Re: [Mysql-cocoa-users] Problem compiling a Universal Binary

[Jeff K. <kop...@gm...>]

If I'm not mistaken, I've always gotten the binary from the MySQL  
folks (mysql.com).  I'm guessing that a Universal Binary will need to  
be compiled to get a Universal Binary libmysqlclient.a.

Sound sane?

I'll give it a whirl, but I'm notoriously bad at getting these things  
to work.  If anyone else has done this before, go for it...no ego to  
bruise here. :)

-Jeff.


On Jun 17, 2005, at 8:36 AM, Aaron Jacobs wrote:

> Of course, perhaps I should have tried this before I sent the e- 
> mail.  I guess I built the wrong target when I tested before.  I'm  
> getting an error, but I don't think it's the same as yours.  My  
> error is because the copy of libmysqlclient I have is built for ppc  
> only, so it won't link it to an i386 binary.
>
> Does anyone know how I could build an i386 copy of libmysqlclient.a?
>
> Aaron
>
>
> On Jun 17, 2005, at 7:29 AM, Aaron Jacobs wrote:
>
>
>> I'm not using it as a framework or a library or anything.  All I  
>> have in my project is the source for the classes I need as well as  
>> the libmysql_client library and the libmysql headers, since it  
>> depends on those.  The only warning I get is "Possible missing  
>> call to [super dealloc]" for MCPNumber.  I've been meaning to fix  
>> that myself, but the MCPNumber class is designed really weirdly  
>> and I don't know if there's a reason for that.
>>
>> Aaron
>>
>>
>> On Jun 16, 2005, at 11:38 PM, Jeff Kopmanis wrote:
>>
>>
>>
>>> I haven't gotten the same errors twice, unfortunately.
>>>
>>> The current one is:
>>> /usr/bin/libtool: can't locate file for: -lmysqlclient.12.0.0
>>> /usr/bin/libtool: file: -lmysqlclient.12.0.0 is not an object  
>>> file (not allowed in a library)
>>>
>>> The last one complained about cclobj not being present for  
>>> gcc-3.3, which was puzzling, since gcc-4.0 is supposed to be the  
>>> current compiler (I checked that with gcc_select).
>>>
>>> Did you have to set any paths or other particulars?  Also, there  
>>> are alot of warnings regarding stuff like bad return types, bad  
>>> typing and such, but nothing that claimed it was going to break  
>>> stuff.
>>>
>>> Oh...I'd assumed that if there are precompiled headers that I'd  
>>> need to recompile those as well...is that true?
>>>
>>> I'll admit my programming skills are a bit rusty, but this really  
>>> seems to be less obvious than it could be.  Serves me right for  
>>> staying in system admin for so long...
>>>
>>> Thanks,
>>>
>>> -Jeff. :)
>

Jeff Kopmanis
kop...@gm...
http://www.kopmanis.com