Regular expression

PugSA has asked for the wisdom of the Perl Monks concerning the following question:

Hi Monks

I postes in chatterbox but thought to post a more accurate picture of what I am trying to accomplish Thank you for your time

I am tryning to match in $record CPOSTA_1221170039_C1_F1 but it includes the whole path when it matches

This is what I match => 49\backup\rbgmst02_dd2\stu1\CPOSTA_1221170039_C1_F1

how do I get it not to include the \ and why does it match the \ for interest sake?

I have tried [^\\] but it then just does not match anything

$record = "FRAG 1 1 2000000 0 0 0 0 \\\\172.20.13.49\\backup\\rbgmst02
+_dd2\\stu1\\CPOSTA_1221170039_C1_F1 rbgmst02 65536 0 0 -1 0 *NULL* 12
+22379639 1 65537 0 0 0 0 0 0 0";

if($record =~ m|FRAG|)
{
    if($record =~ m|([A-z0-9]+[_][0-9]+[_]C[0-9]+[_][A-z0-9]+)|)
    {
        #print $2."\n";
        push(@netbck_list,$1);

        print $1."\n";

    } #if($record =~ m|FRAG(.*?)\/([A-z0-9_-]+[_][0-9]+[_]C[0-9]+[_][A
+-z0-9]+)|)
    else
    {
        #print "FRAG RECORD NOT MATCHED => $record \n";
    }
} #if($record =~ m|FRAG|)
[download]

Comment on Regular expression Select or Download Code

Replies are listed 'Best First'.
Re: Regular expression by johngg (Canon) on Sep 16, 2008 at 14:36 UTC
Corion and shmem have both suggested an approach using a split on whitespace. You would need to take precautions if the path can include spaces. This can be done by isolating the path field using three-argument splits from either end of the record. use strict; use warnings; my @records = ( qq{FRAG 1 1 2000000 0 0 0 0 \\\\172.20.13.49\\backup\\rbgmst02_dd2\ +\stu1\\CPOSTA_1221170039_C1_F1 rbgmst02 65536 0 0 -1 0 NULL 1222379 +639 1 65537 0 0 0 0 0 0 0}, qq{FRAG 1 1 2000000 0 0 0 0 \\\\172.20.13.49\\backup\\rbgmst02_dd2\ +\New Folder\\Space File rbgmst02 65536 0 0 -1 0 NULL 1222379639 1 6 +5537 0 0 0 0 0 0 0}, ); foreach my $record ( @records ) { my @leftFields = split m{\s+}, $record, 9; my $path = ( reverse map { $_ = reverse } split m{\s+}, reverse( $leftFields[ 8 ] ), 18 )[ 0 ]; my $basename = ( split m{\\}, $path )[ -1 ]; print qq{$path\n $basename\n}; } [download] The output, `\\172.20.13.49\backup\rbgmst02_dd2\stu1\CPOSTA_1221170039_C1_F1 CPOSTA_1221170039_C1_F1 \\172.20.13.49\backup\rbgmst02_dd2\New Folder\Space File Space File` [download] I hope this is of interest. Cheers, JohnGG	[reply] [d/l] [select]
Re: Regular expression by Fletch (Bishop) on Sep 16, 2008 at 13:23 UTC
Rather than trying to roll your own there's the core File::Basename module . . . The cake is a lie. The cake is a lie. The cake is a lie.	[reply]
Re: Regular expression by Corion (Patriarch) on Sep 16, 2008 at 13:27 UTC
Also in the chatterbox, I pointed out File::Basename and File::Spec to you. Both will only work for your filenames if you're on Win32. If you're trying this on a non-Windows platform, the mentioned alternative of using ~~`/([^\\+])$/`~~`/([^\\]+)$/` to extract the filename part will still work, with the minor change of using whitespace instead of end of line as the anchor. Personally, I would approach your log parsing task by splitting on whitespace and then looking closer at the filename part: `my @entries = split "", $record; if ($record[0] eq 'FRAG') { warn "File is $record[8]"; $record[8] =~ /([^\\+])$/ or die "Malformed file entry: '$record[8]'"; warn "Basename is '$1'"; };` [download] Update: Fixed bad typo that rendered the RE useless.	[reply] [d/l] [select]
Re^2: Regular expression by PugSA (Beadle) on Sep 16, 2008 at 13:38 UTC
I'm on a Solaris machine and tried `([^\\+])` but can't match it at end of line thats why I posted the code here I'll try the split Thank you	[reply] [d/l]
Re^3: Regular expression by shmem (Chancellor) on Sep 16, 2008 at 14:05 UTC
You could use File::Basename, split and tr: `use File::Basename; $record = "FRAG 1 1 2000000 0 0 0 0 \\\\172.20.13.49\\backup\\rbgmst02 +_dd2\\stu1\\CPOSTA_1221170039_C1_F1 rbgmst02 65536 0 0 -1 0 NULL 12 +22379639 1 65537 0 0 0 0 0 0 0"; if($record =~ m\|FRAG\|) { my $str = (split " ", $record)[8]; $str =~ tr[\\][/]; print basename($str),"\n"; } __END__ CPOSTA_1221170039_C1_F1` [download] Alternative: `$record = "FRAG 1 1 2000000 0 0 0 0 \\\\172.20.13.49\\backup\\rbgmst02 +_dd2\\stu1\\CPOSTA_1221170039_C1_F1 rbgmst02 65536 0 0 -1 0 NULL 12 +22379639 1 65537 0 0 0 0 0 0 0"; if($record =~ m\|FRAG\|) { $record =~ /(?<=\\)(\w+)(?:\s\|$)/; print "$1\n"; } __END__ CPOSTA_1221170039_C1_F1` [download] See perlre.	[reply] [d/l] [select]
Re: Regular expression by pjotrik (Friar) on Sep 16, 2008 at 13:45 UTC
`A-z` [download] That's why, there's several non-alphabetic characters between 'Z' and 'a', including the backslash. `[A-Za-z0-9]` or `[[:alnum:]]` will match alphanumeric characters. Update: When you're not sure why your regex doesn't work, `use re 'debug';`	[reply] [d/l] [select]
Re: Regular expression by Grey Fox (Chaplain) on Sep 16, 2008 at 14:38 UTC
Lots of good regex info if you search around Perl Monks try Perlmonks Tutorial: Regular Expressions Tutorial, the Basics (for BEGINNERS) I have also found http://www.regular-expressions.info/quickstart.html very helpful working with regex's. Lots of good examples. I found tools like http://www.txt2re.com/ or Eclipses regex tool to be helpful in debugging my regex commands. Don't forget to use a \ to escape out special characters when using regex. Good Luck. -- Grey Fox "We are grey. We stand between the darkness and the light" B5	[reply]