Regexp Arity
 

Just ran into a need to know how many parenthetical groupings a Regexp has.
Would #arity for Regexp be a good idea?

T.

Re: Regexp Arity
 

Hi,

In message "Re: Regexp Arity"
on Thu, 23 Sep 2004 03:26:44 +0900, "trans. (T. Onoma)" <transami@runbox.com> writes:

|Just ran into a need to know how many parenthetical groupings a Regexp has.
|Would #arity for Regexp be a good idea?

I'm not sure if it should be named 'arity'.

matz.

Re: Regexp Arity
 

On Thu, 23 Sep 2004 03:54:31 +0900, Yukihiro Matsumoto
<matz@ruby-lang.org> wrote:
> In message "Re: Regexp Arity"
> on Thu, 23 Sep 2004 03:26:44 +0900, "trans. (T. Onoma)" <transami@runbox.com> > |Just ran into a need to know how many parenthetical groupings a Regexp has.
> |Would #arity for Regexp be a good idea?
> I'm not sure if it should be named 'arity'.

Regexp#groups ?
Regexp#groupcount?

-austin
--
Austin Ziegler * halostatue@gmail.com
* Alternate: austin@halostatue.ca
: as of this email, I have [ 6 ] Gmail invitations

Re: Regexp Arity
 

On Wednesday 22 September 2004 20:26, trans. (T. Onoma) wrote:
> Just ran into a need to know how many parenthetical groupings a Regexp has.
> Would #arity for Regexp be a good idea?


hmm...

irb(main):006:0> parser = Parser.new('a(b(?:c(x)c)d)')
[snip]
irb(main):007:0> class << parser
irb(main):008:1> attr_reader :number_of_captures
irb(main):009:1> end
=> nil
irb(main):010:0> parser.number_of_captures
=> 2
irb(main):011:0>


why have I never thought of this?

--
Simon Strandgaard

Re: Regexp Arity
 

On Wednesday 22 September 2004 02:54 pm, Yukihiro Matsumoto wrote:
> Hi,
>
> In message "Re: Regexp Arity"
>
> on Thu, 23 Sep 2004 03:26:44 +0900, "trans. (T. Onoma)"
<transami@runbox.com> writes:
> |Just ran into a need to know how many parenthetical groupings a Regexp
> | has. Would #arity for Regexp be a good idea?
>
> I'm not sure if it should be named 'arity'.

Perhaps not, but I really can't think of a better word. Everything else seems
long or odd sounding:

#group_count
#parentheticals_count
#number_of_subexpressions

Hmm... that brings up a good point. Would zero-width positive lookheads and/or
non-backreferencing groups be counted?

T.

--
( o _ カラチ
// trans.
/ \ transami@runbox.com

I don't give a damn for a man that can only spell a word one way.
-Mark Twain

Re: Regexp Arity
 

Hi --

On Thu, 23 Sep 2004, trans. (T. Onoma) wrote:

> On Wednesday 22 September 2004 02:54 pm, Yukihiro Matsumoto wrote:
> > Hi,
> >
> > In message "Re: Regexp Arity"
> >
> > on Thu, 23 Sep 2004 03:26:44 +0900, "trans. (T. Onoma)"
> <transami@runbox.com> writes:
> > |Just ran into a need to know how many parenthetical groupings a Regexp
> > | has. Would #arity for Regexp be a good idea?
> >
> > I'm not sure if it should be named 'arity'.
>
> Perhaps not, but I really can't think of a better word. Everything else seems
> long or odd sounding:
>
> #group_count
> #parentheticals_count
> #number_of_subexpressions
>
> Hmm... that brings up a good point. Would zero-width positive lookheads and/or
> non-backreferencing groups be counted?

To back up slightly: I can't help wondering under what conditions you
would need to know this. Can you present the problem you were trying
to solve? Maybe there's a simpler way.


David

--
David A. Black
dblack@wobblini.net

Re: Regexp Arity
 

On Wednesday 22 September 2004 21:47, David A. Black wrote:
> > <transami@runbox.com> writes:
> > > |Just ran into a need to know how many parenthetical groupings a Regexp
> > > | has. Would #arity for Regexp be a good idea?
[snip]
> > Hmm... that brings up a good point. Would zero-width positive lookheads
> > and/or non-backreferencing groups be counted?
>
> To back up slightly: I can't help wondering under what conditions you
> would need to know this. Can you present the problem you were trying
> to solve? Maybe there's a simpler way.


Maybe I should advertise some more for my regexp package

irb(main):001:0> require 'regexp'
=> true
irb(main):002:0> puts /this(feels(like)lisp|scheme)/.tree
+-Sequence
+-Inside set="t"
+-Inside set="h"
+-Inside set="i"
+-Inside set="s"
+-Group capture=1
+-Alternation
+-Sequence
| +-Inside set="f"
| +-Inside set="e"
| +-Inside set="e"
| +-Inside set="l"
| +-Inside set="s"
| +-Group capture=2
| | +-Sequence
| | +-Inside set="l"
| | +-Inside set="i"
| | +-Inside set="k"
| | +-Inside set="e"
| +-Inside set="l"
| +-Inside set="i"
| +-Inside set="s"
| +-Inside set="p"
+-Sequence
+-Inside set="s"
+-Inside set="c"
+-Inside set="h"
+-Inside set="e"
+-Inside set="m"
+-Inside set="e"
=> nil
irb(main):003:0>

to install just type

rpa install re

--
Simon Strandgaard

Re: Regexp Arity
 

On Wednesday 22 September 2004 03:47 pm, David A. Black wrote:
> To back up slightly: I can't help wondering under what conditions you
> would need to know this. Can you present the problem you were trying
> to solve? Maybe there's a simpler way.

I'm pattern matching a document. When a match is found the match/submatches
are passed to various procs. If the regexp has groupings I want to pass the
group matches, if not just the whole match. I could just go by the matchdata
length, but the regexp is stored in a dedicated "token" class and that class
may need to be initialized differently based on that "arity".

But granted, I did just work out a different approach, so as of this moment I
won't need it. Although I see no good reason that information shouldn't be
available. As I think Simon pointed out with his engine, "it is in there".

Thanks,
T.

Re: Regexp Arity
 

On Thursday, September 23, 2004, 4:56:23 AM, Austin wrote:

> On Thu, 23 Sep 2004 03:54:31 +0900, Yukihiro Matsumoto
> <matz@ruby-lang.org> wrote:
>> In message "Re: Regexp Arity"
>> on Thu, 23 Sep 2004 03:26:44 +0900, "trans. (T. Onoma)"
>> <transami@runbox.com> > |Just ran into a need to know how many
>> parenthetical groupings a Regexp has.
>> |Would #arity for Regexp be a good idea?
>> I'm not sure if it should be named 'arity'.

> Regexp#groups ?
> Regexp#groupcount?

Regexp#ngroups

Gavin

Re: Regexp Arity
 

* Gavin Sinclair <gsinclair@soyabean.com.au> [Sep 23, 2004 01:50]:
> > Regexp#groups ?
> > Regexp#groupcount?
>
> Regexp#ngroups

Best by far, so far,
nikolai

--
::: name: Nikolai Weibull :: aliases: pcp / lone-star / aka :::
::: born: Chicago, IL USA :: loc atm: Gothenburg, Sweden :::
::: page: www.pcppopper.org :: fun atm: gf,lps,ruby,lisp,war3 :::
main(){printf(&linux["\021%six\012\0"],(linux)["have"]+"fun"-97);}