Perl Regex Tutorial

Carlos   

Posted at 7:27am on Monday, August 27th, 2007

Very usefull

Carlos   

Posted at 7:31am on Monday, August 27th, 2007

Why this regex "^-\(w+)" does not retreive enything from here "-lolo",

Thank you very much

Perlfect   

Posted at 7:35am on Monday, August 27th, 2007

Carlos, the backslash should be before the "w+", not the parenthesis. Your regex should read "^-(\w+)", without the quotes.

Eddie   

Posted at 12:51pm on Tuesday, September 11th, 2007

I put (/w+) in your tutorial running on \fredupperlower/clipped/file.txt and it matches /clipped, which I want it to. But in my script, it's not working. The sequence is:

$dir = $File::Find::dir;
$dir =~ /(/w+)/;
$dirname = $1;
print "directory name: $dirnamen";

Eddie   

Posted at 1:16pm on Tuesday, September 11th, 2007

I tried using:

$dir =~ m!/(w+)!;

and that seemed to work (better than I expected, since it returns "clipped" instead of "/clipped" which is easier to work with for my purposes).

Agus   

Posted at 10:19am on Thursday, September 13th, 2007

I need your, how to validate phone number. eg: (021)087655 or 021-56745.

thanks

Sleve   

Posted at 6:53pm on Monday, September 24th, 2007

Agus: (d{3})d{6} or
d{3}-d{5}
I forget how to make it so that "either" will grab the line.
(((d{3})d{6})|(d{3}-d{5})) should have done it, I think, but it failed to catch the second one in capture groups.

Sleve   

Posted at 6:56pm on Monday, September 24th, 2007

Agus: I put it in correctly, but the comment engine stripped out some back slashes. There should be a backslash before each parenthesis in the first one listed.

Kokok   

Posted at 12:30am on Thursday, January 10th, 2008

This page is great to test my regex

Siidheesh   

Posted at 11:03pm on Friday, October 24th, 2008

Now that my regex, (TEXT|MATH)[_](.+)[(](.+)[)], is working correctly, how do I enclose the regex in an 'if' statement to test a variable against it and get the $1, $2 and $3 variables.

Karyn   

Posted at 8:38am on Monday, November 17th, 2008

Excellent - saved me a lot of time from running the script, changing the regexpr, running it again, re-changing the regexpr etc. Thanks!

Dennis   

Posted at 9:32pm on Wednesday, December 31st, 2008

Question: How do I validate the 'regular expression' itself before using it? I have an application that allows users to enter regular expressions to 'mask' data and it blows up if they enter an invalid regular expression. Would like to check the regex first for validity to prevent it from blowing up the script when it is actually used... Thanks in advance, Dennis

thyme   

Posted at 10:45am on Tuesday, January 13th, 2009

Which regular expression can I use to extract the FOUR-DIGIT number in the text "my name is mgt, I have 1234 at someplace"

Please help!!!

Score_Under   

Posted at 1:25pm on Thursday, January 15th, 2009

@Thyme:

bd{4}b

Score_Under   

Posted at 1:25pm on Thursday, January 15th, 2009

Sorry, it filtered the backslashes. There was a backslash before the "b", "d" and other "b".

soger   

Posted at 11:05am on Sunday, January 18th, 2009

Dennis: to validate a regular expression you can do something like this:

eval {''=~/$user_regex/};
if ($@) {
print "Invalid regular expression: $@";
} else {
do_something($user_regex);
}

alper   

Posted at 10:17am on Wednesday, February 4th, 2009

better than other online regex tools.. only thing is the text limit in box. is it really accepting only 36 characters ot is it my browser?

Steven the Easily Amused   

Posted at 10:25am on Monday, February 16th, 2009

Really Nice... just wish I could have a longer string as alper noted above.

vobb   

Posted at 10:25am on Monday, February 23rd, 2009

Nice. Simple, effective and visual.

Nandu   

Posted at 10:08pm on Sunday, April 26th, 2009

I need a clarofication for the following:-
For Input :cathatcatcat
Case 1)the regex ^(c|a|t|h)+$ results in the last "t" being matched
BUT
Case 2)^(c|a|t)+$ results in 0 matches
In (case 1) and (case 2) shouldn't the first "c" get matched?Please explain..

kuddus   

Posted at 9:51am on Monday, May 4th, 2009

fine

Scott   

Posted at 12:40pm on Thursday, May 7th, 2009

Thanks for the tool. It helped me to study for my final!

Benit   

Posted at 10:21am on Tuesday, June 2nd, 2009

I want to capture the first occurrent of AND in the following statement :
C:/Test/AND/Check/Run_AND/AND.exe

I want to replace from C: to first occurrence of AND by some other path.
Please help me.

