Re: Match multiple lines

How do I best get it to take everything following the header:

<$fh_in>;  # Skip header
print $fh_out $_ while <$fh_in>;
[download]

Or if you don't mind loading the entire input file into memory:

<$fh_in>;  # Skip header
{ local $/; print { $fh_out } <$fh_in>; }
[download]

By the way, '.' indicates the current directory, so there's no need to load Cwd.

By the way, $_ != /\.ktf$/ makes no sense. You probably meant $_ !~ /\.ktf$/ (long for !/\.ktf$/). But isn't that inverted? I think you want /\.ktf$/ (short for $_ =~ /\.ktf$/).

So you get:

use strict;
use warnings;

use File::Find qw( find )

open(my $fh_out, '>', 'D:\\mergedfile.ktf')
    or die("Can't create output file \"D:\\mergedfile.ktf\": $!\n");

my $first = 1;

find(sub {
    return if !/\.ktf\z/;

    print("$File::Find::name\n");
    open(my $fh_in, '<', $File::Find::name)
        or die("Can't open file \"$File::Find::name\": $!\n");

    <$fh_in> if !$first;
    $first = 0;

    print $fh_out $_ while <$fh_in>;
}, '.');
[download]

Comment on Re: Match multiple lines Select or Download Code

Replies are listed 'Best First'.
Re^2: Match multiple lines by repellent (Priest) on Sep 19, 2009 at 16:03 UTC
I think other Kartex files also have headers, and that each header counts as two lines. So something like: `find(sub { return if !/\.ktf\z/; print("$File::Find::name\n"); open(my $fh_in, '<', $File::Find::name) or die("Can't open file \"$File::Find::name\": $!\n"); while (my $line = <$fh_in>) { next if $. <= 2; print $fh_out $line; } }, '.');` [download]	[reply] [d/l]
Re^3: Match multiple lines by WeeDie (Novice) on Sep 19, 2009 at 16:27 UTC
Yes exactly. The header is the 2 first lines of each file. I want a script that takes all lines following the header, from all .ktf files in the current directory, and merges them into one file, ultimately with the header added to it. I'm sorry, I don't understand the $fh_in and $fh_out variables. This is my first perl script >_< If someone could give me a more detailed reply on how to read and merge all lines following the 2 first lines, in all .ktf files in a directory, and write it to a file, I would be eternally grateful! :)	[reply]
Re^4: Match multiple lines by repellent (Priest) on Sep 19, 2009 at 18:01 UTC
WeeDie, did you try the solution provided by ikegami at the end of his post? Then try substituting the `find()` call with the one in my post? Please state how those solutions differ from your expectations. The `$fh_in` and `$fh_out` are filehandle variables, much like `MYFILE` in your OP. Since you're new to Perl, you can think of those filehandle variables as another way to write `GPXDATA` and `MYFILE`, respectively. Oh, and welcome to Perl! :)	[reply] [d/l] [select]
Re^4: Match multiple lines by ikegami (Patriarch) on Sep 20, 2009 at 16:45 UTC
I'm sorry, I don't understand the $fh_in and $fh_out variables. You used global vars `GPXDATA` and `MYFILE`. I switched to using lexical vars with more meaningful names: `$fh_in` and `$fh_out`. ("fh" stands for "file handle".) If someone could give me a more detailed reply on how to read and merge all lines following the 2 first lines, in all .ktf files in a directory, and write it to a file, I would be eternally grateful! :) I don't know how we can get more detailed then providing the code that does that.	[reply] [d/l] [select]
Re^2: Match multiple lines by WeeDie (Novice) on Sep 19, 2009 at 15:48 UTC
Thank you for your reply, I think you misunderstand me. I need a regex pattern that matches everything following the header to the end of the file. The file contains multiple lines. The script is able to match the first line, following the header, of each file and append them together into a single file. But I don't want the regex pattern to stop at the first newline, I want it to match all characters, even newline. I've also been confused by the if ($_ != /\.ktf$/) code, but it's the only thing that worked. It returns true if filename ends with .ktf. It looks inverted to me though... anyway, it works so...	[reply]
Re^3: Match multiple lines by WeeDie (Novice) on Sep 19, 2009 at 16:09 UTC
For example, this is the results running the script in a directory containing 10 ktf files. %,0,N59°3.248' E13°2.656',65.99,2009-09-16 09:51:42,,,$ %,0,N59°3.577' E13°2.460',53.97,2009-09-16 10:05:38,,,$ %,0,N59°3.576' E13°2.462',55.89,2009-09-16 10:07:51,,,$ %,0,N59°3.648' E13°2.924',55.89,2009-09-16 10:14:10,,,$ %,0,N59°3.652' E13°2.937',55.41,2009-09-16 10:19:17,,,$ %,0,N59°3.454' E13°3.092',55.89,2009-09-16 10:25:05,,,$ %,0,N59°3.453' E13°3.467',58.30,2009-09-16 10:31:30,,,$ %,0,N59°3.451' E13°3.466',57.34,2009-09-16 10:36:06,,,$ %,0,N59°3.473' E13°3.605',60.70,2009-09-16 10:39:35,,,$ %,0,N59°3.452' E13°3.467',56.37,2009-09-16 10:42:49,,,$ [download] The regex pattern stops at newline instead of matching until the end of file. I've tried the m modifier with no luck... Example: print MYFILE /^(%,0,.*)/ms	[reply] [d/l]