Parse an array with regular expression

Question

I have JS file and I need to parse that array in it:

something: ['Prefix.All','Ignore.Me','Prefix.Another']

I need to get all elements from "something" array with defined prefix. Now I can get it, if array will contain only one element. My regexp:

String: something: ['Prefix.All']
Can get element with:
/something\s*:\s*\[['"](Prefix[a-zA-Z0-9.]+)['"]]/

But how to find many elements in array? And how to find more than one "something" array in file?

Is your something always going to start with Prefix.All I would suggest you to split it by comma... — bonCodigo
– bonCodigo, Commented Jan 15, 2013 at 7:37
What regex engine are you using for this? Unless it's .NET, you can't do it with a single regex. — Tim Pietzcker
– Tim Pietzcker, Commented Jan 15, 2013 at 7:56
uh, take string, find first [, find last ]. take substring in [ ], split it by ,. find all substrings, that have prefix Prefix. Why you think, that regexp can really helps you? — gaussblurinc
– gaussblurinc, Commented Jan 15, 2013 at 8:08
In that case I will find: required:[..], anything:[..], another:[..] e.t.c — nkuhta
– nkuhta, Commented Jan 15, 2013 at 8:11

Tim Pietzcker · Accepted Answer · 2013-01-15 08:19:42Z

3

In PHP you need to do it in two steps (which is probably better anyway - cramming everything into a single regex isn't always the best idea).

First, match the array you're looking for. Note that this regex requires that there are no brackets between the delimiters [ and ], so nested arrays or strings that contain brackets will cause the regex to fail (as in "match the wrong text"):

if (preg_match('/\bsomething: \[([^[\]]*)\]/', $subject, $regs)) {
    $result = $regs[1];
}

This will find the first occurrence of a string like something: ['Prefix.All','Ignore.Me','Prefix.Another'] and put 'Prefix.All','Ignore.Me','Prefix.Another' into $result.

Then you can get all the matches from that:

preg_match_all('/\bPrefix[^\']*/', $result, $final, PREG_PATTERN_ORDER);
$final = $final[0];

answered Jan 15, 2013 at 8:19

Tim Pietzcker

337k59 gold badges520 silver badges572 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Jay Gattuso · Accepted Answer · 2013-01-15 07:58:26Z

0

In regex you can group a token with (), and ask for it many times with *.

So: (something)*would give you every instance of the word 'something'.

Further: (__your__token__here__)* finds every instance of your token.

Paste some of your array here: http://gskinner.com/RegExr/ and use the regex builder to test your regex

Does that help?

edited Jan 15, 2013 at 7:58

answered Jan 15, 2013 at 7:34

Jay Gattuso

4,14812 gold badges40 silver badges51 bronze badges

4 Comments

nkuhta Over a year ago

I tried simple example. Find create[\d] in that string: "create2 create3 create4". But ()* don't work. Here is my repexp: /(create[\d])*/ . It can't do that. Mistake?

Jay Gattuso Over a year ago

See edit - I always use an online regex tester to test what I want to do until I get it right - without seeing your array its a bit hard to figure out what you are trying to achieve. For your example, you just need (create[\d])

Tim Pietzcker Over a year ago

(something)* matches "", "something", "somethingsomething" etc., but it doesn't do at all what the OP wants.

Tim Pietzcker Over a year ago

Not really. In his example, two of the "tokens" he's looking for are separated by a token he does not want to match, so * wouldn't work.

Collectives™ on Stack Overflow

Parse an array with regular expression

2 Answers 2

Comments

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related