[ILUG] Perl RE question

[Fergal Daly]

That RE also matches

^947,356

so I guess you can only use real live characters/character classes in []
what you really want is

/(^|,)9/

It's only on stroke more,

Fergal

On Wed, May 30, 2001 at 06:28:18PM +0100, Niall O Broin wrote:
> I have a list of numbers in a string like
> 
> 
> 123,456,789
> 
> and I need to find out if any of them begin with 9. This seemed correct
> 
> #!/usr/bin/perl
> $numlist = $ARGV[0];
> if ($numlist =~ /[,^]9/) {
>   print "$numlist matched\n"
> } else {
>   print "$numlist didn't match\n"
> }
> 
> but it doesn't work. Now, I have solved my problem. It's Perl, and TIMTOWTDI
> as we Perlmongers say. I could have split the string, but I chose to do
> 
> if (",$numlist" =~ /,9/)
> 
> which works fine. But why won't my program above work ? I had started off
> with 
> 
> if ($numlist =~ /[^,]9/)
> 
> but realised (in time, believe it or not) that this was of course a negating
> character class. I tried escaping the ^ before I RTFM and simply made the ^
> not the first character, as TFM says to do. I also tried \A instead of ^ but
> although this works outside of a character class, it does not work as above.
> 
> I just tried this with grep too. Given a file, nums, with these lines
> 
> 123,987 - MATCH
> 123,456 - NO MATCH
> 987,123 - MATCH
> 
> I think that  grep "[,^]9" nums    should match the lines marked MATCH but
> it doesn't.  egrep ",9|^9" nums  does match the lines marked MATCH so why
> the hell doesn't the alternation work ? What is wrong with my interpretation
> of both the grep and the Perl manuals ? (The alternative is that they're
> both wrong, and that seems passing unlikely).
> 
> 
> 
> Regards,
> 
> 
> 
> Niall
> 
> 
> 
> -- 
> Irish Linux Users' Group: ilug at linux.ie
> http://www.linux.ie/mailman/listinfo/ilug for (un)subscription information.
> List maintainer: listmaster at linux.ie

--