Perl: /^(?:(d+)|::)* what does that equate to?

[macawgumbo]

what exactly would /^(?:(d+)|::)* equate to if you say would including it in a grep statement?

[Hu]

what exactly would /^(?:(d+)|::)* equate to if you say would including it in a grep statement?

That expression expands as:

^: Match at beginning
*: zero or more of these
(?:()): groups like bare parentheses, but does not affect backreferences
d+: one or more lowercase d
|: alternative
::: a pair of colons

This should be parsed as: a sequence of d interspersed with pairs of colons, zero or more times. I think the equivalent grep would be: grep -E '^((d+)|::)*'.

[frostschutz]

...and this grep is pretty much pointless (unless you're using color grep or something), as it returns every line. Every line has a beginning of the line, and it contains ((d+)|::) at least zero times.

[Hu]

...and this grep is pretty much pointless (unless you're using color grep or something), as it returns every line. Every line has a beginning of the line, and it contains ((d+)|::) at least zero times.

Sure, but he didn't say it had to be useful. Besides, since he left off the trailing slash for his perl pattern, he might have some other expression that he is appending to this.

[nutbar]

^(?:(d+)|::)*

If you're still having a bit of trouble reading this, I find it easier to read it from the inside out by sectioning things off separated by the () brackets. What it translates to is close to what Hu said, but not exactly.

Matching from the beginning of the line, find one or more "d"'s or a set of colons "::" zero or more times.

The ^ matches the beginning of a line.
The () brackets are submatches (just think of it as a way to separate one set of match settings from another).
The (?:...) sequence means a submatch but the ?: part tells it to NOT store this submatch as a backreference (unused in this query) - useful since you're only allowed 9 submatches usually (what ever the amount may be, it's finite).
The (...|...) sequence is a submatch where the pipe is just like saying OR between the various matches it separates.
The + quantifier means one or more occurances
The * quantifier means zero or more occurances

If you want to begin to really understand regular expressions, I suggest you start with basic RE's by reading the regex man page (which also covers extended RE's as well). Once you have a firm grasp on POSIX RE's, you can understand Perl RE's much easier.

[wynn]

If you want to begin to really understand regular expressions, I suggest you start with basic RE's by reading the regex man page (which also covers extended RE's as well). Once you have a firm grasp on POSIX RE's, you can understand Perl RE's much easier.

"Mastering Regular Expressions: Powerful Techniques for Perl and Other Tools" by Jeffrey E. F. Friedl is very good too.

[macawgumbo]

Yes, this statement is absolutely pointless, but somebody thought I should spend time figuring it out. You can see how productive we are here in the office throwing these types of expressions around to solve. Thanks for the recommendations on where to look for reference as I am sure I will be reading up on it in the near future.

[wynn]

...and when you're tired of life, you can always try to fathom "perl one-liners" like those in http://www.consultix-inc.com/consultix/ ... rlcmds.pdf