[boost] regex guinea pigs wanted

From: John Maddock (jm_at_[hidden])
Date: 2003-04-08 06:44:37

Folks,

I'm about ready to release a new version of boost regex - to date there are
three main changes:

1) A brand new search/match algorithm implementation that's up to an order
of magnitude faster than the old one in some cases.  Performance should now
be on a par with or better than PCRE, GRETA, and GNU regex in most cases -
see for example
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/boost/boost/libs/r
egex/performance/Attic/vc71-results.htm?rev=1.1.2.1 although please note
that this needs updating.

2) Matching rules follow perl when the expression is compiled as a perl
expression (the default), and POSIX when compiled as a POSIX extended or
basic regex.

3) The only new feature added so far is support for independent
sub-expressions, more will follow in due course.

I'm looking for a few heavy regex users to check this out - the code is in
cvs under a branch called "regex-4" - which you will need to check out into
a fresh directory (not on top of an existing cvs working copy).

Still to do before release:

Update the docs (badly needs doing).
Add regex iterators.

Many thanks.

John Maddock.