[FFmpeg-devel] [PATCH] [issue 566] return error for rgb 16-bit pnm files
Justin Ruggles
justin.ruggles
Sun Sep 14 17:25:06 CEST 2008
Justin Ruggles wrote:
> Michael Niedermayer wrote:
>> On Sun, Sep 14, 2008 at 10:53:44AM -0400, Justin Ruggles wrote:
>>> Hi,
>>>
>>> Patch attached to resolve issue 566. It would be nice to implement
>>> PIX_FMT_RGB48/64, but for now we at least need an error instead of
>>> silently generating bad output.
>>>
>>> Thanks,
>>> Justin
>>> Index: libavcodec/pnm.c
>>> ===================================================================
>>> --- libavcodec/pnm.c (revision 15319)
>>> +++ libavcodec/pnm.c (working copy)
>>> @@ -106,7 +106,13 @@
>>> else
>>> avctx->pix_fmt = PIX_FMT_GRAY8;
>>> } else if (depth == 3) {
>>> + if (maxval < 256) {
>>> avctx->pix_fmt = PIX_FMT_RGB24;
>>> + } else {
>>> + av_log(avctx, AV_LOG_ERROR, "16-bit components are only supported for grayscale\n");
>>> + avctx->pix_fmt = PIX_FMT_NONE;
>>> + return -1;
>>> + }
>>> } else if (depth == 4) {
>>> avctx->pix_fmt = PIX_FMT_RGB32;
>>> } else {
>> ok
>>
>>
>>> @@ -131,6 +137,10 @@
>>> avctx->pix_fmt = PIX_FMT_GRAY16BE;
>>> if (s->maxval != 65535)
>>> avctx->pix_fmt = PIX_FMT_GRAY16;
>>> + } else if(s->maxval >= 256) {
>>> + av_log(avctx, AV_LOG_ERROR, "16-bit components are only supported for grayscale\n");
>>> + avctx->pix_fmt = PIX_FMT_NONE;
>>> + return -1;
>>> }
>> the if(s->maxval >= 256) {
>> can be factored out
>
> New patch attached.
oops... and again...
-Justin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: issue_566_pnm_rgb48.diff
Type: text/x-patch
Size: 1374 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080914/ce2be18e/attachment.bin>
More information about the ffmpeg-devel
mailing list