http://www.perlmonks.org?node_id=1022239

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

How could I gather the variables from a multi-part if with RegEx's , or do I need to break it down into separate parts? 
while(<IFILE>) {
   if ((@LINE =~ /#\s+(\w+)\s+(\d+).+/) && (@LINE =~ /!\s+(\w+)\s+(\d+
+).+/)) {
      $Word=$1;
      $TimeStamp=$2;
      $Word2=$3;
      $TimeStamp2=$4;
      print "X $Word X $TimeStamp X $Word2 X $TimeStamp2 x";
   }
}

[download]
Thanx, all help much appreciated. 
# add 1362072030 dc=ifdsgroup,dc=com
dn: uid=nce9834xdho,ou=user,ou=nce,ou=prod 
! port 1362 dc=isgrp,dc=com 
changetype: add
ifastCreateDate: 1362072030649
objectClass: account
objectClass: ifastUser
objectClass: top
ifastModifyDate: 1362072030649
uid: nce9834xdho
cn: nce9834xdho
ifastProducts: WEB
ifastGivenName: Omar
ifastLanguage: en
ifastFailedLogons: 0
ifastLastLogon: 15764
ifastEmailAddr: omar.hafez@td.com
ifastSurname: Hafez
structuralObjectClass: account
entryUUID: e587875c-1616-1032-93e5-4b7d087c5768
createTimestamp: 20130228172030Z
modifyTimestamp: 20130228172030Z
# end add 1362072030

# add 1362072030 dc=ifdsgroup,dc=com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# end add 1362072030

[download]
The suggestion by choroba seems to work, and is part of the solution, but I can't seem to get it working with more than one instance of it in the same if statement: Works: 
undef $/;
$/="\n\n";

while(<IFILE>) {
   my @RESULT=();
   my @LINES=split (/\n/, $_);
   if ((grep  /objectClass:\s+account/, @LINES)
    && (grep  /objectClass:\s+ifastUser/, @LINES)
    #&& (my ( $PRODS ) = /ifastProducts:\s+(.+)\s*$/, $_)
    && (my ( $USER ) = /dn: uid=(\w+?),.+/, $_)) {
      push(@RESULT, $USER);
      #push(@RESULT, $PRODS, $USER);
      print join "|", @RESULT;
      #print OFILE join "|", @RESULT;
      #print OFILE "\n";
      print "\n";
   }
}

[download]
But this does not: 
undef $/;
$/="\n\n";

while(<IFILE>) {
   my @RESULT=();
   my @LINES=split (/\n/, $_);
   if ((grep  /objectClass:\s+account/, @LINES)
    && (grep  /objectClass:\s+ifastUser/, @LINES)
    && (my ( $PRODS ) = /ifastProducts:\s+(.+)\s*$/, $_)
    && (my ( $USER ) = /dn: uid=(\w+?),.+/, $_)) {
      #push(@RESULT, $USER);
      push(@RESULT, $PRODS, $USER);
      print join "|", @RESULT;
      #print OFILE join "|", @RESULT;
      #print OFILE "\n";
      print "\n";
   }
}

[download]

Replies are listed 'Best First'.
Re: Get variables from if with regex
by choroba (Cardinal) on Mar 07, 2013 at 14:53 UTC
    In your particular case, it is not so hard to combine the regexes into one: 
    while (<IFILE>) {
    if (my ($word, $time_stamp) = /[#!]\s+(\w+)\s+(\d+)./) {
        print "[ $word $time_stamp ]\n";
    }
}

    [download]
    Note that the binding operator =~ imposes a scalar context on its left operand, so matching an array does not do what you think.
    لսႽ† ᥲᥒ⚪⟊Ⴙᘓᖇ Ꮅᘓᖇ⎱ Ⴙᥲ𝇋ƙᘓᖇ
[reply]
[d/l]
[select]
      Thank You
[reply]
      Thanx, all help much appreciated. 
      # add 1362072030 dc=ifdsgroup,dc=com
dn: uid=nce9834xdho,ou=user,ou=nce,ou=prod 
! port 1362 dc=isgrp,dc=com 
changetype: add
ifastCreateDate: 1362072030649
objectClass: account
objectClass: ifastUser
objectClass: top
ifastModifyDate: 1362072030649
uid: nce9834xdho
cn: nce9834xdho
ifastProducts: WEB
ifastGivenName: Omar
ifastLanguage: en
ifastFailedLogons: 0
ifastLastLogon: 15764
ifastEmailAddr: omar.hafez@td.com
ifastSurname: Hafez
structuralObjectClass: account
entryUUID: e587875c-1616-1032-93e5-4b7d087c5768
createTimestamp: 20130228172030Z
modifyTimestamp: 20130228172030Z
# end add 1362072030

