Extracting text from a string using regex

ramya2005 has asked for the wisdom of the Perl Monks concerning the following question:

I have a question which is an extension of the solution that I got for my previous problem. Extracting specific text info from the config file ouput
I have a string as specified in the code below. I need a mechanisam to extract the authentication information.
For me the right authentication information are
1. authentication open <and/or followed by something>
eg1: authentication open network-eap
eg2: authentication open
2. authentication shared <and/or followed by something>
That means I should pick only the ones that start with authentication. Not the one starting with 'aaa' in the example below.
Also when I am looping I should strip of the part which I extracted from the original string.

Could some one tell me how to do it?

   my $str = "aaa authentication login\x0d\n ssid Rich\x0d\n authentic
+ation open\x0d\n authentication shared \x0d\n";
    
   my $authentication;  

   while ($str =~/authentication (\S+)/){
      $authentication = $authentiction. $1. " ";            
   }

   # At the end of the while loop
   # $authentication should have the value 
   # $authentication = open shared
[download]

Any input provided will be much appreciated! Thanks.

Comment on Extracting text from a string using regex Download Code

Replies are listed 'Best First'.
Re: Extracting text from a string using regex by holli (Abbot) on Dec 20, 2005 at 19:07 UTC
Your code is close but has two problems. First you do not use the g modifier in your regex, so your code runs into an endless loop. Second, your have a typo in this line: `$authentication .= $authentiction. $1. " ";` [download] That tells me you are using my but you are not using strict! Remember, always `use strict`. If you change that code to `$authentication .= $authentication. $1. " ";` [download] all is fine. That all boils down to `while ($str =~/authentication ?(\S+)?/g){ $authentication .= $authentiction. $1. " "; #more perlish }` [download] or even `$authentication = join " ", $str =~ /authentication ?(\S+)?/g; #even + more perlish` [download] holli, /regexed monk/	[reply] [d/l] [select]
Re: Extracting text from a string using regex by GrandFather (Saint) on Dec 20, 2005 at 19:46 UTC
Note that `\r` should generally be used for cr rather than `\x0d`. The following uses split to break the string into candidate sub strings, then uses a regex with a code evaluation expression in it to extract the text you want. A more conventional alternative is given too. `use strict; use warnings; my $str = "aaa authentication login\r\n ssid Rich\r\n authentication o +pen\r\n authentication shared \r\n"; my $authentication = ''; my @candidates = split /[\r\n] +/, $str; /^authentication\s+(\w)(?{$authentication.= $1 . ' '})/ for @candidat +es; print $authentication;` [download] Alternate extract code: `for (@candidates) { $authentication.= $1 . ' ' if /^authentication\s+(\w)/; }` [download] Either prints: `open shared` [download] DWIM is Perl's answer to Gödel	[reply] [d/l] [select]
Re: Extracting test from a string using regex by davido (Cardinal) on Dec 20, 2005 at 18:49 UTC
Something like this, if I understand the question correctly... `m/^(authentication\s(shared\|open)\s?(.+)?)/` [download] ...would be a reasonable choice for the regexp. $2 will contain shared or open. $3 will be defined and will contain whatever comes after shared or open, if something exists. That should get you going in the right direction. Dave	[reply] [d/l]