Strange regex problem

[Dagur Páll Ammendrup]

This code is enough to reproduce it for me:

#!/usr/bin/env python

import re

file = ''' <tr>
           <td class='forum1' width='5%' valign='middle'><{F_ACTIVE}></td>
           <td class='forum2' width='95%'><b>$total</b> 
$birth_lang<br>$birthusers</td>
         </tr>
<!--START:skin_boards-->
<!--IBF_START_FUNC|ActiveUsers|$active-->
         <tr>
            <td class='category' colspan='2'>$active[TOTAL] 
{ibf.lang.active_users}</td>
         </tr>
[.. A lot of HTML ..]
<!--IBF_END_FUNC|subheader-->
<!--END:skin_boards-->
<!--IBF_START_FUNC|append_view|$url=""-->
'''

locations = []

p = re.compile('<!--START:(\w*?)-->')
m = p.search(file)
while m:
     begin = m.end()
     match = m.group(1)
     n = re.search('<!--END:%s-->' % match,file,begin)
     m = p.search(file,begin)



If i change the 2n line from the bottom to this:

q = re.compile('<!--END:%s-->' % match)
n = q.search(file,begin)

this problem does not happen. Is there any change in speed because of 
this? Not that it matters much for the script I'm making.

-Dagur







Rene Pijlman wrote:
> Miki Tebeka:
> 
>>OP:
>>
>>>literal 45
>>>literal 73
>>>literal 66
>>>literal 70
>>>literal 95
>>>literal 71
> 
> 
> It looks like debug output. Are you running in some sort of
> debug mode?
> 
> 
>>2. Can you give us a sample input file?
> 
> 
> And some complete code we can run to reproduce the problem.
>