Front page | perl.perl5.porters |
Postings from October 2009
On Thu, Oct 29, 2009 at 12:46:30PM +0000, Paul LeoNerd Evans wrote:
> On Wed, 28 Oct 2009 09:25:30 +0200
> Yuval Kogman <nothingmuch@woobling.org> wrote:
>
> > How difficult would it be to introduce special chars which aren't
> > charclasses, which are probably more suitable for what people want anyway
> > (things that agree with grok_number, with rules for natural numbers,
> > integers, decimal fractions, and floating point notation)?
> >
> > Seems like the distinction between matching a character that is a digit vs.
> > matching ascii digits is mostly about what you do with the numbers
> > afterwords. Perhaps it's better to just remove the extra duplication?
>
> Vim uses foo vs \_foo to distinguish whether a linefeed is included or
> not; e.g.
>
> abc. <= literal followed by anything except linefeed
> abc\. <= literal followed by anything including linefeed
>
> Maybe we can find some suitable mangling to apply to \w, \d, \s, etc...
> to say "with extra Unicode chars like these"
\begin{not-really-serious}
I suggest \ḋ, \ṡ, and \ẇ for the Unicode character classes, and
\d, \s, \w for the ASCII versions.
For those not able to read my suggestions, it's
\N{LATIN SMALL LETTER D WITH DOT ABOVE} \x{1E0B}
\N{LATIN SMALL LETTER S WITH DOT ABOVE} \x{1E61}
\N{LATIN SMALL LETTER W WITH DOT ABOVE} \x{1E87}
\end{not-really-serious}
Abigail
Thread Previous
|
Thread Next