[m-rev.] Re: for review: parsing_utils improvements

[Paul Bone]

On Thu, Oct 01, 2009 at 05:30:38PM +1000, Ian MacLarty wrote:
> On Thu, Oct 01, 2009 at 01:49:48PM +1000, Ralph Becket wrote:
> > Ian MacLarty, Wednesday, 30 September 2009:
> > > On Wed, Sep 30, 2009 at 12:12 PM, Ian MacLarty
> > > <maclarty at csse.unimelb.edu.au> wrote:
> > > > Here are some benchmarks I did with the sparql parser with an input of
> > > > 4.8M (times are an average of 3 runs):
> > > 
> > > Correction: 48M (48 100 031 bytes).
> > 
> > Sorry, I thought I'd replied to this before: you've convinced me this is
> > a good idea!
> 
> Here's an interdiff for review.  Note the whitespace bug fix.  It was
> using char.is_whitespace for the first character, but then calling
> skip_whitespace/3 which calls the user defined whitespace function.
> I changed it so it calls char.is_whitespace for all characters which is
> its documented behaviour.

I'm pretty sure I noticed this when I reviewed Ralph's patch that introduced
user defined whitespace.  I thought that the behaviour was correct but did not
realise that the documentation did not agree with the implementation.  Perhaps
it's the documentation that should be changed?  (I'm not sure though, it's a
vague memory).


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 489 bytes
Desc: Digital signature
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20091002/fb5a3cce/attachment.sig>