On 09/03/2011 04:52 PM, Anton Khirnov wrote: > > On Sat, 03 Sep 2011 01:24:55 +0200, David Goldwich <[email protected]> > wrote: >> >> The +7 comes from these lines in the original content decoding code >> >>> while (taglen > 1 && q - dst < dstlen - 7) { >> >> so it just ignores the last 7 bytes of dst. To be honest, I have no clue >> why it checks for dstlen-7 here, nor was there any documentation about >> it. I can't see any reason why it should be like this, but I didn't feel >> like breaking something, so I adopted my buffers by making them 7 bytes >> bigger. (The ttag parser uses a constant 512 byte buffer which is way >> longer than most text tag content, so it's not noticeable there most of >> the time). > > The answer lies in libavutil/common.h -- PUT_UTF8 macro writes up to 7 bytes.
Ah, thanks for pointing that out, now it's clear. _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
