On Tue, Oct 08, 2019 at 01:55:27PM +0200, Moritz Barsnick wrote: > On Tue, Oct 08, 2019 at 11:32:09 +0200, Michael Niedermayer wrote: > > > A side effect is that this coincidentally breaks enforcement of > > > strict MIME headers against servers running motion < 3.4.1, where > > > the boundary announcement in the HTTP headers incorrectly used the > > > prefix "--", which exactly matched this bug's behavior. > > > > Can this be worked around, by for example detecting that case and > > supporting it ? > > Or maybe i misunderstand > > It could, but should we really? > - Strict MIME enforcement never really worked upto now, due to a > programming bug (possibly a misunderstanding of the spec) in ffmpeg. > This makes me think that hardly anyone ever even used > -strict_mime_boundary. > - It would now, after this patch, happen to fail with a server which > implements the same misinterpretation of the spec. >
> If you want to support both interpretations, you'd need to search for > two different boundaries now. Quite some bloat, if you ask me. I don't > know if it's worth the effort, i.e. whether this will even hit anyone, > as "-strict_mime_boundary" is off by default. I hoped that this server version can easily be detected so only one search is needed > > I vote against, but I probably don't decide. I abstain from voting, was more a thought that more stuff working out of the box wouldnt be bad thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Avoid a single point of failure, be that a person or equipment.
signature.asc
Description: PGP signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
