Incremental parser (was: Backtick Hickup)

[A. Pagaltzis]

* Michel Fortin <michel.fortin at michelf.com> [2007-08-14 16:45]:

> I'm totally not convinced that creating a byte-by-byte parser

> in Perl  or PHP is going to be very useful.


Perl has this handy /c switch for regular expressions that means
“continue” and has the effect that a failed match does not reset
the start position for /g matches on the string. That way you can
write parsers using a separate /gc pattern for each token. This
would be hardly any slower than the current Perl implementation
which also runs lots of separate regexes.

Regards,
-- 
Aristotle Pagaltzis // <http://plasmasturm.org/>