[mod_python] mysqldb calls cause segfault

Hi Nicolas,

Thanks for your help, I've added one line to your script to see if the
interpreter uses 2 or 4-byte wide unicode strings (the root of the problem).

the script:

import sys
def handler(req):
     req.content_type = 'text/plain'
     req.write(sys.version+'\n')
     req.write(sys.exec_prefix+'\n')
     req.write(`sys.maxunicode`+'\n')
     return apache.OK


the output:

2.3.4 (#1, Dec 30 2004, 12:39:04)
[GCC 3.3.5 (Debian 1:3.3.5-5)]
/usr
65535

So there it is, black & white, it uses 2-byte wide Unicode chars, that
causes the trouble, but how do I track down which python interpreter is
called? To my knowledge, python >=2.3 uses 4-byte wide unicode, so how
do I track down which python interpreter is used? (or maybe more simple
question is how do I force the mod_python/mysqldb extension to use the
"good" python interpreter)?

Balint

Nicolas Lehuen wrote:
> Are you sure mod_python is using the same Python interpreter as the one 
> you made the test in ?
> 
> Try running this handler :
> 
> from mod_python import apache
> import sys
> def handler(req):
>     req.content_type = 'text/plain'
>     req.write(sys.version+'\n')
>     req.write(sys.exec_prefix+'\n')
>     return apache.OK
> 
> Regards,
> Nicolas
> 
> 2005/10/9, Erdi Balint <erdibalint at freemail.hu 
> <mailto:erdibalint at freemail.hu>>:
> 
>     Thank you, Nicolas, that fixed that problem, but now I have another one,
>     when importing a module from the MySQLdb module, I get the following
>     exception:
> 
>        File "/var/www/trivia/control/trivia.py", line 5, in _connect
>          db = MySQLdb.connect(user="xxx",passwd="xxx",db="xxx")
> 
>        File "/usr/lib/python2.3/site-packages/MySQLdb/__init__.py", line 66,
>     in Connect
>          return Connection(*args, **kwargs)
> 
>        File "/usr/lib/python2.3/site-packages/MySQLdb/connections.py", line
>     117, in __init__
>          from converters import conversions
> 
>        File "/usr/lib/python2.3/site-packages/MySQLdb/converters.py", line
>     40, in ?
>          import array
> 
>     ImportError: /usr/lib/python2.3/lib-dynload/array.so: undefined symbol:
>     PyUnicodeUCS4_FromUnicode
> 
>     I've come across a python faq item, that says the following about the
>     cause of the error:
> 
>     "...
>     If instead the name of the undefined symbol starts with PyUnicodeUCS4,
>     the problem is the reverse: Python was built using 2-byte Unicode
>     characters, and the extension module was compiled using a Python with
>     4-byte Unicode characters.
>     ...
> 
>     You can check the size of the Unicode character a Python interpreter is
>     using by checking the value of sys.maxunicode:
> 
>      >>> import sys
>      >>> if sys.maxunicode > 65535:
>     ...     print 'UCS4 build'
>     ... else:
>     ...     print 'UCS2 build'
> 
>     "
>     Ok, so in theory, my Python interpreter uses 2-byte Unicode characters,
>     but I've checked it (I am python 2.3.5) with the above code snipplet,
>     and it is not so, I am using 4-byte Unicode characters? Then why is the
>     error?
> 
>     Thank you for your help,
>     Balint