Rob/James Thanks for the input guys, still not quite there, damm nice code though...
the data files I process are typically 3000 lines long and full of blank lines, i use HTML::PARSER to strip the pages. sod's law say's the example I gave cannot be trusted ( and it can't) a line like 2 (5) cannot be expected, what can be expected is that after a count of 23 there will be the next number OR 22 there will be a blank line followed by the next sequence number. So the line will just be 2 instead of 2 (5) This is my predicament, had it been the above pattern I could have worked with it. I look forward to your thoughts, above all else it's making me think how I approach this. Steve -----Original Message----- From: James Edward Gray II [mailto:[EMAIL PROTECTED] Sent: 13 October 2003 18:26 To: [EMAIL PROTECTED] Subject: Re: design - help Looks like Rob beat me to the draw this time, but I thought I would add one little tidbit. On Monday, October 13, 2003, at 12:15 PM, Rob Dixon wrote: > if ( /(\d+)\s+\(\d+\)/ ) { I think at least anchoring the pattern to the front of the line is a good idea, just so we don't match some arbitrary junk in the middle of one of those text lines. I would use: /^(\d+)\s+\(\d+\)/ They probably both work exactly the same, but I prefer to anchor when I can just to reduce my chances for error. James -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
