re module - error with non-greedy matches

[Derek Thomson]

Hi Paul,

Paul Wright wrote:

> In article <3C3C4FFA.1040104 at dstc.qut.edu.au>,
> Derek Thomson  <derekt at dstc.qut.edu.au> wrote:
> 
>>I'm having some trouble porting a Perl program to Python. It extracts C 
>>functions from the source and performs some analysis.
>>
> 
> This is peripherally related to your question, but: when I'm doing
> analysis of C, I usually connect Python to Exuberent Ctags
> <http://ctags.sf.net/> via a pipe, rather than trying to parse C with
> regexps.


No, I wouldn't try parsing C with regexes. The above is pretty much all 
the parsing needed.

> (Actually, I usually get ctags to do the hard stuff and then do
> the easy stuff in Python with regexps). I'd recommend this method as a
> workaround.  ctags -x is a useful option, as it produces nicely
> delimited output.


I'm sure this works, but I really need to know that Python regexes 
aren't broken before I can seriously use Python for all my scripting 
needs. I use it for some things (CORBA, GUIs, Java/C/C++ scripting) but 
I'd like to try it as a drop-in replacement for Perl. Yes, I realize 
that some things are done differently in different languages, but the re 
module documentation clearly states that non-greedy matching is 
supported, and it's something I do need on the odd occasion.

Thanks anyway for the help!

Regards,
Derek.