Improve code for DVB VBI data

libavcodec/codec_desc.c: move to correct location and correct description
libavcodec/codec_id.h: move to correct location in data codecs
libavformat/mpegts-mythtv.c: reuse teletext_descriptor code for VBI_teletext_descriptor

libmythbase/iso639.h: add used include
libmythbase/stringutil.h: add new funtion split_sv()
libmythtv/decoders/avformatdecoder.cpp:
AvFormatDecoder::ScanTeletextCaptions() was specific to AV_CODEC_ID_DVB_TELETEXT,
looking only at the teletext_descriptor in the PMT.

ScanTeletextCaptions() no longer uses the MythTV exported PMT, using what FFmpeg
already exported when it parsed PMTs.  Both AV_CODEC_ID_DVB_TELETEXT and
AV_CODEC_ID_DVB_VBI (if there is EBU teletext data in the VBI data) now add to
the tracks list and multiple streams are now supported.

mythtv/external/FFmpeg/libavcodec/codec_desc.c

@@ -3622,12 +3622,6 @@ static const AVCodecDescriptor codec_descriptors[] = {
         .name      = "mpeg2vbi",
         .long_name = NULL_IF_CONFIG_SMALL("ivtv proprietary embedded VBI captions"),
     },
-    {
-        .id        = AV_CODEC_ID_DVB_VBI,
-        .type      = AVMEDIA_TYPE_DATA,
-        .name      = "dvb_vbi",
-        .long_name = NULL_IF_CONFIG_SMALL("dvb teletext"),
-    },
     {
         .id        = AV_CODEC_ID_DSMCC_B,
         .type      = AVMEDIA_TYPE_DATA,
@@ -3720,6 +3714,12 @@ static const AVCodecDescriptor codec_descriptors[] = {
         .name      = "lcevc",
         .long_name = NULL_IF_CONFIG_SMALL("LCEVC (Low Complexity Enhancement Video Coding) / MPEG-5 LCEVC / MPEG-5 part 2"),
     },
+    {
+        .id        = AV_CODEC_ID_DVB_VBI,
+        .type      = AVMEDIA_TYPE_DATA,
+        .name      = "dvb_vbi",
+        .long_name = NULL_IF_CONFIG_SMALL("DVB VBI data"),
+    },
     {
         .id        = AV_CODEC_ID_MPEG2TS,
         .type      = AVMEDIA_TYPE_DATA,

mythtv/external/FFmpeg/libavcodec/codec_id.h

@@ -577,7 +577,6 @@ enum AVCodecID {
     /* MythTV */
     /* teletext codecs */
     AV_CODEC_ID_MPEG2VBI,
-    AV_CODEC_ID_DVB_VBI,
 
     /* DSMCC codec */
     AV_CODEC_ID_DSMCC_B,
@@ -598,6 +597,7 @@ enum AVCodecID {
     AV_CODEC_ID_BIN_DATA,
     AV_CODEC_ID_SMPTE_2038,
     AV_CODEC_ID_LCEVC,
+    AV_CODEC_ID_DVB_VBI,
 
 
     AV_CODEC_ID_PROBE = 0x19000, ///< codec_id is not known (like AV_CODEC_ID_NONE) but lavf should attempt to identify it

mythtv/external/FFmpeg/libavformat/mpegts-mythtv.c

@@ -1917,6 +1917,7 @@ int ff_mythtv_parse_mpeg2_descriptor(AVFormatContext *fc, AVStream *st, int stre
             }
         }
         break;
+    case VBI_TELETEXT_DESCRIPTOR:
     case 0x56: /* DVB teletext descriptor */
         {
             uint8_t *extradata = NULL;
@@ -2070,15 +2071,6 @@ int ff_mythtv_parse_mpeg2_descriptor(AVFormatContext *fc, AVStream *st, int stre
         sti->stream_identifier = 1 + get8(pp, desc_end);
         st->component_tag     = sti->stream_identifier - 1;
         break;
-    case VBI_TELETEXT_DESCRIPTOR:
-        language[0] = get8(pp, desc_end);
-        language[1] = get8(pp, desc_end);
-        language[2] = get8(pp, desc_end);
-        language[3] = 0;
-
-        if (language[0])
-            av_dict_set(&st->metadata, "language", language, 0);
-        break;
     case METADATA_DESCRIPTOR:
         if (get16(pp, desc_end) == 0xFFFF)
             *pp += 4;

mythtv/libs/libmythbase/iso639.h

@@ -2,6 +2,7 @@
 #ifndef ISO_639_2_H
 #define ISO_639_2_H
 
+#include <cctype>
 #include <vector>
 
 #include <QMap>

mythtv/libs/libmythbase/stringutil.h

@@ -5,6 +5,9 @@
 #include <format>
 #endif
 
+#include <string_view>
+#include <vector>
+
 #include <QByteArray>
 #include <QString>
 
@@ -60,6 +63,32 @@ inline bool naturalSortCompare(const QString &a, const QString &b,
 MBASE_PUBLIC QString formatKBytes(int64_t sizeKB, int prec=1);
 MBASE_PUBLIC QString formatBytes(int64_t sizeB, int prec=1);
 
+/**
+Split a `std::string_view` into a `std::vector` of `std::string_view`s.
+
+@param s String to split, may be empty.
+@param delimiter String to determine where to split.
+@return Will always have a size >= 1.  Only valid as long as the data
+        referenced by s remains valid.
+*/
+inline std::vector<std::string_view> split_sv(const std::string_view s, const std::string_view delimiter)
+{
+    // There are infinitely many empty strings at each position, avoid infinite loop
+    if (delimiter.empty())
+        return {s};
+    std::vector<std::string_view> tokens;
+    size_t last_pos = 0;
+    size_t pos = 0;
+    do
+    {
+        pos = s.find(delimiter, last_pos);
+        tokens.emplace_back(s.substr(last_pos, pos - last_pos));
+        last_pos = pos + delimiter.size();
+    }
+    while (pos != std::string_view::npos);
+    return tokens;
+}
+
 } // namespace StringUtil
 
