# Regexp Capture Access **URL:** \ **Category:** Ruby\ **Created:** [April 2, 2008, 3:20pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679 "2008-04-02T15:20:11Z")\ **Posts on this page:** 8\ **Page:** 1
**Author:** ![ole](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/o/fbc32d/32.png) [@ole](https://www.ruby-forum.com/u/ole)\ **Post date:** [April 2, 2008, 3:20pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679/1 "2008-04-02T15:20:11Z")
Given that: %r{(\w,)+}.match(‘a,b,c’)[0] #=\> “a,b,” and %r{(\w,)+}.match(‘a,b,c’)[1] #=\> “b,” How do I access the capture that contains “a,”? ---
**Author:** ![ole](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/o/fbc32d/32.png) [@ole](https://www.ruby-forum.com/u/ole)\ **Post date:** [April 2, 2008, 4:05pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679/2 "2008-04-02T16:05:18Z")
Try: %r{(\w,)+}.match(‘a,b,c’).to\_s[0…1] #=\> “a,” 2008/4/2, Oliver S. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid): ---
**Author:** ![ole](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/o/fbc32d/32.png) [@ole](https://www.ruby-forum.com/u/ole)\ **Post date:** [April 2, 2008, 4:28pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679/3 "2008-04-02T16:28:35Z")
On Wed, Apr 2, 2008 at 3:20 PM, Oliver S. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > Posted via [http://www.ruby-forum.com/](http://www.ruby-forum.com/). Use scan “a,b,c”.scan(/\w+,/) → [‘a,’, ‘b,’] and in order to have the whole match you have to join the result of scan again. HTH Robert – [http://ruby-smalltalk.blogspot.com/](http://ruby-smalltalk.blogspot.com/) * * * Whereof one cannot speak, thereof one must be silent. Ludwig Wittgenstein ---
**Author:** ![ole](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/o/fbc32d/32.png) [@ole](https://www.ruby-forum.com/u/ole)\ **Post date:** [April 2, 2008, 4:32pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679/4 "2008-04-02T16:32:08Z")
The reason why it’s acting like this is because Regexes try to find the largest possible match from your expression. Hence, your expression %r{(\w,)+}.match(‘a,b,c’)[0] will return the first match (and largest) “a,b,” Im sure you’re asking this for a bigger reason, so you have to refine your regular expression for that specific purpose. if you just want to split based on a “,” then simply do ‘a,b,c’.split(’,’) which returns an array [“a”, “b”, “c”] If any other, please post the actual problem. HTH Mutahhir. ---
**Author:** ![ole](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/o/fbc32d/32.png) [@ole](https://www.ruby-forum.com/u/ole)\ **Post date:** [April 2, 2008, 4:33pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679/5 "2008-04-02T16:33:19Z")
On Wed, Apr 2, 2008 at 8:20 AM, Oliver S. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > Given that: > > %r{(\w,)+}.match(‘a,b,c’)[0] #=\> “a,b,” > > and > > %r{(\w,)+}.match(‘a,b,c’)[1] #=\> “b,” > > How do I access the capture that contains “a,”? Maybe leave the plus symbol (+) out? r = /(\w,)/ r.match(‘hi,a,b,c’)[1] =\> “i,” I’d probably use #scan, or even #split, instead. Todd ---
**Author:** ![ole](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/o/fbc32d/32.png) [@ole](https://www.ruby-forum.com/u/ole)\ **Post date:** [April 2, 2008, 5:28pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679/6 "2008-04-02T17:28:25Z")
Nevermind. I was trying to parse CSV. I thought pre-written tools weren’t capable of understanding quoted values with commas inside but I was wrong so I can use them. I do find it amazing that you can’t access multiple matches for a single sub-pattern though. That’s a serious limitation IMO. ---
**Author:** ![ole](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/o/fbc32d/32.png) [@ole](https://www.ruby-forum.com/u/ole)\ **Post date:** [April 2, 2008, 4:40pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679/7 "2008-04-02T16:40:30Z")
On Wed, Apr 2, 2008 at 9:27 AM, Robert D. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > Use scan > > “a,b,c”.scan(/\w+,/) → [‘a,’, ‘b,’] > and in order to have the whole match you have to join the result of scan again. > > HTH > Robert Yeah, I was thinking of that regexp also, but I assumed the OP wanted the last word letter before a comma. Oliver, what is it that you’re looking for? Todd ---
**Author:** ![ole](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/o/fbc32d/32.png) [@ole](https://www.ruby-forum.com/u/ole)\ **Post date:** [April 2, 2008, 5:46pm UTC](https://www.ruby-forum.com/t/regexp-capture-access/133679/8 "2008-04-02T17:46:03Z")
On Wed, Apr 2, 2008 at 10:28 AM, Oliver S. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > Nevermind. I was trying to parse CSV. I thought pre-written tools > weren’t capable of understanding quoted values with commas inside but I > was wrong so I can use them. > > I do find it amazing that you can’t access multiple matches for a single > sub-pattern though. That’s a serious limitation IMO. I’m not sure I understand this correctly, because I think your pattern was wrong. Todd