry->language());
break;
case TRACK_CODEC_ID:
track_entry->set_codec_id(DECODER_TRY_ALLOC(String::from_byte_string(TRY_READ(streamer.read_string()))));
dbgln_if(MATROSKA_TRACE_DEBUG, "Read Track's CodecID attribute: {}", track_entry->codec_id());