Beefy Boxes and Bandwidth Generously Provided by pair Networks
Your skill will accomplish
what the force of many cannot

Re^2: Search and Copy

by Athanasius (Chancellor)
on Nov 23, 2012 at 15:53 UTC ( #1005287=note: print w/replies, xml ) Need Help??

in reply to Re: Search and Copy
in thread Search and Copy

Assuming the specification is, “print the key phrase together with the following text up to 250 characters,” the regex would be better as /(web site.{0,250})/, which also matches when the key phrase is followed by fewer than 250 characters of text before the end of the file. As this match is greedy, it will match the largest number of characters up to 250.

Athanasius <°(((><contra mundum

Replies are listed 'Best First'.
Re^3: Search and Copy
by mikebailey (Initiate) on Nov 23, 2012 at 16:35 UTC
    This is what I have so far... open (OUTPUT, ">Results.txt") || die ("Could not open file results.txt; $OS_ERROR"); open( INFILE, "Textfile.txt" )or die("Can not open input file: $!"); while (<INFILE>) { if ($ARG =~ /Something/ ) { print OUTPUT $ARG ; } } close (OUTPUT);

      Again, please use <code> tags and read How do I post a question effectively?

      I have found strict to be very helpful to identifying and avoiding bugs in my code -- see Use strict warnings and diagnostics or die. If I were going to write your posted code, it might look more like:

      use strict; use warnings; open (my $out, ">", "Results.txt") or die ("Could not open file Result +s.txt; $!"); open (my $in, "<", "Textfile.txt") or die ("Can not open input file: $ +!"); local $/; while (<$in>) { if (/(web site.{250})/i) { print $out $1; } }

      Changes that I made include:

      1. I swapped to lexical file handles and 3 argument open, which are considered better practice for a number of reasons. See perlopentut. In particular, this gives strict more power to help and removes the need for explicit close.
      2. I corrected inconsistency between your file name and error message; file names are generally case sensitive.
      3. I swapped to slurp mode using $/. Given the large number of characters you are interested in, it is unlikely they will all fall on the same line.
      4. Your while(<>) loop read data into $_ not $ARG, so I corrected that.
      5. I swapped your regular expression to the regular expression I posted above, with the addition of the s modifier. This makes it so . also matches new lines, and is essential when working in slurp mode.

      You may consider going to to gather some learning resources before trying to run too far.

      #11929 First ask yourself `How would I do this without a computer?' Then have the computer do it the same way.

      open (OUTPUT, ">Results.txt") open( INFILE, "Textfile.TXT" ) while (<INFILE>) { if ($ARG =~ /(Something.{0,250})/ ) { print OUTPUT $ARG ; } } close (OUTPUT);

      The "regexp" suggested is not working.


        The regex would work if you match it against the right variable. Where is $ARG coming from? Use strict and warnings; and if someone gives you a code, taking just a small part of it if you do not understand the rest would probably not work.
        لսႽ ᥲᥒ⚪⟊Ⴙᘓᖇ Ꮅᘓᖇ⎱ Ⴙᥲ𝇋ƙᘓᖇ

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://1005287]
and all is quiet...

How do I use this? | Other CB clients
Other Users?
Others imbibing at the Monastery: (2)
As of 2018-01-21 09:44 GMT
Find Nodes?
    Voting Booth?
    How did you see in the new year?

    Results (227 votes). Check out past polls.