regexp and valid-sre? Michael Montague (26 Nov 2013 03:34 UTC)
Re: regexp and valid-sre? Alex Shinn (26 Nov 2013 12:44 UTC)
Re: regexp and valid-sre? Peter Bex (26 Nov 2013 14:25 UTC)
Re: regexp and valid-sre? Michael Montague (26 Nov 2013 18:00 UTC)
Re: regexp and valid-sre? Peter Bex (26 Nov 2013 18:21 UTC)
Re: regexp and valid-sre? Michael Montague (26 Nov 2013 19:09 UTC)
Re: regexp and valid-sre? John Cowan (26 Nov 2013 18:24 UTC)
Re: regexp and valid-sre? Michael Montague (26 Nov 2013 19:17 UTC)
Re: regexp and valid-sre? Peter Bex (26 Nov 2013 19:23 UTC)
Re: regexp and valid-sre? Kevin Wortman (26 Nov 2013 19:52 UTC)
Re: regexp and valid-sre? Michael Montague (26 Nov 2013 19:59 UTC)
Re: regexp and valid-sre? Kevin Wortman (27 Nov 2013 23:33 UTC)
Re: regexp and valid-sre? John Cowan (27 Nov 2013 23:42 UTC)
Re: regexp and valid-sre? Arthur A. Gleckler (30 Nov 2013 14:55 UTC)
Re: regexp and valid-sre? Michael Montague (26 Nov 2013 18:02 UTC)
Re: regexp and valid-sre? John Cowan (26 Nov 2013 18:19 UTC)
Re: regexp and valid-sre? Michael Montague (26 Nov 2013 19:11 UTC)

Re: regexp and valid-sre? Michael Montague 26 Nov 2013 19:11 UTC

On 11/26/2013 10:19 AM, John Cowan wrote:
> Michael Montague scripsit:
>
>> I propose dropping the requirement that 'regexp' can be called with
>> an already compiled <re>. It provides no additional functionality
>> and it makes the specification of 'regexp' less clear.
> It provides the ability to compile repeatedly used parts of regular
> expressions.  In a context where a large number of regular expressions
> are in use that have factors in common, this can save a good deal of
> time and possibly space.

That would require calling 'regexp' with a <sre> containing an <re>. I
don't see where in the grammar that you are allowed to have an <re> as
part of an <sre>.