Perl Extract specific string

[MacLing]

Hi Monks,

could you give me perl script to extract the specific string like data below.

Source file : input.txt

06.02.2013 12.24.01:909 5807225321 INFO {EXT:httpadapter:17:14:}[0]RULEZ HTTPADAPTER,msisdn:637584930382,ud:Pan,trxtime:20130206122401 ResponseDeltaTime:31 ms ResponseCode:200 ResponseBody:OK

output1.txt

06.02.2013,637584930382,Pan,20130206,200

output2.txt

06-02-2013,637584930382,Pan,2013-02-06,200

[davido]

We're here to help, tutor, guide, teach, pat each other on the back, brag, show off, socialize, learn, ponder, share, collaborate, argue, and distract ourselves from our real $work. Not included in that list is write free programs on demand.

Maybe you already know this and just didn't know how to ask or what to ask for, so let's try again: What have you tried, and what part are you having trouble with? How can we help you in this learning process?

[CountZero]

Before we answer we need some more info.
• Are these three lines of data actually ONE line or indeed THREE lines? Tip: enclose your data in <code> ... </code> tags. That makes it easier to see the actual format and also things like square brackets and such.
• Can you show some more lines of your data? Especially when you have to write perhaps a regular expression to extract the data it is always good to see some more data, to check how "regular" your data structure really is.

[MacLing]

Hi CountZero,

It One line. and structure data is same. and there is a lots of data in each file.

anyway thanks for your reply, i will test it. Please give me the script which can read from input file and create the output file

[CountZero]

Perl is such a powerful language, you only need a few lines of code:

use Modern::Perl;

while (<DATA>) {
    my ( $date, $msisdn, $ud, $trxtime, $response ) =
      map { /:([^:]+)/ ? $1 : $_ } ( split /[\s+,]/ )[ 0, 6, 7, 8, 11 
+];
    say join ',', $date, $msisdn, $ud, substr( $trxtime, 0, 8 ), $resp
+onse;
}

__DATA__
06.02.2013 12.24.01:909 5807225321 INFO {EXT:httpadapter:17:14:}[0]RUL
+EZ [HTTPADAPTER],msisdn:637584930382,ud:Pan,trxtime:20130206122401 Re
+sponseDeltaTime:31 ms ResponseCode:200 ResponseBody:OK

Output:

06.02.2013,637584930382,Pan,20130206,200

[Anonymous Monk]

I agree, with great minds that has given there wisdom on this. However, since is your first time in the house, I will cut you some slack on the rule of the house which you can read for yourself here How do I post a question effectively?.

I can think of three ways of getting to your desired output:

• see Text::CSV or Test::CSV_XS. Since you presented a Comma seperated string
• use split,conditional loop "if" and regex like this:

  #!/usr/bin/perl
  use warnings;
  use strict;
  
  my $str = <<'_STR_';
  06.02.2013 12.24.01:909 5807225321 INFO {EXT:httpadapter:17:14:}[0]RUL
  +EZ HTTPADAPTER,msisdn:637584930382,ud:Pan,trxtime:20130206122401 Resp
  +onseDeltaTime:31 ms ResponseCode:200 ResponseBody:OK
  
  _STR_
  
  my $wanted_data;
  for ( split /,/, $str ) {
      if (/(\d{2}\.\d{2}\.\d{4})/) {
          my $format_date = $1;
          $format_date =~ s/\./-/g;
          $wanted_data .= $format_date;
      }
      elsif (/msisdn:(.+)/) {
          $wanted_data .= ',' . $1;
      }
      elsif (/ud:(.+)/) {
          $wanted_data .= ',' . $1;
      }
      elsif (/trxtime:(\d{8}).+Code:(\d+)/sm) {
          my ( $format_date, $code ) = ( $1, $2 );
          $format_date =~ s/(\d{4})(\d{2})(\d{2})/$1-$2-$3/;
          $wanted_data .= ',' . $format_date . ',' . $code;
      }else{ print "data given is not true"}
  }
  
  print $wanted_data, $/;
  

