[FFmpeg-devel] [PATCH] Made minor changes to get the decklink avdevice code to build using Visual C++

[Marton Balint]

On Sun, 7 May 2017, Marton Balint wrote:

>
> On Fri, 5 May 2017, Aaron Levinson wrote:
>
>> On 5/5/2017 4:57 PM, Marton Balint wrote:
>>> 
>>> 
>>> On Fri, 5 May 2017, Aaron Levinson wrote:
>>> 
>>>> On 4/16/2017 4:11 PM, Aaron Levinson wrote:
>>>>> On 4/15/2017 6:13 AM, Aaron Levinson wrote:
>>>>>> On 4/15/2017 4:19 AM, Marton Balint wrote:
>>>>>>>
>>>>>>> On Thu, 13 Apr 2017, Aaron Levinson wrote:
>>>>>>>
>>>>>>>> On 4/13/2017 1:23 PM, Hendrik Leppkes wrote:
>>>>>>> [...]
>>>>
>>>> Ping, and in addition, I've provided a new patch below since the
>>>> original won't apply anymore.  I've also adjusted the patch text a
>>>> little bit.  The original patch went through a few different reviews.
>>>> These changes are necessary to get the decklink code to build using
>>>> MSVC on Windows.
>>>>
>>>> Thanks,
>>>> Aaron Levinson
>>>>
>>>> 
> ------------------------------------------------------------------------------------
>>>> [...]
>>>>         case PTS_SRC_WALLCLOCK:
>>>> -            pts = av_rescale_q(wallclock, AV_TIME_BASE_Q, time_base);
>>>> +        {
>>>> +            /* MSVC does not support compound literals like
>>>> AV_TIME_BASE_Q
>>>> +             * in C++ code (compiler error C4576) */
>>>> +            // pts = av_rescale_q(wallclock, AV_TIME_BASE_Q, time_base);
>>> 
>>> I'd rather remove the old code, i see no point in keeping it as a comment.
>>
>> Done--new patch with this change follows at the end.
>>
>>>> [...]
>>> 
>>> The decklink part seems fine otherwise, maybe you should get an ACK from
>>> Hendrik for the configure part, which I have no opinion about.
>>
>> Hendrik already reviewed the configure changes last month, and the 
>> configure part of the patch went through three iterations till it was 
>> determined that it was adequate, although it wouldn't hurt to have 
>> another look since then.
>
> Ok, will apply this tomorrow.

Pushed, thanks.

Marton