I went ahead and implemented Perl Compatable Regular Expression support as well as a module which uses template haskell to check regular expressions at runtime. The full set of modules is RRegex - PCRE if available, else Posix, (compatable with Text.Regex) RRegex.PCRE - Perl compatable regular expressions RRegex.Posix - posix regular extensions (compatable with Text.Regex.Posix) RRegex.TH - template haskell compile-time checked regular expressions RRegex.Syntax - perl like overloaded matching syntax Along the way, I extende the Posix interface, as the one in Text.Regex.Posix was lacking in several key ways. among the more interesting changes are ability to use (?options) syntax with posix expressions all of PCRE goodness. which is a lot. much better error messages (compile-time when using TH) example:
Fail: user error (PCRE Regular Expression Error: [a-z_A-Z]+\(.*); ^ unmatched parentheses) notice it points out the exact spot of the error. ability to pull out offset,length integer pairs, rather than extracting the substrings all the =~ stuff mentioned in my previous email.
see the sort-of homepage here: http://haskell.org/hawiki/RegexSyntax and haddock documentation here: http://repetae.net/john/computer/haskell/hsregex/hsregex/docs/index.html John -- --------------------------------------------------------------------------- John Meacham - California Institute of Technology, Alum. - john@foo.net ---------------------------------------------------------------------------