howto grep in perl

bishal · 07-13-2004, 11:33 AM

Hello all

I want to grep the only the selected text in perl. Can anyone tell me how to do that and exclude the remaining file in perl. Below is my output of my perl scripts and I want to grep "unlock" only and exclude the rest. How can i do it with split command?

[abc@root]/sbin/test.pl

# Power lock status: unlock

keefaz · 07-13-2004, 02:52 PM

I assume you want to extract the word 'unlock' for Power lock status value

Code:

$str = "# Power lock status: unlock";
if( $str =~ /Power lock status/ )
{
    $str =~ s/^(.*)\: //;
    print "$str\n";
}

bishal · 07-14-2004, 01:46 AM

thanks kefaz it worked out. Can you explain me that second and 4th line of string and what does that tilde do?

keefaz · 07-14-2004, 07:12 AM

if( $str =~ /Power lock status/ )
Here, the regular expression (between / and / ) is used to match something so the =~ action related to $str is to search the pattern between / and / in it.

$str =~ s/^(.*)\: //;
Here the regular expression syntax is s///, it is to do a search and replace like :
$str = "this milk shake is too hot";
$str =~ s/hot/cold/;

This will replace hot by cold in the $str string.
^ : search from beginning of string
(.*) : any character match
"\: " : match a ':' folowed by a space

So the sequence s/^(.*)\: //; will replace all characters followed by a ':' and a space by nothing (//), in other words, it will delete all characters followed by a ':' and a space ( the ':' and space inclued).

bishal · 07-15-2004, 12:37 AM

Thank you keffaz

bishal · 07-18-2004, 06:32 AM

ok keefaz I have another question. How to ignore the last line also in perl. Example:
$line= /# milk in liter : 10.12[ltr]/;

I want only "10.12"

How to do that?

eric.r.turner · 07-18-2004, 07:19 AM

Quote:

Originally posted by bishal
I have another question. How to ignore the last line also in perl. Example:
$line= /# milk in liter : 10.12[ltr]/;

I want only "10.12"

How to do that?

Code:

#!/usr/bin/perl -w

$str = '# milk in liter: 10.12[ltr]';

if ( $str =~ /^.*:\s([0-9\.]+).*$/ ) {
  print( "$1\n" );
}

"^" matches the beginning of the line.

".*" matches zero or more of any character ("." matches any character, and "*" tells how many of them to match.) In this case, it matches "# milk in liter"

":\s" matches the colon followed by the space.

"([0-9\.]+)" is the key to the whole thing. Anything that matches the expression between "(" and ")" can be used later via the "$1" variable. See how we use it in the print statement? If you have multiple sets of "(" and ")" then you access them later via "$1", "$2", etc. The "[" and "]" represent a "character class". It means to match any of the characters between the "[" and "]". In this case we match any number 0 through 9, or the dot. The dot has to be escaped with a "\" otherwise it's the same as "any character". The "+" afterword means match one or more of the characters in the preceding character class. So, "[0-9\.]+" matches 10.12, 0.5, 10, ..1, 1.2.3.4, etc. We ignore the rest of the line by matching zero or more of anything, again using ".*".

"$" matches the end of the line.

If you wanted to capture the units (i.e. ltr) then you could do:

Code:

#!/usr/bin/perl -w

$str = '# milk in liter: 10.12[ltr]';

if ( $str =~ /^.*:\s([0-9\.]+).*\[(.*)\].*$/ ) {
   print( "$1\t$2\n" );
}

We'll leave it as an exercise for the reader to figure out that regular expression.

Eric

bishal · 07-19-2004, 01:32 AM

thanks eric, it worked out.