[Wrapper-user] Re: [PATCH] Support for Mac OS X

[Leif M. <le...@ta...>]

Andy,
    I got the patch that you supplied committed, but when I went up to the
Sourceforge compile farm to test it out, I found that the OSX host is not
currently available.  Not sure if it is down or gone.  I will have to 
check back
later.
    Are you still having the crash problems that you had mentioned before?
Those will need to get fixed before I can do a release with OSX support.
If the SourceForge OSX host comes back online, I will take a look at it.
Otherwise my hands are tied.  If you do a build with DEBUG enabled the
generated binary will kick out a large amount of output as the server is
being started up.  That might provide some clues as to where the crash
you were seeing is originating.   Otherwise, I always use the low tech
"printf" method of tracking these things down. :-/

You should be able to enable debug output with the following change in
your makefile :
COMPILE = gcc -O3 -Wall -pedantic -D_DEBUG

Cheers,
Leif

Andy Barnett wrote:

> See attached gzip'd tar file containing "patchfile.txt" created with 
> "cvs diff -u" using the latest code from CVS, plus the new 
> "Makefile.macosx" file.
>
> ~Andy
>
>
>
> On Sunday, April 13, 2003, at 09:48  PM, Leif Mortenson wrote:
>
>> Andy,
>>    Go ahead and send the files to me directly.  I'll take a look at 
>> them and try
>> to getting it running on the MasOSX machine running on the SF compile
>> farm.   If I have any Questions I may ask you to try running a test 
>> build for me
>> or something.
>>
>> Cheers,
>> Leif
>>
>> Andy Barnett wrote:
>>
>>> Would you like me to post the patched files to this mailing list or 
>>> to  the Patches section of wrapper.sourceforge.net? Also, is a "cvs 
>>> diff  -u" patchfile sufficient?
>>>
>>> FYI, I still need to put in IFDEFs in the C code to mark my Mac  
>>> changes. I know a lot more Java than I know C, so I'm going to use 
>>> the  advice in this message as my guide for Mac OS X IFDEFs.
>>>      <http://www.omnigroup.com/mailman/archive/macosx-dev/2000-May/ 
>>> 001923.html>
>>>
>>> One last note, I get Segmentation Faults when  
>>> "wrapper.logfile.loglevel" is set to INFO. I found that setting this 
>>> to  DEBUG made the Segmentation Faults go away. Perhaps this points 
>>> to a  bug somewhere. I'll try to reproduce this problem again and 
>>> post the  console output of the SegFault; I've done zero C coding on 
>>> Macs so the  error output didn't help me much.
>>>
>>> Cheers,
>>> ~Andy
>>