[FFmpeg-devel] matroskadec: prevent access of elements after freeing

When using the decode interrupt feature of ffmpeg may causing crashes by accessing previous freed pointers in matroska_read_close. The attached patch will reset nb_elem to zero after freeing the elements because ffmpeg normally tests for nb_elem. Feedback for sure is warmly welcome. Regards Mi

[FFmpeg-devel] [PATCH] avformat/matroskadec: Prevent access of elements after freeing by set nb_elem to zero.

Hi folks, retry, because my eMail address was changed. I’m using the decode interrupt feature of ffmpeg for getting a fast user reaction. It looks like this may causing some nasty crashes by accessing previous freed pointers in matroska_read_close. The attached patch will reset nb_elem to zero

Re: [FFmpeg-devel] [PATCH 2/3] hls: get/set protocol options through the AVIOContext

Dear friends, I fully agree with Hendrik on that point and I'm trapping into the same problems Regards Michael -Ursprüngliche Nachricht- Von: ffmpeg-devel [mailto:ffmpeg-devel-boun...@ffmpeg.org] Im Auftrag von Hendrik Leppkes Gesendet: Montag, 14. März 2016 12:23 An: ffmpeg-devel@ffmp

Re: [FFmpeg-devel] [PATCH 1/3] Revert "hls: Add and use a memebr of AVIOInternal rather than abuse opaque"

Dear friends, I fully agree with Hendrik on that point and I'm trapping into the same problems Regards Michael -Ursprüngliche Nachricht- Von: ffmpeg-devel [mailto:ffmpeg-devel-boun...@ffmpeg.org] Im Auftrag von Hendrik Leppkes Gesendet: Montag, 14. März 2016 12:23 An: ffmpeg-devel@ffmp

Re: [FFmpeg-devel] [PATCH 3/3] hls: handle crypto in the protocol checks

Dear friends, from my point of view the patch looks ok. I was just about to submit something similar because it run into the same problem (not working AES encrypted HLS). Regards Michael -Ursprüngliche Nachricht- Von: ffmpeg-devel [mailto:ffmpeg-devel-boun...@ffmpeg.org] Im Auftrag v

[FFmpeg-devel] [PATCH] avformat/http: Add crypto to default whitelist

Dear friends, I think we missed the crypto in the default_whitelist in case of http. Otherwise encrypted HLS will fail with [hls,applehttp @ 0x2af39c00] playlist[0] open_input [http://playertest.longtailvideo.com/adaptive/oceans_aes/oceans_aes-audio=65000-video=236000-14.ts], start_seq_no

[FFmpeg-devel] [PATCH] avformat/http: detect end of received on chunked persistenc

Dear friends, we had a problem with an HLS server which was using chunked http transfer over persistence http connection. Even if we have received all data (chunksize = 0) the read of a m3u8 playlist was blocked till the server tears down the TCP connection. Please find attached a patch for avfor

[FFmpeg-devel] [PATCH] avformat/http: detect end of received on chunked persistence

Dear friends, we had a problem with an HLS server which was using chunked http transfer over persistence http connection. Even if we have received all data (chunksize = 0) the read of a m3u8 playlist was blocked till the server tears down the TCP connection. Please find attached a patch for avfor

[FFmpeg-devel] [PATCH] avformat/hls: Do not iterate to next sequence number if interrupted is requested

Hi, In general it's a good and useful feature iterating to the next sequence if open_input failes. But if the application is using the abort callback possibility we shall be aware of this and returning with a meaningful error code, otherwise we just increases sequence numbers. Feedback for sure