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

All if i have a simple loop that runs over each line of the file. like 
 for( split(/[\r\n]+/, $content) ) {
    if(($_ =~ m/35=8/)&&($_ =~ m/$tag1/)&&($_ =~ m/$tag2/)){
     printf $_;
    }elsif(($_ =~ m/35=9/)&&($_ =~ m/$tag1/)) {
     printf $_;
    }
 }

[download]

Now the above would match the string i am looking for, but the \r\n splits out the end of my strings line, so the message is. 
Message dump:
8=1034=DFG55=TEST90=23435=D
19 06/02/23 11:18:33:984 Sent Message

[download]

So when the code above runs it prints out.
8=1034=DFG55=TEST90=23435=D

[download]

But what i need is to catch the next line 
19 06/02/23 11:18:33:984 Sent Message

[download]

Replies are listed 'Best First'.
Re: Matching string, then getting next line
by inman (Curate) on Mar 08, 2006 at 10:39 UTC
    Hi minixman

    It would be worth trying a slightly different approach to the problem. You have mentioned that your content is coming from a file, it would be easier to work with the file itself line by line, look for the match and then do something with the next line. e.g. 

    while (<MYFILE>){
    chomp; #remove line feed
    if (/do the match here/) {
        my $nextLine = <MYFILE>;
        # do stuff with the next line here.
    }
}

    [download]

    This reads the content of the file handle, line by line placing each line into the default $_ variable. You can test to see if you have a match and then take another line from the file.

    If you have all of your content in a list then you can substitute the reading of a line from the file with a shift of a line form the list.

[reply]
[d/l]
      Hi THe problem with that is, that when perl reads my file which was created on windows, it treats the whole file as one line, that is why i had to use FILE::Slurp to split out the lines. So what i need to do is say okay if you match my line, place the value into a variable and then get the next line whatever it is and place that into a variable.
[reply]
        I guess that you are trying this with a file that has been generated on Windows and then transferred to a UNIX box. You end up with extra ^M chars when you look at the file using vi. Try trimming the surplus whitespace using a regex 
        #!/usr/bin/perl -w

use strict;

while (<MYFILE>){
    s/\s+$//;
    if (/NEXT LINE MARKER/) {
        my $nextLine = <MYFILE>;
        $nextLine =~ s/\s+$//;
        print "$nextLine\n";
        # do stuff with the next line here.
    }
}

        [download]
[reply]
[d/l]
        > THe problem with that is, that when perl reads my file
        > which was created on windows, it treats the whole file as one line

        that's odd. most MSDOS/Windows files use CR/LF as the end-of-line marker, but it sounds like your file uses just CR...are you sure it's a DOS file and not a Mac file?

        to read MSDOS/Windows text files, you need to set the input record separator to CR/LF rather than newline (actually, this isn't strictly necessary. if you don't do it, you just end up with a CR at the end of each line. chomp will get rid of it)

        i.e.

        $/="\r\n";

        to read Mac text files, set it to just CR:

        $/="\r";

        see the man page for 'perlvar' for more info.

[reply]
Re: Matching string, then getting next line
by lukeyboy1 (Beadle) on Mar 08, 2006 at 16:46 UTC
    One thing that's good to note is that I'd personally say don't just assume that $_ holds what you think it does. My advice would be NOT to use $_, particularly when using regular expressions.

    My reasoning is that anonymous variables almost always store up a world of trouble for you, and although I haven't looked at your example in any detail, I would think that doing multiple matches MIGHT have the effect of altering $_......

    Don't take this the wrong way, though. But you'll find that a combination of use strict; and named variables will also make the code waaay more obvious!

    Of course, I could just be talking total cack.... :)
[reply]
Re: Matching string, then getting next line
by graff (Chancellor) on Mar 09, 2006 at 03:19 UTC
    It would be easier if the file content were stored in an array (or if, as suggested above, you were reading it one line at a time). Something like this -- with some of the initialization steps filled in -- could be worthwhile: 
    open( INPUT, "some.file" ) or die $!;

my @content;
{
    local $/;
    @content = split( /[\r\n+]/, <INPUT> );
}
my $tag1 = "whatever";
my $tag2 = "blah";

my $i = 0;
while ( $i < $#content ) { # don't test very last line
    $_ = $content[$i++];
    print "$content[$i]\n"
        if (( /35=8/ and /$tag1/ and /$tag2/ ) or
            ( /35=9/ and /$tag1/ ));
}

    [download]

    I wanted to include the file open and read steps to show how to slurp it directly into an array (rather than slurping to a scalar then splitting to an array, which keeps two copies of the whole file in memory, which makes me itchy).

    BTW, notice that you were using printf with a single scalar arg. Someday, if that string happened to contain "%d" or something similar, you might find the output to be different from what you intended.

[reply]
[d/l]

Back to Seekers of Perl Wisdom