[FFmpeg-devel] [PATCH 1/2] aadec: improve eof detection

[Michael Niedermayer]

On Sun, Jan 28, 2018 at 12:01:13AM +0100, Karsten Otto wrote:
> Remember the end position of audio content in the file and check it during
> read_packet. There always seems to be other data beyond it, which could be
> misinterpreted as more audio. Also add some extra avio_read error checks,
> to bail early in case of a broken/truncated file.
> ---
>  libavformat/aadec.c | 11 +++++++++--
>  1 file changed, 9 insertions(+), 2 deletions(-)
> 
> diff --git a/libavformat/aadec.c b/libavformat/aadec.c
> index 8d39b1d9ba..d6b6c125da 100644
> --- a/libavformat/aadec.c
> +++ b/libavformat/aadec.c
> @@ -46,6 +46,7 @@ typedef struct AADemuxContext {
>      struct AVTEA *tea_ctx;
>      uint8_t file_key[16];
>      int64_t current_chapter_size;
> +    int64_t content_end;
>  } AADemuxContext;
>  
>  static int get_second_size(char *codec_name)
> @@ -197,6 +198,7 @@ static int aa_read_header(AVFormatContext *s)
>      }
>      start = TOC[largest_idx].offset;
>      avio_seek(pb, start, SEEK_SET);
> +    c->content_end = start + largest_size;
>      c->current_chapter_size = 0;
>  
>      return 0;
> @@ -214,6 +216,11 @@ static int aa_read_packet(AVFormatContext *s, AVPacket *pkt)
>      int ret;
>      AADemuxContext *c = s->priv_data;
>  
> +    // are we at the end of the audio content?
> +    if (avio_tell(s->pb) >= c->content_end) {
> +        return AVERROR_EOF;
> +    }
> +
>      // are we at the start of a chapter?
>      if (c->current_chapter_size == 0) {
>          c->current_chapter_size = avio_rb32(s->pb);
> @@ -234,7 +241,7 @@ static int aa_read_packet(AVFormatContext *s, AVPacket *pkt)
>      // decrypt c->current_codec_second_size bytes
>      blocks = c->current_codec_second_size / TEA_BLOCK_SIZE;
>      for (i = 0; i < blocks; i++) {
> -        avio_read(s->pb, src, TEA_BLOCK_SIZE);
> +        if (avio_read(s->pb, src, TEA_BLOCK_SIZE) < 0) return AVERROR_EOF;

I suspect this should forward the error code or the test should be
for != TEA_BLOCK_SIZE 


[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Does the universe only have a finite lifespan? No, its going to go on
forever, its just that you wont like living in it. -- Hiranya Peiri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: not available
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20180128/e17a4822/attachment.sig>