Re: forgot: (was: Re: query-replace-regexp behaves weird (bug?))
>>>> "Matt" == Matt Tucker
<tuck(a)whistlingfish.net> writes:
Matt> -- Jani Averbach <jaa(a)cc.jyu.fi> spake thusly:
> On Sat, 25 May 2002, Norbert Koch wrote:
>
>>
>> I guess, [:space:] is a character class (correct wording?) and I
>> believe, XEmacs doesn't grok this. As an effect, the regexp matcher
>> sees a set of single characters to match against.
>>
>
> Ouh, of cource, I have to keep a break
>
> [[:space:]]
Matt> XEmacs won't grok that, either. The character class for
Matt> space is "\s-". Take a look at the info page
Matt> "(xemacs)Regexps"; do:
Hi Matt,
I think
C-h i g (xemacs)Syntax Entry
needs a patch.
It only lists:
` '
The class of whitespace characters.
even though
`-'
is supported also and has the benefit of being compatible with Emacs,
AFAIK.
How about this change, for which I could submit a patch:
`-'
The class of whitespace characters. Please don't use the
formerly advertised ` ', which is not supported by GNU Emacs.
Matt> C-h i g (xemacs)Regexps
Matt> and search on that page for "\sCODE". Note that \s- will
Matt> pick up newlines also, which is often not what you want. You
Matt> can also do:
Matt> [ <tab>]
Right, interactively one has to type
[ SPC C-q TAB ]
even though when written as a string in lisp it is generally
preferable to use "[ \t]"
Matt> where <tab> is the tab key and not the text "<tab>".
This
Matt> should catch all whitespace other than newlines (ignoring
Matt> things like carriage-return, vertical-tab and form-feed,
Matt> which aren't likely to be in your code).
--
Adrian Aichner
mailto:adrianï¼ xemacs.org
http://www.xemacs.org/