[FFmpeg-devel] _t POSIX namespace issues

Alex Converse alex.converse
Fri Dec 12 01:36:51 CET 2008


On Thu, Dec 11, 2008 at 3:31 AM, Diego Biurrun <diego at biurrun.de> wrote:

> So names ending in _t are reserved by POSIX.  Nonetheless we have tons
> of them.  I thought about fixing these issues myself, but coming up with
> good new names is tedious, so I propose to have each maintainer address
> the problem in the files they maintain.  Here is a the result of a grep
> for problematic names with maintainers, in no particular order:
>
> [...]
>
> Aurelien Jacobs:
> [...]
> libavcodec/vp56.c:    vp56_frame_t ref_frame;
> libavcodec/vp56data.h:} vp56_frame_t;
> libavcodec/vp56data.h:} vp56_mb_t;
> libavcodec/vp56data.h:} vp56_tree_t;
> libavcodec/vp56.h:typedef int (*vp56_adjust_t)(int v, int t);
> libavcodec/vp56.h:typedef int (*vp56_parse_header_t)(vp56_context_t *s,
> const uint8_t *buf,
> libavcodec/vp56.h:typedef struct vp56_context vp56_context_t;
> libavcodec/vp56.h:typedef struct vp56_mv vp56_mv_t;
> libavcodec/vp56.h:typedef void (*vp56_default_models_init_t)(vp56_context_t
> *s);
> libavcodec/vp56.h:typedef void (*vp56_filter_t)(vp56_context_t *s, uint8_t
> *dst, uint8_t *src,
> libavcodec/vp56.h:typedef void (*vp56_parse_coeff_models_t)(vp56_context_t
> *s);
> libavcodec/vp56.h:typedef void (*vp56_parse_coeff_t)(vp56_context_t *s);
> libavcodec/vp56.h:typedef void
> (*vp56_parse_vector_adjustment_t)(vp56_context_t *s,
> libavcodec/vp56.h:typedef void (*vp56_parse_vector_models_t)(vp56_context_t
> *s);
> libavcodec/vp56.h:} vp56_macroblock_t;
> libavcodec/vp56.h:} vp56_model_t;
> libavcodec/vp56.h:} vp56_range_coder_t;
> libavcodec/vp56.h:} vp56_ref_dc_t;
> libavcodec/vp5.c:    sizeof(vp56_context_t),
> [...]
>

Why did these change like this?:
vp56_context_t -> VP56Context
vp56_frame_t -> VP56Frame
vp56_mv_t -> VP56mv
vp56_mb_t -> VP56mb
vp56_range_coder_t -> VP56RangeCoder
vp56_ref_dc_t -> VP56RefDc

But this change like that?:
vp56_model_t -> Vp56Model

Regards,
Alex Converse




More information about the ffmpeg-devel mailing list