Front page | perl.perl5.porters |
Postings from November 2010
Abigail wrote:
> On Mon, Nov 29, 2010 at 11:22:20AM -0700, karl williamson wrote:
>> Dr.Ruud wrote:
>>> On 2010-11-29 03:18, karl williamson wrote:
>>>
>>>> I think that a utf8 regex pattern with the /d modifier should have
>>>> unicode semantics. But I'm open to contrary opinions.
>>>>
>>>> I had always assumed that it should, but it doesn't currently.
>>> How to define "a utf8 regex pattern"?
>>>
>>> Will /\d$re/ be a "utf8 regex pattern" if $re has the utf8-flag on,
>>> even if it only contains code points <= 255?
>>>
>> Yes.
>
>
> So, this will be fixed?
>
> my $e = "\xE8";
> my $re = "\N{WHITE SMILING FACE}";
>
> say $e =~ /[\w$re]/ ? "Match" : "No match";
> say $e =~ /\w|$re/ ? "Match" : "No match";
>
> __END__
> Match
> No match
>
>
Well, in my undelivered code space they both match. I don't understand
what the Match/No match is after the __END__. If you think they should
return different values, please explain.
Thread Previous
|
Thread Next