[FFmpeg-cvslog] matroskadec: introduce resync function.

[Sean McGovern]

ffmpeg | branch: master | Sean McGovern <gseanmcg at gmail.com> | Mon May 27 18:11:50 2013 -0400| [8835c554ff506992c47f6e347c74216ae073f0fa] | committer: Anton Khirnov

matroskadec: introduce resync function.

This allows handling matroska files with errors.
Fixes test4.mkv and test7.mkv from the official Matroska test suite,
and by extension Bugzilla #62.

Based on a patch by Reimar Doffinger <Reimar.Doeffinger at gmx.de>

Signed-off-by: Anton Khirnov <anton at khirnov.net>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=8835c554ff506992c47f6e347c74216ae073f0fa
---

 libavformat/matroskadec.c |   47 +++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 43 insertions(+), 4 deletions(-)

diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c
index d88b34f..9b116b0 100644
--- a/libavformat/matroskadec.c
+++ b/libavformat/matroskadec.c
@@ -557,6 +557,35 @@ static EbmlSyntax matroska_clusters_incremental[] = {
 
 static const char *const matroska_doctypes[] = { "matroska", "webm" };
 
+static int matroska_resync(MatroskaDemuxContext *matroska, int64_t last_pos)
+{
+    AVIOContext *pb = matroska->ctx->pb;
+    uint32_t id;
+    matroska->current_id = 0;
+    matroska->num_levels = 0;
+
+    /* seek to next position to resync from */
+    if (avio_seek(pb, last_pos + 1, SEEK_SET) < 0)
+        goto eof;
+
+    id = avio_rb32(pb);
+
+    // try to find a toplevel element
+    while (!pb->eof_reached) {
+        if (id == MATROSKA_ID_INFO     || id == MATROSKA_ID_TRACKS      ||
+            id == MATROSKA_ID_CUES     || id == MATROSKA_ID_TAGS        ||
+            id == MATROSKA_ID_SEEKHEAD || id == MATROSKA_ID_ATTACHMENTS ||
+            id == MATROSKA_ID_CLUSTER  || id == MATROSKA_ID_CHAPTERS) {
+                matroska->current_id = id;
+                return 0;
+        }
+        id = (id << 8) | avio_r8(pb);
+    }
+eof:
+    matroska->done = 1;
+    return AVERROR_EOF;
+}
+
 /*
  * Return: Whether we reached the end of a level in the hierarchy or not.
  */
@@ -1362,6 +1391,7 @@ static int matroska_read_header(AVFormatContext *s)
     MatroskaChapter *chapters;
     MatroskaTrack *tracks;
     uint64_t max_start = 0;
+    int64_t pos;
     Ebml ebml = { 0 };
     AVStream *st;
     int i, j, res;
@@ -1392,8 +1422,16 @@ static int matroska_read_header(AVFormatContext *s)
     ebml_free(ebml_syntax, &ebml);
 
     /* The next thing is a segment. */
-    if ((res = ebml_parse(matroska, matroska_segments, matroska)) < 0)
-        return res;
+    pos = avio_tell(matroska->ctx->pb);
+    res = ebml_parse(matroska, matroska_segments, matroska);
+    // try resyncing until we find a EBML_STOP type element.
+    while (res != 1) {
+        res = matroska_resync(matroska, pos);
+        if (res < 0)
+            return res;
+        pos = avio_tell(matroska->ctx->pb);
+        res = ebml_parse(matroska, matroska_segment, matroska);
+    }
     matroska_execute_seekhead(matroska);
 
     if (!matroska->time_scale)
@@ -2286,7 +2324,6 @@ static int matroska_parse_cluster(MatroskaDemuxContext *matroska)
                                      pos);
         }
     ebml_free(matroska_cluster, &cluster);
-    if (res < 0)  matroska->done = 1;
     return res;
 }
 
@@ -2296,9 +2333,11 @@ static int matroska_read_packet(AVFormatContext *s, AVPacket *pkt)
     int ret = 0;
 
     while (!ret && matroska_deliver_packet(matroska, pkt)) {
+        int64_t pos = avio_tell(matroska->ctx->pb);
         if (matroska->done)
             return AVERROR_EOF;
-        ret = matroska_parse_cluster(matroska);
+        if (matroska_parse_cluster(matroska) < 0)
+            ret = matroska_resync(matroska, pos);
     }
 
     if (ret == AVERROR_INVALIDDATA && pkt->data) {