GNU bug report logs - #36167
[PATCH] Replace manually crafted hex regexes with [[:xdigit:]]

Previous Next

Full log

Message #95 received at 36167 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Andy Moreton <andrewjmoreton <at> gmail.com>
Cc: 36167 <at> debbugs.gnu.org
Subject: Re: bug#36167: [PATCH] Replace manually crafted hex regexes with
 [[:xdigit:]]
Date: Wed, 12 Jun 2019 19:07:29 +0300

> From: Andy Moreton <andrewjmoreton <at> gmail.com>
> Date: Wed, 12 Jun 2019 12:44:12 +0100
> 
> >> The [:alnum:] and [:alpha:] are based on (unspecified values of) the Unicode
> >> general-category property, but [:digit:] is not. Thus [:alnum:] includes
> >> other numeric characters that are not matched by [:digit:].
> >
> > It's true that [:alnum:] includes more numerical characters that
> > [:digit:], but what exactly needs to be clarified here?  Assuming you
> > mean clarified in the manual, that is.
> 
> As noted by Paul Eggert, the POSIX behaviour is different. It may be
> worth a note in the manual to draw attention to this difference.

It's strange to say in the manual what we do NOT do.  The current text
says:

  ‘[:digit:]’
       This matches ‘0’ through ‘9’.  Thus, ‘[-+[:digit:]]’ matches any
       digit, as well as ‘+’ and ‘-’.

I believe you suggest to add to this something like "Note that Posix
interpretation of '[:digit:]' is different"?  Given the crystal
clarity of the current text, wouldn't such addition confuse the
reader?

Btw, where's the reference for a different interpretation by Posix?  I
cannot find anything to that effect; even the Unicode UTS18
(http://www.unicode.org/reports/tr18/) says, while describing the
Posix equivalent of the Unicode regexp notations: "Non-decimal numbers
(like Roman numerals) are normally excluded".  What am I missing?

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.