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

Hello all,

I'm trying to simplify my subroutine that is supposed to open an (LDIF)text file, read in its contents into an array then print to a new file while excluding the lines that exist in another array.

I can currently accomplish this with just the word I want to exclude, but of course this leads to bloated code because I have to continuously open and close the file on each removal.

Here is what I have so far: 

sub GROOMLDIF {
     # Delete LDAP-internal fields
      my @erasethese = qw(structuralObjectClass entryUUID creatorsName
+ modifiersName createTimestamp modifyTimestamp entryCSN);
   
      open(FILE,"< data/all.ldif");
       my @LINES = <FILE>;
      close(FILE);
      open(FILE,"> data/groomed.ldif");
       foreach my $LINE (@LINES) {
        my @array = split(/\:/,$LINE);
        print FILE $LINE unless ($array[0] eq "$erasethese[0]");
       }
      close(FILE);

      open(FILE,"< data/groomed.ldif");
          @LINES = <FILE>;
      close(FILE);
      open(FILE,"> data/groomed.ldif");
       foreach my $LINE (@LINES) {
        my @array = split(/\:/,$LINE);
        print FILE $LINE unless ($array[0] eq "$erasethese[1]");
       }
      close(FILE);
}

[download]

I would like to combine all of them into one open/close. I've tried a nested for-loop that incremented with no luck.

Any help would be appreciated.

Replies are listed 'Best First'.
Re: Clearing lines in a file based on an array containing the lines
by almut (Canon) on Apr 22, 2009 at 20:32 UTC

    When you put the names of the fields in a hash 

    my %erasethese = map { $_ => 1 } qw(structuralObjectClass entryUUID
                                    creatorsName modifiersName createT
+imestamp
                                    modifyTimestamp entryCSN);

    [download]

    you can then write 

        print FILE $LINE unless exists $erasethese{$array[0]};

    [download]

    which would test if $array[0] is any of the keywords (which - if I've understood you correctly - is what you want to achieve).

[reply]
[d/l]
[select]
      Thank you Almut... that was awesome. I was 'this' close (*holding fingers very close together.*)

      Thank you monks for your assistance. Much appreciated.
[reply]
Re: Clearing lines in a file based on an array containing the lines
by ramrod (Priest) on Apr 22, 2009 at 20:27 UTC
    You can eliminate the second "grooming" by comparing against your elements at the same time: 

    print FILE $LINE unless ($array[0] eq "$erasethese[0]" || $array[0] eq
+ "$erasethese[1]" );

    [download]
    As far as only opening the file once, you can try opening it for both reading and writing (+<)
[reply]
[d/l]
Re: Clearing lines in a file based on an array containing the lines
by toolic (Bishop) on Apr 22, 2009 at 20:33 UTC
    I think grep might help you out, especially if you need to check against all of the @erasethese items. Since you did not provide a small sample of your input file, I will only offer this untested code (which does check against all @erasethese items): 
    sub GROOMLDIF {
    # Delete LDAP-internal fields
    my @erasethese = qw(
        structuralObjectClass
        entryUUID
        creatorsName
        modifiersName
        createTimestamp
        modifyTimestamp
        entryCSN
    );
   
    open my $fh_in , '<', "data/all.ldif"     or die "can not open dat
+a/all.ldif: $!";
    open my $fh_out, '>', "data/groomed.ldif" or die "can not open dat
+a/groomed.ldif: $!";
    while (<$fh_in>) {
        my $line = $_;
        my @thing = (split /:/)[0];
        print $fh_out $line unless (grep {$thing eq $_} @erasethese);
    }
}

    [download]
[reply]
[d/l]
[select]
Re: Clearing lines in a file based on an array containing the lines
by jwkrahn (Abbot) on Apr 22, 2009 at 23:33 UTC

    Perhaps this will work better for you (UNTESTED): 

    sub GROOMLDIF {
    # Delete LDAP-internal fields

    my $erasethese = qr/\A(?:structuralObjectClass|entryUUID|creatorsN
+ame|modifiersName|createTimestamp|modifyTimestamp|entryCSN):/;

    open my $IN,  '<', 'data/all.ldif'     or die "Cannot open 'data/a
+ll.ldif' $!";
    open my $OUT, '>', 'data/groomed.ldif' or die "Cannot open 'data/g
+roomed.ldif' $!";

    while ( <$IN> ) {
        print $OUT $_ unless /$erasethese/;
    }
}

    [download]
[reply]
[d/l]
Re: Clearing lines in a file based on an array containing the lines
by NiJo (Friar) on Apr 23, 2009 at 18:12 UTC 
    $command = join (' | grep -v ', @erase_these)
system "cat $in_file" . $command . " > $out_file"

    [download]
    Least amount of your code, C speed, use of multiple cores for free

    Limitation: maximum command length on shell

[reply]
[d/l]

Back to Seekers of Perl Wisdom