Re: PRD parser expontentially slower with larger files

>>>>> "Jan" == Jan Sundberg <[email protected]> writes:

Jan> However, it gets very slow with larger files. If I generate data files with adjustable size with:

It's a known problem, unfixable without breaking something, unless
you gave a pragma to stop supporting some of the features.

The problem is that PRD uses "nibbling" to scan the string, which looks good
on paper, but doesn't scale when you start getting larger data.  The proper
approach ("inchworming") wasn't available when Damian started writing PRD
(perl 5.003), so we're stuck with the current implementation.

See <http://www.stonehenge.com/merlyn/UnixReview/col55.html> for
a better description of the terms.

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<[email protected]> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!