[FFmpeg-devel,v5,3/3] aadec: improve seeking in mp3 content

Submitted by Karsten Otto on July 11, 2018, 9:53 p.m.

Details

Message ID 20180711215325.84466-1-ottoka@posteo.de
State New
Headers show

Commit Message

Karsten Otto July 11, 2018, 9:53 p.m.
MP3 frames may not be aligned to aa chunk boundaries. When seeking,
calculate the expected frame offset in the target chunk. Adjust the
timestamp and truncate the next packet accordingly.

This solution works for the majority of tested audio material. For
some rare encodings with mp3 padding or embedded id3 tags, it will
mispredict the correct offset, and at worst skip an extra frame.
---
 libavformat/aadec.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

Comments

Michael Niedermayer July 12, 2018, 11:48 p.m.
On Wed, Jul 11, 2018 at 11:53:25PM +0200, Karsten Otto wrote:
> MP3 frames may not be aligned to aa chunk boundaries. When seeking,
> calculate the expected frame offset in the target chunk. Adjust the
> timestamp and truncate the next packet accordingly.
> 
> This solution works for the majority of tested audio material. For
> some rare encodings with mp3 padding or embedded id3 tags, it will
> mispredict the correct offset, and at worst skip an extra frame.
> ---
>  libavformat/aadec.c | 17 ++++++++++++++---
>  1 file changed, 14 insertions(+), 3 deletions(-)
> 
> diff --git a/libavformat/aadec.c b/libavformat/aadec.c
> index 17ad20686b..2d78def6e8 100644
> --- a/libavformat/aadec.c
> +++ b/libavformat/aadec.c
> @@ -37,6 +37,7 @@
>  #define TEA_BLOCK_SIZE 8
>  #define CHAPTER_HEADER_SIZE 8
>  #define TIMEPREC 1000
> +#define MP3_FRAME_SIZE 104
>  
>  typedef struct AADemuxContext {
>      AVClass *class;
> @@ -50,6 +51,7 @@ typedef struct AADemuxContext {
>      int64_t current_chapter_size;
>      int64_t content_start;
>      int64_t content_end;
> +    int seek_offset;
>  } AADemuxContext;
>  
>  static int get_second_size(char *codec_name)
> @@ -228,6 +230,7 @@ static int aa_read_header(AVFormatContext *s)
>      ff_update_cur_dts(s, st, 0);
>      avio_seek(pb, start, SEEK_SET);
>      c->current_chapter_size = 0;
> +    c->seek_offset = 0;
>  
>      return 0;
>  }
> @@ -266,6 +269,8 @@ static int aa_read_packet(AVFormatContext *s, AVPacket *pkt)
>      // is this the last block in this chapter?
>      if (c->current_chapter_size / c->current_codec_second_size == 0) {
>          c->current_codec_second_size = c->current_chapter_size % c->current_codec_second_size;
> +        if (c->seek_offset > c->current_codec_second_size)
> +            c->seek_offset = 0;
>      }
>  
>      // decrypt c->current_codec_second_size bytes
> @@ -293,12 +298,13 @@ static int aa_read_packet(AVFormatContext *s, AVPacket *pkt)
>      if (c->current_chapter_size <= 0)
>          c->current_chapter_size = 0;
>  
> -    ret = av_new_packet(pkt, written);
> +    ret = av_new_packet(pkt, written - c->seek_offset);

Please document how the existing checks gurantee that written >= c->seek_offset


thanks

Patch hide | download patch | download mbox

diff --git a/libavformat/aadec.c b/libavformat/aadec.c
index 17ad20686b..2d78def6e8 100644
--- a/libavformat/aadec.c
+++ b/libavformat/aadec.c
@@ -37,6 +37,7 @@ 
 #define TEA_BLOCK_SIZE 8
 #define CHAPTER_HEADER_SIZE 8
 #define TIMEPREC 1000
+#define MP3_FRAME_SIZE 104
 
 typedef struct AADemuxContext {
     AVClass *class;
@@ -50,6 +51,7 @@  typedef struct AADemuxContext {
     int64_t current_chapter_size;
     int64_t content_start;
     int64_t content_end;
+    int seek_offset;
 } AADemuxContext;
 
 static int get_second_size(char *codec_name)
@@ -228,6 +230,7 @@  static int aa_read_header(AVFormatContext *s)
     ff_update_cur_dts(s, st, 0);
     avio_seek(pb, start, SEEK_SET);
     c->current_chapter_size = 0;
+    c->seek_offset = 0;
 
     return 0;
 }
@@ -266,6 +269,8 @@  static int aa_read_packet(AVFormatContext *s, AVPacket *pkt)
     // is this the last block in this chapter?
     if (c->current_chapter_size / c->current_codec_second_size == 0) {
         c->current_codec_second_size = c->current_chapter_size % c->current_codec_second_size;
+        if (c->seek_offset > c->current_codec_second_size)
+            c->seek_offset = 0;
     }
 
     // decrypt c->current_codec_second_size bytes
@@ -293,12 +298,13 @@  static int aa_read_packet(AVFormatContext *s, AVPacket *pkt)
     if (c->current_chapter_size <= 0)
         c->current_chapter_size = 0;
 
-    ret = av_new_packet(pkt, written);
+    ret = av_new_packet(pkt, written - c->seek_offset);
     if (ret < 0)
         return ret;
-    memcpy(pkt->data, buf, written);
+    memcpy(pkt->data, buf + c->seek_offset, written - c->seek_offset);
     pkt->pos = pos;
 
+    c->seek_offset = 0;
     return 0;
 }
 
@@ -342,7 +348,12 @@  static int aa_read_seek(AVFormatContext *s,
     c->current_chapter_size = chapter_size - chapter_pos;
     c->chapter_idx = 1 + chapter_idx;
 
-    ff_update_cur_dts(s, s->streams[0], ch->start + chapter_pos * TIMEPREC);
+    // handle extra offset for unaligned frames
+    if (s->streams[0]->codecpar->codec_id == AV_CODEC_ID_MP3) {
+        c->seek_offset = (MP3_FRAME_SIZE - chapter_pos % MP3_FRAME_SIZE) % MP3_FRAME_SIZE;
+    }
+
+    ff_update_cur_dts(s, s->streams[0], ch->start + (chapter_pos + c->seek_offset) * TIMEPREC);
 
     return 1;
 }