Perl Guru Forums: Perl Programming Help: Regular Expressions: regex removing similar enteries

CGI/Perl Guide | Learning Center | Forums | Advertise | Login

Site Search: in


	MAIN INDEX	SEARCH POSTS	WHO'S ONLINE	LOG IN

Home: Perl Programming Help: Regular Expressions:
regex removing similar enteries

pnajafi
New User

Jan 7, 2011, 10:04 AM

Post #1 of 4 (8981 views)

regex removing similar enteries	Can't Post

Hi There,

I have the following text (it is not a regex rather a file path followed by space followed by more text):

/next/ whatevertext
/next whatevertext
/Next watevertext
/next/next wordswords
/next/ words

Result of the match should be as follows:

The first 3 lines:

the result of this match should be only one line:
/next whatevertext
(given that "whatevertext" is also the same in all the 3 lines)
this is because "\next", "\Next" and "\next\" are all similar and I want only the one lowercase with no slash at the end.

The last two lines are selected as well, because "/next/next" and "/next/" are different.

My attempted regex Pattern is:

Code

/((.*)(\/)? (.*)){1}/i

So what I did in the pattern is to say zero or more characters (.*), alternatively ending with a slash (\/)? then space then zero or more characters (.*).

Now ((.*)(\/)? (.*)){1} means that all in the bracket happens at most once. And case insensitive is set using i flag ; however upon testing I am not getting the result I am looking for.

I think the problem lies in the fact that I am looking for something that happens at most once but it happens 3 times, so I would only (probably get only the last 2 distinct lines). So how do I go about it?
Any help is appreciated.

BillKSmith
Veteran

Jan 7, 2011, 3:12 PM

Post #2 of 4 (8973 views)

Re: [pnajafi] regex removing similar enteries [In reply to]	Can't Post

I think this is what you want. Note the /xms modifiers.

Code

use strict; 
use warnings; 
my $INTEXT = 
     "/next/ whatevertext\n" 
    ."/next whatevertext\n" 
    ."Next watevertext\n" 
    ."next/next wordswords\n" 
    ."next/ words\n" 
; 
my ($line2,$line45) = $INTEXT =~ /.+\n     #skip first line 
                              (.+\n)       #capture second line 
                              .+\n         #skip third line 
                              (.+\n.+\n)\Z #Capture last two lines 
/xms; 
my $out_text = $line2.$line45; 
print $out_text;

Good Luck,
Bill

pnajafi
New User

Jan 8, 2011, 9:48 PM

Post #3 of 4 (8929 views)

Re: [BillKSmith] regex removing similar enteries [In reply to]	Can't Post

Thanks BillkSmith.

I have not tested this yet, but the code seems to rather follow things rather specifically to this case. I am reading a file that has lines such as:

\dir\dir1\ someurl
\dir2\dir3 someurl2
\dir2 someotherurl3
\dir2\ someotherurl3
\dir3\ someotherurl4
\Dir3 someotherurl4

and the result of the match will be:

.....
\dir\dir1\ someurl
\dir2\dir3 someurl2
\dir2 someotherurl3
\dir3 someotherurl4
.....

your code seems like it is only limited to the 4 lines, where is this file I am reading could have arbitrary number of lines.

BillKSmith
Veteran

Jan 9, 2011, 2:07 PM

Post #4 of 4 (8897 views)

Re: [pnajafi] regex removing similar enteries [In reply to]	Can't Post

It now seems that you want to delete lines whose final field has already been seen. (That assumes that the missing 'h' in line 3 of your first post was a typo).

Do not try to use a single regular expression. Use grep.

Code

use strict; 
use warnings;  
 
my $INTEXT = 
     '\dir\dir1\ someurl'   . "\n"  
    .'\dir2\dir3 someurl2'  . "\n" 
    .'\dir2 someotherurl3'  . "\n" 
    .'\dir2\ someotherurl3' . "\n" 
    .'\dir3\ someotherurl4' . "\n" 
    .'\Dir3 someotherurl4'  . "\n" 
;  
 
 
my @lines = split "\n", $INTEXT;  
 
my %seen=(); 
my @results = grep {!seen($_)} @lines; 
{ 
local $, = "\n"; 
print @results; 
}  
 
 
{ 
%seen=(); 
sub seen { 
    my ($line) = @_; 
    my ($field) = $line =~ /\s*(\S+)$/; 
    return $seen{uc $field}++; 
} 
}

Good Luck,
Bill

Search for (options)

Web Applications & Managed Hosting Powered by Gossamer Threads
Visit our Mailing List Archives