Front page | perl.perl5.porters |
Postings from December 2011
On 12/23/2011 10:03 PM, Brian Fraser wrote:
>
>
> On Sat, Dec 24, 2011 at 12:37 AM, Karl Williamson
> <public@khwilliamson.com <mailto:public@khwilliamson.com>> wrote:
>
> On 04/13/2011 08:33 PM, brian d foy wrote:
>
> In article<3020.1302738770@chthon__>, Tom Christiansen
> <tchrist@perl.com <mailto:tchrist@perl.com>> wrote:
>
>
> What's wrong with quotemeta() someday changing to quote say,
> \p{Pattern_Syntax}
> character if and when Perl makes use of them?
>
>
> Why not do that now (as in 5.15) with the current characters that it
> would quote. Then, it's already there when we want to expand it.
>
>
> This seems like a reasonable solution. Does anyone have objections?
>
>
> I do, sorta. \p{Pattern_Syntax} isn't a superset of what quotemeta()
> currently uses; things like NULL and the control characters aren't
> included in the former, but are quoted by the latter.
> Seems easy enough to resolve though: quotemeta should escape all
> non-word character in the ASCII range, or those matching \p{Pattern_Syntax}.
>
Whatever solution we adopt should not change ASCII-range characters.
But since portions of the original thread have been cut-off, I want to
re-iterate that this is a form of the Unicode bug. quotemeta behaves
differently depending on whether the string is encoded in UTF-8 or not.
This has caused problems in the field. And I think it should be fixed
to be consistent whatever the encoding, in 5.16.
Thread Previous