 #endif // STRINGUTIL_H_

mythtv/libs/libmythtv/decoders/avformatdecoder.cpp

@@ -108,6 +108,8 @@ enum V4L2_MPEG_LINE_TYPES : std::uint8_t {
 #include "mythvideoprofile.h"
 #include "remoteencoder.h"
 
+using namespace std::string_view_literals;
+
 #define LOC QString("AFD: ")
 
 // Maximum number of sequential invalid data packet errors before we try
@@ -1738,57 +1740,51 @@ void AvFormatDecoder::ScanTeletextCaptions(int av_index)
     if (!m_tracks[kTrackTypeTeletextCaptions].empty())
         return;
 
-    MythAVBufferRef pmt_buffer {get_pmt_section_for_AVStream_index(m_ic, av_index)};
-    if (!pmt_buffer.has_buffer())
+    AVStream* st = m_ic->streams[av_index];
+    const AVDictionaryEntry* language_dictionary_entry =
+        av_dict_get(st->metadata, "language", nullptr, 0);
+
+    if (language_dictionary_entry == nullptr ||
+        language_dictionary_entry->value == nullptr ||
+        st->codecpar->extradata == nullptr
+       )
     {
         return;
     }
-    const ProgramMapTable pmt(PSIPTable(pmt_buffer.data()));
-
-    for (uint i = 0; i < pmt.StreamCount(); i++)
-    {
-        if (pmt.StreamType(i) != StreamID::PrivData)
-            continue;
 
-        const desc_list_t desc_list = MPEGDescriptor::ParseOnlyInclude(
-            pmt.StreamInfo(i), pmt.StreamInfoLength(i),
-            DescriptorID::teletext);
+    std::vector<std::string_view> languages {StringUtil::split_sv(language_dictionary_entry->value, ","sv)};
 
-        for (const auto *desc : desc_list)
+    if (st->codecpar->extradata_size != static_cast<int>(languages.size() * 2))
+    {
+        return;
+    }
+    for (size_t i = 0; i < languages.size(); i++)
+    {
+        if (languages[i].size() != 3)
         {
-            const TeletextDescriptor td(desc);
-            if (!td.IsValid())
-                continue;
-            for (uint k = 0; k < td.StreamCount(); k++)
-            {
-                int type = td.TeletextType(k);
-                int language = td.CanonicalLanguageKey(k);
-                uint magazine = td.TeletextMagazineNum(k);
-                if (magazine == 0)
-                    magazine = 8;
-                uint pagenum  = td.TeletextPageNum(k);
-                uint lang_idx = (magazine << 8) | pagenum;
-                StreamInfo si {av_index, 0, language, lang_idx};
-                if (type == 2 || type == 1)
-                {
-                    TrackType track = (type == 2) ? kTrackTypeTeletextCaptions :
-                                                    kTrackTypeTeletextMenu;
-                    m_tracks[track].push_back(si);
-                    LOG(VB_PLAYBACK, LOG_INFO, LOC +
-                        QString("Teletext stream #%1 (%2) is in the %3 language"
-                                " on page %4 %5.")
-                            .arg(QString::number(k),
-                                 (type == 2) ? "Caption" : "Menu",
-                                 iso639_key_toName(language),
-                                 QString::number(magazine),
-                                 QString::number(pagenum)));
-                }
-            }
+            continue;
+        }
+        int language = iso639_str3_to_key(languages[i].data());
+        uint8_t teletext_type = st->codecpar->extradata[i * 2] >> 3;
+        uint8_t teletext_magazine_number = st->codecpar->extradata[i * 2] & 0x7;
+        if (teletext_magazine_number == 0)
+            teletext_magazine_number = 8;
+        uint8_t teletext_page_number = st->codecpar->extradata[i * 2 + 1];
+        if (teletext_type == 2 || teletext_type == 1)
+        {
+            TrackType track = (teletext_type == 2) ?
+                                kTrackTypeTeletextCaptions :
+                                kTrackTypeTeletextMenu;
+            m_tracks[track].emplace_back(av_index, 0, language,
+                (static_cast<unsigned>(teletext_magazine_number) << 8) | teletext_page_number);
+            LOG(VB_PLAYBACK, LOG_INFO, LOC +
+                QString("Teletext stream #%1 (%2) is in the %3 language on page %4 %5.")
+                    .arg(QString::number(i),
+                         (teletext_type == 2) ? "Caption" : "Menu",
+                         iso639_key_toName(language),
+                         QString::number(teletext_magazine_number),
+                         QString::number(teletext_page_number)));
         }
-
-        // Assume there is only one multiplexed teletext stream in PMT..
-        if (!m_tracks[kTrackTypeTeletextCaptions].empty())
-            break;
     }
 }
 
@@ -2241,7 +2237,8 @@ int AvFormatDecoder::ScanStreams(bool novideo)
             }
             case AVMEDIA_TYPE_DATA:
             {
-                ScanTeletextCaptions(static_cast<int>(strm));
+                if (par->codec_id == AV_CODEC_ID_DVB_VBI)
+                    ScanTeletextCaptions(static_cast<int>(strm));
                 m_bitrate += par->bit_rate;
                 LOG(VB_PLAYBACK, LOG_INFO, LOC + QString("data codec (%1)")
                         .arg(AVMediaTypeToString(par->codec_type)));