Re^4: Parse::RecDescent and need for speed

Mis-remembered. It's about pos(), but not pos(). What got added in 5.005 was m//gc, so a failed match didn't reset pos().

Comment on Re^4: Parse::RecDescent and need for speed

Replies are listed 'Best First'.
Re^5: Parse::RecDescent and need for speed by Anonymous Monk on Feb 16, 2005 at 14:41 UTC
Fair enough. Too bad Damian didn't realise that pos() was assignable, even with 5.004.	[reply]
Re^6: Parse::RecDescent and need for speed by merlyn (Sage) on Feb 16, 2005 at 15:04 UTC
Uh, I'm absolutely sure he realized that. The problem is the awkwardness of having to put a "save pos, see if this works, no? then restore pos" junk around every match, while still retaining the scalar-ness or list-ness of the context in which it's happening. This is why /c was added! -- Randal L. Schwartz, Perl hacker Be sure to read my standard disclaimer if this is a reply.	[reply]