In article <[EMAIL PROTECTED]>, Dave Love <[EMAIL PROTECTED]> writes:
> Kenichi Handa <[EMAIL PROTECTED]> writes:
>> Unicode doesn't say which syntax a character should have in
>> Emacs.
> I know it doesn't in detail -- the Emacs syntax codes don't actually
> map directly to the Unicode properties. Categories can do that if
> appropriate.
>> Even if we give guillemets "generic string
>> delimiter" syntax, it doesn't mean emacs thinks it's not a
>> punctuation character. What emacs thinks is just that its
>> SYNTAX is not "punctuation character", which doesn't
>> conflict wiht Unicode's definition about a character.
> I don't understand what you mean,
What I mean is that there's no confliction between these:
1) Emacs gives the syntax other than punctuation to character X.
2) Unicode categorizes X as a punctuation character.
> but if you give these things generic
> string syntax (a) it's contrary to what the Emacs doc says about how
> text modes should behave, (b) they have to be assumed to be balanced
> and (c) they have to be assumed not to nest like  â â â âÂâ Â.
That is a meaningful point to be discussed. But, as I don't
use those characters, I don't know what is the most
convenient behaviour of Emacs for users.
>> So, it seems that even Unicode thinks that they are supposed
>> to be used as a pair in some sense.
> Yes, typically, but they are not always used in pairs and as far as I
> remember, only the low-9 Latin quotation marks are used consistently.
Even '(' and ')' are not always used in pairs as what I
wrote above, but don't you think it is convenient (in
general) that they have open/close paren syntaxes as now?
So, what is the more convenient behaviour for users (in
general) as to those guillemets; giving them punctuation
syntax, generic string syntax, or something else?
---
Ken'ichi HANDA
[EMAIL PROTECTED]
_______________________________________________
Emacs-pretest-bug mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/emacs-pretest-bug
