[FFmpeg-devel] [PATCH] Mark AVFilterPad[] compound literals as const.

[Stefano Sabatini]

On date Saturday 2011-11-05 14:59:24 +0100, Reimar Döffinger encoded:
> GCC 4.6.2 at least still seems to fail to put them in .rodata though.
> 
> Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
> ---
>  libavfilter/af_aconvert.c     |    4 ++--
>  libavfilter/af_aformat.c      |    4 ++--
>  libavfilter/af_anull.c        |    4 ++--
>  libavfilter/af_aresample.c    |    4 ++--
>  libavfilter/af_ashowinfo.c    |    4 ++--
>  libavfilter/af_earwax.c       |    4 ++--
>  libavfilter/af_volume.c       |    4 ++--
>  libavfilter/asink_anullsink.c |    4 ++--
>  libavfilter/asrc_abuffer.c    |    4 ++--
>  libavfilter/asrc_aevalsrc.c   |    4 ++--
>  libavfilter/asrc_anullsrc.c   |    4 ++--
>  libavfilter/sink_buffer.c     |    8 ++++----
>  libavfilter/src_movie.c       |    8 ++++----
>  libavfilter/vf_aspect.c       |    8 ++++----
>  libavfilter/vf_blackframe.c   |    4 ++--
>  libavfilter/vf_boxblur.c      |    4 ++--
>  libavfilter/vf_copy.c         |    4 ++--
>  libavfilter/vf_crop.c         |    4 ++--
>  libavfilter/vf_cropdetect.c   |    4 ++--
>  libavfilter/vf_delogo.c       |    6 +++---
>  libavfilter/vf_deshake.c      |    4 ++--
>  libavfilter/vf_drawbox.c      |    4 ++--
>  libavfilter/vf_drawtext.c     |    4 ++--
>  libavfilter/vf_fade.c         |    4 ++--
>  libavfilter/vf_fieldorder.c   |    4 ++--
>  libavfilter/vf_fifo.c         |    4 ++--
>  libavfilter/vf_format.c       |    8 ++++----
>  libavfilter/vf_frei0r.c       |    8 ++++----
>  libavfilter/vf_gradfun.c      |    4 ++--
>  libavfilter/vf_hflip.c        |    4 ++--
>  libavfilter/vf_hqdn3d.c       |    4 ++--
>  libavfilter/vf_libopencv.c    |    4 ++--
>  libavfilter/vf_lut.c          |    6 +++---
>  libavfilter/vf_mp.c           |    4 ++--
>  libavfilter/vf_null.c         |    4 ++--
>  libavfilter/vf_overlay.c      |    4 ++--
>  libavfilter/vf_pad.c          |    4 ++--
>  libavfilter/vf_pixdesctest.c  |    4 ++--
>  libavfilter/vf_scale.c        |    4 ++--
>  libavfilter/vf_select.c       |    4 ++--
>  libavfilter/vf_setpts.c       |    4 ++--
>  libavfilter/vf_settb.c        |    4 ++--
>  libavfilter/vf_showinfo.c     |    4 ++--
>  libavfilter/vf_slicify.c      |    4 ++--
>  libavfilter/vf_split.c        |    4 ++--
>  libavfilter/vf_transpose.c    |    4 ++--
>  libavfilter/vf_unsharp.c      |    4 ++--
>  libavfilter/vf_vflip.c        |    4 ++--
>  libavfilter/vf_yadif.c        |    4 ++--
>  libavfilter/vsink_nullsink.c  |    4 ++--
>  libavfilter/vsrc_buffer.c     |    4 ++--
>  libavfilter/vsrc_color.c      |    4 ++--
>  libavfilter/vsrc_mptestsrc.c  |    4 ++--
>  libavfilter/vsrc_testsrc.c    |   12 ++++++------
>  54 files changed, 124 insertions(+), 124 deletions(-)
> 
> diff --git a/libavfilter/af_aconvert.c b/libavfilter/af_aconvert.c
> index fef096c..fc3dc8d 100644
> --- a/libavfilter/af_aconvert.c
> +++ b/libavfilter/af_aconvert.c
> @@ -406,12 +406,12 @@ AVFilter avfilter_af_aconvert = {
>      .uninit        = uninit,
>      .query_formats = query_formats,
>  
> -    .inputs    = (AVFilterPad[]) {{ .name            = "default",
> +    .inputs    = (const AVFilterPad[]) {{ .name            = "default",

maybe align the "=" like this:

> -    .inputs    = (AVFilterPad[]) {{ .name            = "default",
> +    .inputs    = (const AVFilterPad[]) {{ .name      = "default",

if you mind, push that otherwise if you verified that it works and
doesn't cause regressions.

And yes I agree putting .name on a dedicated line and reducing indent
may be better, but that's bikeshed so it shouldn't block this patch.
-- 
FFmpeg = Funny and Fundamentalist Monstrous Puritan Exciting Game