Greetings.
Recently I have been working on a number of utilities for regular
expressions, one of which is similer to String::Random. I have taken the
liberty of using the namespaces specified, but would be interested to know
if they present a problem, or if their functionality has been covered
elsewhere. All of the packages are set up in OO style.
(Apologies if I have missed out on some protocol issue here, this my first
post.)
Also a request: Could we have a pragma that can be used to replace the dual
use of use strict and use warnings? It would be cool if
use caution;
was an alias for
use strict;
use warnings;
Also the name seemed apropriate. :-)
-------------------------------------------------------------------
String::RegEx::.....
-- This is the umbrella class that I have used for my classes. I thought
about extending String::Random, but the functionality I have been working on
covers much more than String::Random does.
It provides a single interface into the other classes I have been writing,
so as to enable me to avoid typing long use lists.
Major functions:
String::RegEx::generate()
Takes a regular expression and generates random data based on it. Supports
basically all perl regular expressions that dont involve backreferences and
zero-width assertions, which are currently a work in progress.
An example would be:
my $regex=String::RegEx->new();
warn $regex->generate('(?:(perl|PERL|[Pp]erl) is (?:very ){1,5} cool!\n)+');
Produces (Edited for size...):
PERL is very very cool!
Perl is very very very very cool!
PERL is very cool!
PERL is very very very very cool!
String::RegEx::gen_allcases()
An experimental routine that will generate a set of strings, one per 'part'
of the regex. The idea being that if you are using a rgex as a input
cleaner or the like you use this to see if your pattern would match some
string that you didnt expect. Work In Progress.
String::RegEx::pretty()
Takes a regular expression and outputs it indented and the like...
Example using previous regex. Whitespace in the original is convert to
escape sequences. The pretty version should run under the x modifier just
as the original...
(?:
(
perl
|
PERL
|
[Pp]erl
)
\x20is\x20
(?:very\x20){3,5}
\x20cool!
\n
)+
String::RegEx::build_from_list()
Takes a list of words and generate a regular expression that will match
those words. Inspired by J Friedl and his book. Unfortunately I forgot my
laptop network cable today so no example avail :-(
String::RegEx::make_trans()
Automate the eval'ing of runtime tr///.
(Editor says more detail: Takes a string containing alphabet and target
alphabet and returns a subref that will do the translation. For
convienience)
String::RegEx::build_longest_matcher
Kind of a special purpose tool, takes a list like build_from_list and
returns a sub ref that will match the words in the list LONGEST first.
String::RegEx::build_list_matcher
Returns a subroutine reference that will match the words in the list. Has
options to determine if it will make one huge regex or a bunch of smaller
ones that are applied in turn.
----------------------------------------------------------------------
I really would like to release this stuff to CPAN and would appreciate any
help or advice regarding this that may had.
DeMerphQ
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
