Re: [FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-05 Thread Muhammad Faiz
On Mon, Feb 5, 2018 at 4:29 AM, Nicolas George wrote: > Josh de Kock (2018-02-04): >> If we were to add in APIs which allowed you to register external components >> again, this idea wouldn't work well as indexes wouldn't necessarily >> correspond >> to the component which it

Re: [FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-04 Thread Nicolas George
Josh de Kock (2018-02-04): > The main benefit of the opaque pointers is the flexibility of how components > are stored/represented internally, and being able to change/extend it with > no API changes (only additions). Sure there is the question of 'how efficient > is it?', but it's not really a

Re: [FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-04 Thread wm4
On Sun, 4 Feb 2018 22:29:10 +0100 Nicolas George wrote: > Josh de Kock (2018-02-04): > > If we were to add in APIs which allowed you to register external components > > again, this idea wouldn't work well as indexes wouldn't necessarily > > correspond > > to the component which

Re: [FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-04 Thread Nicolas George
Josh de Kock (2018-02-04): > If we were to add in APIs which allowed you to register external components > again, this idea wouldn't work well as indexes wouldn't necessarily correspond > to the component which it previously did after you register extra components. That is not a problem if the

Re: [FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-04 Thread Josh de Kock
On Sun, Feb 04, 2018 at 02:46:09PM +0100, Nicolas George wrote: > Muhammad Faiz (2018-02-04): > > What about av*iterate(int index)? This makes no sense, it's then not an API for iteration of components. > > I like the idea of an index better than the other options evoked, > especially the

Re: [FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-04 Thread Nicolas George
Muhammad Faiz (2018-02-04): > What about av*iterate(int index)? I like the idea of an index better than the other options evoked, especially the linked-list-like APIs. It is also more similar to the APIs for enumerated types. Another option would be to return the whole list in a mallocated

Re: [FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-03 Thread Muhammad Faiz
On Sat, Feb 3, 2018 at 5:39 PM, wm4 wrote: > On Fri, 2 Feb 2018 19:44:18 + > Josh de Kock wrote: > >> --- >> fftools/cmdutils.c | 2 +- >> libavcodec/avcodec.h | 6 +- >> libavcodec/bitstream_filter.c | 4 ++-- >>

Re: [FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-03 Thread wm4
On Fri, 2 Feb 2018 19:44:18 + Josh de Kock wrote: > --- > fftools/cmdutils.c | 2 +- > libavcodec/avcodec.h | 6 +- > libavcodec/bitstream_filter.c | 4 ++-- > libavcodec/bitstream_filters.c | 29 ++--- > 4 files

[FFmpeg-devel] [PATCH v4 7/7] lavc/bsf: make BSF iteration the same as other iterators

2018-02-02 Thread Josh de Kock
--- fftools/cmdutils.c | 2 +- libavcodec/avcodec.h | 6 +- libavcodec/bitstream_filter.c | 4 ++-- libavcodec/bitstream_filters.c | 29 ++--- 4 files changed, 26 insertions(+), 15 deletions(-) diff --git a/fftools/cmdutils.c