reading from a text file

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

Hello, I have a text file which looks exactly like this:

FIRST Level Description
EQT.FIRST.Id: 2
EQT.FIRST.eqtID_01.allowedType: 4TEST
EQT.FIRST.eqtID_02.allowedType: 5TEST
SECOND Level Description
[download]

And from this file I want to read the values:01,02 and 4TEST,5TEST Here is my code but it doesn't work properly:

my $data_file = "test.txt";
open(FILE, "< $data_file");
while(<FILE>) 
{
 if (  /FIRST Level Description/.. /SECOND Level Description/ )
{    if (/EQT.FIRST.eqtID_(\d+).allowedType:([A-Z0-9]+)/)
    { ($str1,$str2 ) = ($1,$2); }
    print "$str1 \n $str2";
}}
[download]

Thank you for your time.

Comment on reading from a text file Select or Download Code

Replies are listed 'Best First'.
Re: reading from a text file by robartes (Priest) on Jun 07, 2005 at 08:05 UTC
Your second regex does not match because of a space between the `[45]TEST` and and the rest of the line in the input. Try this regex: `/EQT.FIRST.eqtID_(\d+).allowedType:\s+([A-Z0-9])+)/` [download] Notice the extra \s+ (match 1 or more whitespace characters). CU Robartes-	[reply] [d/l] [select]
Re: reading from a text file by ZlR (Chaplain) on Jun 07, 2005 at 08:11 UTC
Also remember that the dot has the special meaning of "any character" in a regex, so while your code is working ( with robartes' correction ) it's not exactly doing what you think . Since you want to match real dots you should probably use \. in your expressions ( escaping the . metacharacter with \ will make it lose its special meaning)	[reply]
Re: reading from a text file by tchatzi (Acolyte) on Jun 07, 2005 at 08:41 UTC
Try something like this (simple though) `#!/usr/bin/perl -w use strict; open(F,"test.txt"); foreach(<F>){ if ($_ =~ /.(\d\d).(\d\w+)/){ print "$1 : $2 \n"; } } close F;` [download] ``The wise man doesn't give the right answers, he poses the right questions.'' TIMTOWTDI	[reply] [d/l]
Re^2: reading from a text file by samy_kumar (Scribe) on Jun 07, 2005 at 12:13 UTC
`if ($_ =~ /.(\d\d).(\d\w+)/){` [download] I think you might have forgot that \w indeed accepts digits. -madasamy-	[reply] [d/l]
Re^3: reading from a text file by tchatzi (Acolyte) on Jun 07, 2005 at 14:05 UTC
No i didn't forget. But its either a '\d' inside the '()' like this (\d\w+) or a '\s' outside like this \s(\w+) Do you see any big difference? Cause if you don't use either of them and write it like this `($_ =~ /.(\d\d).(\w+)/)` [download] Then the only thing that $2 will have inside, will be a single 'T'. So i suggest you study a bit of regex more, before coming to conclusions ``The wise man doesn't give the right answers, he poses the right questions.'' TIMTOWTDI	[reply] [d/l]
Re^3: reading from a text file by tchatzi (Acolyte) on Jun 07, 2005 at 14:18 UTC
The only thing you can take out is the first ' .* ' Which i put there for educational reasons, cause the guy doesn't know too much of regex, so there is no need to give him a hard time, yet. ``The wise man doesn't give the right answers, he poses the right questions.'' TIMTOWTDI	[reply]
Re: reading from a text file by aukjan (Friar) on Jun 07, 2005 at 08:11 UTC
Also you might want to change '.' into '\.' to specifically match a '.' and not 'any character'. This will prevent any unsuspected behaviour. .:\| If it can't be fixed .. Don't break it \|:.	[reply]


Perl-Sensitive Sunglasses
	PerlMonks