my substitution problem:

[Madams]

I have a string,and a substitution:
'11+b^10+s'=~s/(?<!\^)(?:\d+\+)//g;
it should result in:
   'b^10+s' NOT 'b^1s' !
it works if the number to the right of the "^" is less than 10 but not if greater.
i want to strip all numbers followed by a "+" (including the "+") but NOT if they are preceeded by a "^".

what should i do? (regexi drive mike crazy...(is that the proper plural?)

[tilly]

Regexen problems? Well the rule is that it will work really really hard to find a match. So figure out why it is finding the unwanted match and deny it.

Well on the second substitution it is clearly able to match "0+" after a 1. Well does that fit the rule? Why yes, the preceeding "1" is not "^", so it can match What to do? Why deny it a digit there. That gives us:

s/(?<![\^\d])(?:\d+\+)//g;
#     ^^^^^^ <- was just \^ before

which does what you asked.

[Madams]

Hey tilly!
thanks (for the fix and the plural i LIKE regexen as a word), i think i was getting google eyed trying to figure that out.. i was relatively sure it involved my not specifying the negassert correctly but i just continually woofed it! regexen are one way to kill your eyesight ;)

after a while they look like /{][P)(*^8(*(/ over and over :)
p.s. still like the multfor "loop" you clued me in on (still studying it cause it's a cool hack!)

Mucho gracias!
madams.

[japhy]

In an effort to shamelessly plug my modules, I produce the output of the OGRE:

NODE                     EXPLANATION
----------------------------------------------------------------------
(?-imsx:                 group, but do not capture (case-sensitive)
                         (with ^ and $ matching normally) (with . not
                         matching \n) (matching whitespace and #
                         normally):
----------------------------------------------------------------------
  (?<!                     look behind to see if there is not:
----------------------------------------------------------------------
    \^                       '^'
----------------------------------------------------------------------
  )                        end of look-behind
----------------------------------------------------------------------
  (?:                      group, but do not capture:
----------------------------------------------------------------------
    \d+                      digits (0-9) (1 or more times (matching
                             the most amount possible))
----------------------------------------------------------------------
    \+                       '+'
----------------------------------------------------------------------
  )                        end of grouping
----------------------------------------------------------------------
)                        end of grouping
----------------------------------------------------------------------

As tilly pointed out, the problem is that in the string "^123", "123" isn't a valid match, but "23" is.

japhy -- Perl and Regex Hacker