Hey JM,
My somewhat more robust pattern:
^([[:alpha:]]+://.+?[?&]mod=)(.+?)(:.+)
Detailed examination:
^ Beginning of line
( Start capture
[[:alpha:]] Alphabetic characters
+ One or more
:// Literal character sequence
.+? Any character 1 or more non-greedy
[?&] Either “?” or “&”
mod= Literal character sequence
) End capture
( Start capture
.+? Any character 1 or more non-greedy
) End capture
( Start capture
: Literal character
.+ Any character 1 or more greedy
) End capture
To analyze in on regex101.com you have to adjust the pattern slightly by escaping the two forward slashes and adding the ‘m’ switch.
^([[:alpha:]]+:\/\/.+?[?&]mod=)(.+?)(:.+)
Forward slashes are used by regex101 as pattern delimiters - i.e. they delimit the beginning and the end of the pattern – and this is why they must be escaped in the pattern for regex101.com.
You must realize the analyzer is representing code-based RegEx usage like in Perl, JavaScript, others… – and NOT a GUI-based search/replace mechanism. This is confusing if you've never seen a programming language do regular expressions.
Perl search/replace syntax looks like this:
s/<search-pattern>/<replace-pattern>/<switches>;
An actual working Perl script that replaces every character (.) with a bullet (•).
#! /usr/bin/env perl -sw
my $strVar = "Now is the time for all good men to come to the aid of their country.";
$strVar =~ s/./•/g;
print $strVar;
Seeing actual examples of sed, awk, and Perl give this business of delimiters and switches context that helps users understand instead of just memorize.
Back to the pattern – here it is on regex101.com:
Here it is with free-spacing turned on and comments:
Generally I prefer to use a RegEx analyzer app on my system to the online ones, but of course that's more difficult to share.
RegExRX is one of the better inexpensive RegEx analyzers. ($4.99 direct or from the app-store, and there's a demo.)
It doesn't give you the comprehensive explanation regex101.com does, but you don't have to fool with the pseudo code separators and such in //.
I think this makes it a little closer to what the user sees in GUI-based systems.
The best analyzer I'm acquainted with is the Windows App RegexBuddy.
(When I upgrade my hardware next time (and have more memory) I'll be able to use Parallels and will buy a copy despite the fairly hefty price – $39.95 U.S.)
-Chris