Anonymous   

Posted at 4:54pm on Thursday, June 11th, 2009

This is great tool. Thanks for making it available. I have one feature request: I'd like to be able to test regex expressions larger than 40 characters.

keith   

Posted at 6:12pm on Wednesday, June 17th, 2009

Nandu: In the first case, your regex is saying "match if the first character of the input is 'c' or 'a' or 't' or 'h' and every subsequent character is one of those characters, and the last character is one of those characters. So your result matches (bold) for all characters in the first case, and the group (in green) ends matching the last character t.

In the second case, the first character is in the group, but with 'h' omitted, when it gets to the 'h' in the input, it fails and there is no match.

In other words, your regex is saying: match if the input is a string of characters made up of any combination of one or more of the characters in the group from beginning to end.

LCB   

Posted at 2:31am on Wednesday, July 1st, 2009

How I can get the string:
11x-A-20 (ST)
1as-50tdt
(Atx-220)

The string can vary from numbers and letters, and combine hyphens (1,2,3,n or none) and parenthesis...

from this lines:
1 11x-A-20 (ST) 831771.8700
1 1as-50tdt 831771.8700
1 (atx-220) 831771.8700

I tried several regex but have no luck to catch it at all
anyone had any idea how to catch the string part of this?

test   

Posted at 6:44pm on Thursday, July 23rd, 2009

Awesome tool

Yash   

Posted at 7:44pm on Tuesday, July 28th, 2009

Thanks, it sorted out my stuff

Jess   

Posted at 11:24am on Wednesday, August 5th, 2009

WOWW!!!! SUPERB!!!!!!! it precisel;y told me wher i went wrong!!!!!

FarmnerGiles   

Posted at 2:54pm on Friday, August 7th, 2009

LCB: Try (d) ([dw-s()]*) d*.d*

BlueINK   

Posted at 8:41am on Monday, August 10th, 2009

LCB This Also Works
[0-9]s(.*)s[0-9]+

Blueink   

Posted at 8:54am on Monday, August 10th, 2009

wierd it auto removed the backlashses
there should be backlash behind the s's which mean Space

Anonymous   

Posted at 3:04am on Friday, August 21st, 2009

very use ful

femilopsy   

Posted at 1:30pm on Monday, September 14th, 2009

please i tried the regex (LA/09A).*(s)(T) to match that a string starts with LA/09A/.

please tell me where i have gone wrong

Dmytro   

Posted at 4:21pm on Thursday, September 17th, 2009

Very useful tool! The only suggestion it would be prefect if you allow longer strings for regex. Thanks guys!

chuck   

Posted at 12:44pm on Thursday, October 1st, 2009

very nice

Nicholas   

Posted at 9:21am on Friday, October 16th, 2009

the minimum character length is quite restrictive.

Nicholas   

Posted at 9:21am on Friday, October 16th, 2009

the minimum character length is quite restrictive.

Nicholas   

Posted at 9:22am on Friday, October 16th, 2009

oops, i meant max...

rohit   

Posted at 12:28am on Saturday, October 31st, 2009

thanks

Emil   

Posted at 8:00pm on Sunday, November 8th, 2009

brilliant

perlsgerl   

Posted at 5:36pm on Saturday, November 14th, 2009

very useful tester. thanks

uderman   

Posted at 6:09pm on Friday, February 26th, 2010

Very usefull, thank you!

Terziyski   

Posted at 7:29pm on Wednesday, March 31st, 2010

Very useful indeed !

kevin   

Posted at 1:17am on Friday, April 2nd, 2010

i'm having some trouble with this.
i have a string 'HASH(0x1f1c6950)'and i only want 0x1f1c6950.
can somebody please help me with this one because i'm very new at this.

thank you

CarlyKim28   

Posted at 10:36pm on Saturday, June 26th, 2010

All people deserve wealthy life and personal loans or secured loan can make it better. Because people's freedom bases on money state.

passer-by   

Posted at 9:37pm on Friday, July 2nd, 2010

You might want to consider expanding the regex field's text input limit.

My regex statement got truncated due to this limit.
^[a-f0-9]{8}-([a-f0-9]{4}-){3}[a-f0-9]{12}$

JWellington   

Posted at 9:16pm on Friday, July 9th, 2010

Ok, I got a string of text, for example: "725.50 87.50 541.50"
I am trying to make sure that each string has at least two instances of .50, why is this not working?

m/.50{2,}/g
I've even tried: m/.{2,}/g
It still does not work. Please help.

Andrew   

Posted at 5:41am on Wednesday, July 14th, 2010

For those who want to type strings longer than 40 chars, use this greasemonkey script.

// ==UserScript==
// @name Bigger regex textboxes
// @namespace http://com.example
// @include http://www.perlfect.com/articles/regextutor.shtml
// ==/UserScript==

var form = document.getElementById("regextutor");

if (form) {
var inputs = form.getElementsByTagName("input");
for (var i = 0; i < inputs.length; i++) {
inputs[i].setAttribute("size", 80);
inputs[i].setAttribute("maxLength", 500);
}
}