G'day mmkstarr,

Welcome to the Monastery.

"I am wondering how far one can get in pearl without learning regex?"

[Note: the language is called Perl and the program is called perl. A "pearl" is something you'll find in an oyster.]

Perl provides much the same constructs as other programming languages. (See "perlintro -- a brief introduction and overview of Perl".) You can write Perl code without using regexes. However, unless you're coding in total isolation, you will often encounter regexes in code written by others: without being able to understand regexes, you won't get very far. As an absolute bare minimum, I suggest you read, and fully comprehend:

• perlintro: Regular expressions
• perlrequick - Perl regular expressions quick start

"Is regex fluency a huge boon to a perl user?"

The short answer is: "Yes". Regexes allow you to write very succinct code, often in a single line, which otherwise would require dozens, or even hundreds, of lines of code to accomplish.

Having said that, it's important not to fall into the trap of always reaching for a regex just because it can be used. Perl provides a number of functions and operators which may be more appropriate than a regex and, more often than not, are much faster than the equivalent regex. Here's some examples:

• Consider whether "index - find a substring within a string" will achieve your goal before reaching for a "regex match".
• Consider whether "substr - get or alter a portion of a string" will achieve your goal before reaching for a "regex substitution".
• When matching an entire string, $string eq 'text' will invariably be a better choice than $string =~ /^text$/.

"If so, what are some of the best regex tutorials you have come upon?"

In the first instance, I suggest you look at:

• perlretut - Perl regular expressions tutorial

There's a lot of documentation on Perl's regular expressions. Use the following, as required, for reference:

• perlop: Regexp Quote-Like Operators
• perlvar: Variables related to regular expressions
• perlre - Perl regular expressions
• perlreref - Perl Regular Expressions Reference
• perlrecharclass - Perl Regular Expression Character Classes
• perlrebackslash - Perl Regular Expression Backslash Sequences and Escapes

Hello mmkstarr, and welcome to the Monastery!

You can program in Perl without using regular expressions. On the other hand, Perl’s built-in regular expression engine is extremely powerful — in fact, it’s one of Perl’s main features — so any effort you put into learning Perl regexen will be richly repaid as you put them to use in your scripts.

The official tutorial is a good starting point: perlretut. After that, you should read through the “Pattern Matching” chapter of the Camel Book,¹ although you can leave the final section, “Fancy Patterns,” till later. But, of course, the only way you’ll really learn regexen is by trial and error: write your own, experiment, and learn from your mistakes!

BTW, the language is called “Perl” (capital P, no A); the interpreter is called “perl” (no capital).

¹Chapter 5 of Programming Perl (4^th Edition, 2012) by Tom Christiansen, brian d foy & and Larry Wall with Jon Orwant.

Hope that helps,

If you're interested in getting a good reference book on regular expressions in general, you might want to get a copy of Mastering Regular Expressions by Jeffrey Friedl. It may not necessarily be the best place to start learning, but it is a very good reference book to own.

As for how useful regular expressions are, I still remember my reaction to an example from that book. I can't remember all of the details but the example was describing the task of finding double occurrences of words (such as "the the" while ignoring case differences. And it should include cases of last word of one line and first word of the next line being the same. And the task was not just finding these occurrences, but highlighting them and copying those lines into another file. My initial thought was that had to be hundreds of lines of code, but the author said it could be done in something like 5 lines of code. At that time, I just couldn't believe that such a complicated task could be accomplished with very little code. When I read that, I knew that I really needed to learn regular expressions.

I use regular expressions in almost every program in one way or another.

Start at the beginning: perlrequick.

Otherwise, just to confirm, regex is really a central skill for using Perl, because it is so deeply integrated within the language. Perl is know to be a language of choice for manipulating text data because of regexes, and it don't use them, your abilities to manipulate data becomes actually quite poor, because many things that can be done with special functions on other languages have to be done with regexes in the broader sense in Perl.

And, BTW, it is much less difficult to learn than you might think: regex may seem impressive if you don't know them, but there is nothing really difficult about them.

Great responses by the other Monks, so I'll just add a couple of things. A couple of debugging tools that can help you understand regexes better are the built-in regex debugger, use re 'debug';, and YAPE::Regex::Explain, both will allow you to display what your regexes are doing, or paste in regexes you find elsewhere and explain to you how they work.

Then, when you're ready to sink a couple dozen hours having "fun" with regexen: http://regex.alf.nu/. I remember using another good one of these (kinds of sites) but can't seem to find it again.

Wow! Before I got bored, I tried to play for a while:

Warmup (207)
Anchors (206)
Ranges (199)
Backrefs (201)
Abba (191)
A man, a plan (169)
Prime (256)
Four (199)
Order (194)
Triples (560)
Glob (348)
Balance (13)
Powers (0)
Long count (250)
Alphabetical (0)
Powers 2 (10)
You have 3003 points.
[download]

($q=q:Sq=~/;[c](.)(.)/;chr(-||-|5+lengthSq)`"S|oS2"`map{chr |+ord
}map{substrSq`S_+|`|}3E|-|`7**2-3:)=~y+S|`+$1,++print+eval$q,q,a,
[download]

That's an awesome little tool there :)

Migth be the www.debuggex.com site?

I also used with profit an old tool in lisp: the regex coach

Also for completeness:

• regex deubugger on CPAN
• Re^2: ppiwx / wxPPI / wxppixregexp ppixregexplain.pl wxPPIxregexplain.pl PPIx::Regexp::xplain because YAPE::Regex::Explain is dead
• Visual (perl/Tk) regex tweaking utility
• VRegExp
• 7 Stages of Regex Users

So the whole list is worth to be boomarked!

L*

There are no rules, there are no thumbs..
Reinvent the wheel, then learn The Wheel; may be one day you reinvent one of THE WHEELS.

(I say "most of what you learn", because I think the Perl regex engine is more extensive - offers a lot more stuff - than other implementations. Still, all the basic concepts and syntax are portable to other scripting languages, SQL and unix shell commands, among other things.)

You can do a lot of programming without using a regex. I know. I did it. But if you use sites like this, you will see posts that mention regexes and then, when you are trying to write something, you will realise that a regex is the best tool. That's the best tutorial you can find - a need to solve a problem of your own. Unless you have to pass an exam, don't try learning something you don't use, as you will quickly forget what you learned. Once you get to wanting to learn something, try writing the regex yourself. If you get stuck, post here with what you are trying to do and what you have written. When I've done this, I've found the Monasteriat extremely helpful.

As others have said, some regexes are very simple. But as I think you realised before the other posts, regexes are a very powerful tool (I don't know them at all well, but I know that much). And in addition to the tools others have mentioned, I have recently taken to using Damian Conway's Regexp::Debugger, which I have found most helpful.

Regards,

John Davies

davido has a nice online Perl Regex Tester that might help while you are learning (and when you start to forget).