# add 1362072030 dc=ifdsgroup,dc=com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# end add 1362072030

      [download]
      The suggestion by choroba seems to work, and is part of the solution, but I can't seem to get it working with more than one instance of it in the same if statement: Works: 
      undef $/;
$/="\n\n";

while(<IFILE>) {
   my @RESULT=();
   my @LINES=split (/\n/, $_);
   if ((grep  /objectClass:\s+account/, @LINES)
    && (grep  /objectClass:\s+ifastUser/, @LINES)
    #&& (my ( $PRODS ) = /ifastProducts:\s+(.+)\s*$/, $_)
    && (my ( $USER ) = /dn: uid=(\w+?),.+/, $_)) {
      push(@RESULT, $USER);
      #push(@RESULT, $PRODS, $USER);
      print join "|", @RESULT;
      #print OFILE join "|", @RESULT;
      #print OFILE "\n";
      print "\n";
   }
}

      [download]
      But this does not: 
      undef $/;
$/="\n\n";

while(<IFILE>) {
   my @RESULT=();
   my @LINES=split (/\n/, $_);
   if ((grep  /objectClass:\s+account/, @LINES)
    && (grep  /objectClass:\s+ifastUser/, @LINES)
    && (my ( $PRODS ) = /ifastProducts:\s+(.+)\s*$/, $_)
    && (my ( $USER ) = /dn: uid=(\w+?),.+/, $_)) {
      #push(@RESULT, $USER);
      push(@RESULT, $PRODS, $USER);
      print join "|", @RESULT;
      #print OFILE join "|", @RESULT;
      #print OFILE "\n";
      print "\n";
   }
}

      [download]
[reply]
[d/l]
[select]
        Do you really need to split the paragraph into lines? You can match a multiline string against a regex, too:
        #!/usr/bin/perl
use warnings;
use strict;

$/ = "\n\n";

while (<>) {
    my @result;
    if (     /objectClass:\s+account/
         and /objectClass:\s+ifastUser/
         and my ($prods) = /ifastProducts:\s+(.+?)\s*$/mg # The /m is 
+important here.
         and my ($user)  = /dn: uid=(\w+?),./g) {
        push @result, $prods, $user;
        print join '|', @result;
        print "\n";
    }
}

        [download]

        But if you need the array, you can have it. The , $_) part of your code does nothing, so remove it. Here is my attempt to do what you need: 

        #!/usr/bin/perl
use warnings;
use strict;

$/ = "\n\n";

while (<>) {
    my @result;
    my @lines = split /\n/;
    if ( (    grep  /objectClass:\s+account/,                  @lines)
         and (grep  /objectClass:\s+ifastUser/,                @lines)
         and (my ($prods) = map /ifastProducts:\s+(.+?)\s*$/g, @lines)
         and (my ($user)  = map /dn: uid=(\w+?),.+/g,          @lines 
+)) {
        push @result, $prods, $user;
        print join '|', @result;
        print "\n";
    }
}

        [download]
        لսႽ† ᥲᥒ⚪⟊Ⴙᘓᖇ Ꮅᘓᖇ⎱ Ⴙᥲ𝇋ƙᘓᖇ
[reply]
[d/l]
[select]
Re: Get variables from if with regex
by daxim (Curate) on Mar 07, 2013 at 14:29 UTC
    Please provide some sample input.
[reply]
Re: Get variables from if with regex
by TomDLux (Vicar) on Mar 09, 2013 at 19:29 UTC

    named captures.

    From the perlre page ...

    " ... as of Perl 5.10.0 you may use named capture buffers .... The notation is (?<name>...) to declare ..." (and $+{name} to access the captured value in code.

    As Occam said: Entia non sunt multiplicanda praeter necessitatem.

[reply]
[d/l]

Back to Seekers of Perl Wisdom