• Or you can use "dispatch table" instead of the ifs and elsifs like this:

  #!/usr/bin/perl
  use 5.014;    ## it wouldn't work for lower version
  use Readonly;
  
  Readonly my $comma => ",";
  
  my $str = <<'_STR_';
  06.02.2013 12.24.01:909 5807225321 INFO {EXT:httpadapter:17:14:}[0]RUL
  +EZ HTTPADAPTER,msisdn:637584930382,ud:Pan,trxtime:20130206122401 Resp
  +onseDeltaTime:31 ms ResponseCode:200 ResponseBody:OK
  
  _STR_
  
  my %data_filter = (
      '1_first_date_format' => sub {
          if ( $_[0] =~ /(\d{2}\.\d{2}\.\d{4})/ ) {
              my $format_date = $1;
              $format_date =~ s/\./-/g;
              return ( $format_date, $comma );
          }
      },
      '2_msisdn' => sub {
          if ( $_[0] =~ /msisdn:(.+)/ ) {
              return ( $1, $comma );
          }
      },
      '3_ud' => sub {
          if ( $_[0] =~ /ud:(.+)/ ) {
              return ( $1, $comma );
          }
      },
      '4_trxtime' => sub {
          if ( $_[0] =~ /trxtime:(\d{8}).+Code:(\d+)/sm ) {
              my ( $format_date, $code ) = ( $1, $2 );
              $format_date =~ s/(\d{4})(\d{2})(\d{2})/$1-$2-$3/;
              return ( $format_date, $comma, $code );
          }
      },
  );
  
  for my $key ( sort keys %data_filter ) {
      for ( split /,/, $str ) {
          print $data_filter{$key}->($_);
      }
  }
  

However, pay attention this "..Can you show some more lines of your data? Especially when you have to write perhaps a regular expression to extract the data it is always good to see some more data, to check how "regular" your data structure really is..." by CountZero and others before.
Hope this help in some ways.

[MacLing]

Thanks CountZero !!!

please help me again.

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

my $str = <<'_STR_';
09.05.2014 09.49.52:359 RID:routerNode1@app1:1662306081,msisdn:7878872
+25696,sid:93889095007001,tid:1405090902095648846024000,status:2,time:
[phone],reason:DELIVRD,refund:null,status2:null

_STR_

my $wanted_data;
for ( split /,/, $str ) {
    if (/(\d{2}\.\d{2}\.\d{4}' '\d{2}\.\d{2}\.\d{2}).+RID:(\d+)/sm) {
        my ( $format_date, $rid )= ( $1, $2 );
        $format_date =~ s/(\d{4})(\d{2})(\d{2})/$1-$2-$3/;
        $wanted_data .= $format_date . ',' . $rid;
    }
    elsif (/msisdn:(.+)/) {
        $wanted_data .= ',' . $1;
    }
    elsif (/sid:(.+)/) {
        $wanted_data .= ',' . $1;
    }
    elsif (/tid:(.+)/) {
        $wanted_data .= ',' . $1;
    }
    elsif (/status:(.+)/) {
        $wanted_data .= ',' . $1;
    }
    elsif (/time:(.+)/) {
        $wanted_data .= ',' . $1;
    }
    else{ print "data given is not true"}
}

print $wanted_data, $/;

the wanted_data like this below :

09-05-2014,09-49-52,routerNode1@app1:1662306081,787887225696,93889095007001,1405090902095648846024000,2,20140509094952

when run, error like below:

data given is not truedata given is not truedata given is not truedata given is not true,787887225696,93889095007001,1405090902095648846024000,2,20140509094952

and please advice, how to read the input file and write to the output file.

Thank you.

[ansh batra]

could you give me ...
you should avoid using this kind of language in any forum
follow the steps given by davido
if you have not tried anything then try it using regular expressions
refer tutorials sections to learn about regular expressions

[Anonymous Monk]

Sure, see perlintro and Re^3: Quick Question about batching files and Re^3: file header change