Re: [Chicken-hackers] need irregex benchmark

[Jim Ursetto]

If I recall correctly, there was a pretty big performance hit for certain 
regexes that include submatches. I think this may have occurred when we 
upgraded irregex to 0.8. I mentioned it on IRC previously, but I don't know if 
I saved the exact test cases. I will check later today. however, I think this 
was an algorithmic problem and not something solvable with compiler 
optimizations. 

The svnwiki-sxml extension uses regexes heavily. It does not process 
particularly slowly from a human perspective, but you could construct a long 
document. 

On May 22, 2011, at 9:53, Felix <address@hidden> wrote:

> Hello!
> 
> 
> I'm looking for code that stresses irregex heavily, because I'd like
> to test optimizer-improvements and because irregex (being very
> portable and generic code without lowlevel hacks) needs every help the
> compiler can give.
> 
> So far I've used the test-suite, but it executes to fast and doesn't
> give any real indication about performance of irregex core operations.
> 
> Any pointers to code, or heavy-duty regexen that need long to execute
> would be very welcome.
> 
> 
> cheers,
> felix
> 
> _______________________________________________
> Chicken-hackers mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/chicken-hackers