http://qs321.pair.com?node_id=986617


in reply to Re^3: Regex Extraction Help
in thread Regex Extraction Help

« Based upon the OP's regex, it appears that the pattern's stable--with one space after the semi-colon »

Oh indeed, my "warning" was meant more like a general tip, I didn't just mean this particular example. Just meant to say that it's a difference in how split vs if(m//) with some rather "strict" regexp typically result in a different level of defensiveness of the code. Again, I mean just typically. I mean hey, "just use split" would've been first answer, too. But you wrote that already, so I had to come up with something nitpicking. ;)

« However, we can ask split to 'test' the format of the input »

Umm... ok, you wrote 'test' in quotes, so alright... ;)

Sure, you can combine the split and trim operation, but still, this split would happily work on any input you throw at it (including undef, with a warning, though). It won't tell you (by not even matching) that your input looks a bit strange there.

Now, again, I am not so much talking about the OP's concrete problem, but was trying to educate a bit on what method to use when, since his usage of \d\d\d\d\d instead of \d{5} suggested that regexen ain't something he works with since years (No offence meant.)

So long,
Flexx

Replies are listed 'Best First'.
Re^5: Regex Extraction Help
by Kenosis (Priest) on Aug 09, 2012 at 22:16 UTC

    You make more good points, and am glad you offered the "general tip," as it helps with developing good programming practices. Anticipating and coding for exceptions can (and does) save many headaches...