Re: Extract individual lines and block of text from large files

in reply to Extract individual lines and block of text from large files

I'm not sure that I completely understand the context of your problem, but a few suggestions. Make a more general expression for the 'NAME: DATE' lines and capture into a hash. You have 4 regex lines to describe the filename, perhaps just one different regex would suffice? When you find the place after the FILENAME where the data capture is supposed to happen, sometimes it works out well to call a subroutine to finish that job off. Here is some code...

#!usr/bin/perl
use warnings;
use strict;
use Data::Dumper;
$Data::Dumper::Sortkeys =1;

my %hash;

sub get_html
{
   while (<DATA>)
   {
      if (/<html>/ .. /<\/html>/) ##see [id://525392]
      {
         chomp;
         push @{$hash{DATA}},$_;
      }
   }
}

while (<DATA>)
{
     if ( (my ($name, $date) =  m/^\s*([\w ]+):\s+([\w-]+)/)) 
     {
        $hash{$name}=$date;
     }

     if (my ($filename) = m/^\s*<FILENAME>\s*(\w+ex(-)?21.*.htm)/i)
     { 
        $hash{FILENAME}=$filename;
        get_html();
     }     
}

print Dumper \%hash;

=Prints: *********
$VAR1 = {
          'CENTRAL INDEX KEY' => '0000786368',
          'CONFORMED PERIOD OF REPORT' => '20081231',
          'DATA' => [
                      '<html>',
                      'blah ',
                      'smore blah',
                      'blahblah',
                      '      **** BODY OF TEXT I WISH TO EXTRACT *****
+',
                      '</html>'
                    ],
          'DATE AS OF CHANGE' => '20090331',
          'FILED AS OF DATE' => '20090331',
          'FILENAME' => 'v144610_ex21.htm',
          'FORM TYPE' => '10-K'
        };
=cut        


__DATA__
*********** Sample text ***************
CONFORMED PERIOD OF REPORT:    20081231     &#61663;------ individual 
+line I want
FILED AS OF DATE:        20090331     &#61663;------ individual line I
+ want
DATE AS OF CHANGE:        20090331     &#61663;------ individual line 
+I want

CENTRAL INDEX KEY:        0000786368    &#61663;------ individual line
+ I want 
        
    FORM TYPE:        10-K    &#61663;------ individual line I want

Whole buncha text here ЕЕЕЕЕ.

</DOCUMENT>
<DOCUMENT>
<TYPE>EX-21
<SEQUENCE>7
<FILENAME>v144610_ex21.htm      &#61663;-----------My starting point
<TEXT>
<html>
blah 
smore blah
blahblah
      **** BODY OF TEXT I WISH TO EXTRACT *****
</html>
</TEXT>
</DOCUMENT>               &#61663;----------- My ending point

**********End of sample text ***********
[download]

In Section Seekers of Perl Wisdom