Can we use regex with cleartool commands?

Question

I am using below command to retrieve baselines.

cleartool lsbl -fmt "%n\n" -comp comp_name@\vob_name -stream stream_name@\vob_name

I am searching for a way to display baselines which is numerically equal to/ less than certain given baseline. Is there any way to achieve it?

Case 1 : If output is

abc_6.2168
abc_7.4587
abc_8.2950
abc_9.3032

If I want to display baseline which is numerically equal to / less (and closest) to abc_8. Hence, the expected result in Case 1 should be : abc_8.2950.

Case 2 : If output is

abc_6.2168
abc_7.4587
abc_9.3032

Expected result should be : abc_7.4587

NOTE : Trying this on Groovy (Jenkins pipeline)

Answer 1

     use strict;
     use warnings;
     use DBI;

     my $bsl_find = $ARGV[0]; #baseline build package name
     my $bsl;
     my $c=0;
     my $mat;
     my $previous_str = q{};
     my $final_baseline;
     my $prev_num_count=1;
     my $prev_num_len=1;
     my $split_strng;
     my $baseline_var = q{};
     my $baseline_file;
     my $all_baseline_file = $ARGV[1]; #file which contains the list of all retreived baselines as per ARGV[0]
     my $app = $ARGV[2]; #the name of the application for which baseline is to be selected


     my $filename = 'D:\\baseline_'.$app.'\\'.'new_'.$all_baseline_file.'.txt';
     $baseline_file = 'D:\\baseline_'.$app.'\\'.'final_'.$all_baseline_file.'.txt';
     $all_baseline_file = 'D:\\baseline_'.$app.'\\'.$all_baseline_file.'.txt';



     open(my $fh, '<:encoding(UTF-8)',$filename)
       or die "Could not open file '$filename' $!";

     while (my $strng = <$fh>) {

       chomp $strng;
       #print "The line is : $strng \n";


       $strng=~ s/^\s+|\s+$//;

        #print " \n strng after trim is $strng.";

        my $num_count = (split '_', $strng)[-1];
         my $num_count_bsl_param = (split '_', $bsl_find)[-1];

        my $num_len = length ($num_count);
         my $num_len_bsl_param = length ($num_count_bsl_param);


        my $a = substr($bsl_find, -$num_len_bsl_param);
        my $b = substr($strng, -$num_len);


        $split_strng = '_'.$a;


        my ($substrng) = split /$split_strng/, $bsl_find;

        if ($substrng =~ m/([^\_]+)$/)  
        {
            $substrng=$1;

        }



        if ( ($a == $b) && (index($strng, $substrng) != -1) )
        {

         print "\n Match found";
         $mat = $strng;
         print "\n baseline found is : $mat";
         $final_baseline = $mat;
         print "\n final bsl is $bsl_find";
         $baseline_var = $strng;
         #exit 0;
         goto label;

        }



        elsif ( ($a < $b) && (index($strng, $substrng) != -1) )

        {

         if ( (grep{/$bsl_find/} $filename) && ($previous_str eq "") ){
           print "\n final baseline decided  : $bsl_find";
          $baseline_var = $bsl_find;
          goto label;
         }

         elsif ( ($previous_str ne "") )

         {
                print "\n final baseline is ...: $previous_str";
                $baseline_var = $previous_str;
                goto label;
         }

        }

        elsif  ( ($a < $b) && ($previous_str ne "") && (index($strng, $substrng) != -1) )

        {



        if ( ($a > $c) && (index($previous_str, $substrng) != -1) )

        {

            print "\n baseline found is : $previous_str";
            $final_baseline = $previous_str;
            print " \n final is $final_baseline";
            $baseline_var = $previous_str;
            goto label;

         }

        }

        elsif ( ($a < $b) && (index($bsl_find, $substrng) != -1) && ($previous_str ne "") && (index($previous_str, $substrng) == -1) )

        {    
            print "\n Baseline not found of type $bsl_find.... final baseline is : $previous_str";
           $baseline_var = $previous_str;
           goto label;

        }

       close(fh);

      }



      if  ($baseline_var eq "")
      {
     open my $fh ,"<",$filename;
     my $last_line;
     $last_line = $_,while (<$fh>);
     print $last_line;

      print " \n Baseline is $last_line";
      $baseline_var = $last_line;
      goto label;
      close(fh);
      }


      label: 
        print " \n\n Writing $baseline_var to $baseline_file...";

        #$baseline_var = $baseline_var.'.';
        $baseline_var=~ s/^\s+|\s+$//;
        print " \n \n baseline_var is $baseline_var. ";

        unlink $baseline_file;

        open(my $fh, '<:encoding(UTF-8)',$all_baseline_file)
        or die "Could not open file '$all_baseline_file' $!";
        while (my $word = <$fh>) {
       chomp $word;
       #print "\n word is $word.";

       if ( $word =~ /\./ )

       {

         if( $word =~ m/$baseline_var\./ )
        {
          print "\n found $baseline_var. in $word";
          open(FH1, '>', $baseline_file) or die $!;

          print FH1 "$word";

        }

       }

       else
       {

          if( $word eq $baseline_var )
        {
          print "\n found $baseline_var. in $word";
          open(FH1, '>', $baseline_file) or die $!;

          print FH1 "$word";

        }

       }

       close(fh);
        }

     close(FH1);

      }

Answer 2

Is there any way to achieve it?

Not with ClearCase/cleartool alone, which means you need to parse its output, and that depends on your OS/shell.

Something like, in Windows CMD shell, with Git For Windows shell in its path:

cleartool lsbl ... | sort -V |awk 'BEGIN{a=$0;FS="._"}$2 ^< 9{print $0;}'|tail -1

(the ^< is needed to escape the <, and prevent the CMD to interpret that as a redirection)