[Ffmpeg-devel] [PATCH] Fix broken compilation on Mac OS X 10.2due to OPT_INT64 support

[Patrice Bensoussan]

On 7 Apr 2007, at 13:01, Patrice Bensoussan wrote:

>
> On 6 Apr 2007, at 19:41, V?ctor Paesa wrote:
>
>> Hi,
>>
>>> On 5 Apr 2007, at 19:28, Baptiste Coudurier wrote:
>>>
>>> Hi,
>>>
>>>> Michael Niedermayer wrote:
>>>>> Hi
>>>>>
>>>>> On Thu, Apr 05, 2007 at 06:56:48PM +0100, Patrice Bensoussan  
>>>>> wrote:
>>>>>> Hello,
>>>>>>
>>>>>> The recent patch to support OPT_INT64 option broke the
>>>>>> compilation on
>>>>>> Mac OS X 10.2. The reason for this is that atoll() is not  
>>>>>> available
>>>>>> on Mac OS X 10.2 (it's only available on 10.3 and later  
>>>>>> versions).
>>>>>>
>>>>>> From the man page:
>>>>>>
>>>>>>      The atoll() function converts the initial portion of the  
>>>>>> string
>>>>>> pointed
>>>>>>      to by nptr to long long integer representation.
>>>>>>
>>>>>>      It is equivalent to:
>>>>>>
>>>>>>            strtoll(nptr, (char **)NULL, 10);
>>>>>>
>>>>>> Simple patch attached to fix the issue.
>>>>>
>>>>> probably ok
>>>>>
>>>>
>>>> my man says:
>>>>
>>>> atoll: SVr4, POSIX.1-2001, 4.3BSD, C99.  C89 and POSIX.1-1996  
>>>> include
>>>> the functions atoi() and atol() only.  atoq(3) is a GNU extension.
>>>>
>>>> strtol() conforms to SVr4, 4.3BSD, C99 and POSIX.1-2001, and  
>>>> strtoll()
>>>> to C99 and POSIX.1-2001.
>>>>
>>>> maybe someone can test on *BSD/Mingw ?
>>>>
>>>
>>> Looks like it should work fine on FreeBSD too (at least 4.3 and  
>>> above):
>>>
>>> http://www.freebsd.org/cgi/man.cgi?
>>> query=strtoll&apropos=0&sektion=0&manpath=FreeBSD+4.3-
>>> RELEASE&format=html
>>>
>>
>> It is defined in the includes for cross-compiling under Cygwin, so  
>> most
>> probably it is a MinGW valid function:
>>
>> $ fgrep strtoll /usr/include/mingw/*
>> /usr/include/mingw/stdlib.h:long long  __cdecl strtoll (const char*
>> __restrict__, char** __restrict, int);
>
> Looks good... anyone willing to apply?
> Patrice


Has this patch been lost? could someone commit the change please? It  
would avoid having to fix it manually when I compile for Mac OS X  
10.2...

Thanks,
Patrice