On Feb 15, Russ Foster said: >> As of Perl 5.6.2 (not released yet), /^(.*?):/ and >> /^([^:]*):/ will have the same efficiency (read: speed). If >> you're curious, currently /^.*?:/ and /^[^:]*:/ have the same >> speed -- it's the capturing that killed .*?, but I have fixed that. > >Just to be clear, it's the capturing (.*?) of the string that suffers the >most (performance-wise), correct? What do you mean you have it fixed--that >future versions of Perl don't suffer a hit?
It's not so much the ACTUAL capturing. Capturing /(.*?):/ and capturing /([^:]*):/ requires the same effort. What killed /(.*?):/ is that normally, Perl's regex engine would optimize /.*?:/ to make .*? jump from one colon to the next in the string. However, because of the internal representation of /(.*?):/, it couldn't do that optimization, and ..*? would move one character at a time, instead of the optimized jump. I made it look a little hard for that optimization. >> And personally, I'd use /([^:]*)/ instead of /^([^:]*):/, >> since they match the same thing (assuming there IS a colon in >> the string). Or I'd use /(.*?):/ instead of /^(.*?):/ but whatever. > >What's the reason for preferring not to use the '^' ? Is it speed? Or just a >desire to make a regex a simple as possible? Personal style. Regexes are already crufty enough -- if I can reduce the amount of content in the regex itself, that's good for me. -- Jeff "japhy" Pinyan [EMAIL PROTECTED] http://www.pobox.com/~japhy/ RPI Acacia brother #734 http://www.perlmonks.org/ http://www.cpan.org/ ** Look for "Regular Expressions in Perl" published by Manning, in 2002 ** <stu> what does y/// stand for? <tenderpuss> why, yansliterate of course. [ I'm looking for programming work. If you like my work, let me know. ] -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
