Front page | perl.perl5.porters |
Postings from October 2009
On Fri, Oct 30, 2009 at 11:30:23AM +0000, David Cantrell wrote:
> On Thu, Oct 29, 2009 at 11:18:05PM +0100, Abigail wrote:
>
> > Noone is suggesting to have \d *start* matching non-ASCII digits.
>
> perldoc perlunicode can be read to indicate that \d does exactly that.
>
> Under "speed", it says:
>
> " As an example, the Unicode properties (character classes) like
> "\p{Nd}" are known to be quite a bit slower (5-20 times) than
> their simpler counterparts like "\d" (then again, there 268
> Unicode characters matching "Nd" compared with the 10 ASCII
> characters matching "d"). "
>
> which looks like saying that \d matches [0123456789]. But it also says
> elsewhere that it does match funny forn characters.
The quote as is suggests that. But it becomes somewhat clearer if
you take the sentence preceeding your quote into account:
In general, operations with UTF-8 encoded strings are still slower.
The point isn't so much \p{Nd} vs. \d, but matching against non-UTF-8
encoded strings, vs matching against UTF-8 encoded strings.
Abigail
Thread Previous
|
Thread Next