GNU bug report logs - #12948
24.1; wrong angle brackets

Previous Next

Package: emacs;

Reported by: Mattias EngdegÄrd <mattiase <at> bredband.net>

Date: Tue, 20 Nov 2012 23:17:02 UTC

Severity: normal

Found in version 24.1

Done: Stefan Monnier <monnier <at> iro.umontreal.ca>

Bug is archived. No further changes may be made.

Full log


Message #11 received at 12948 <at> debbugs.gnu.org (full text, mbox):

From: Mattias EngdegÄrd <mattiase <at> bredband.net>
To: Stefan Monnier <monnier <at> iro.umontreal.ca>
Cc: 12948 <at> debbugs.gnu.org
Subject: Re: bug#12948: 24.1; wrong angle brackets
Date: Sat, 24 Nov 2012 13:39:49 +0100
[ Sorry about the duplicate; now the right recipients ]

23 nov 2012 kl. 22.15 skrev Stefan Monnier:

>> With the TeX input method, \langle and \rangle yield the deprecated
>> characters U+2329 and U+232A rather than the ones at U+27E8 and U 
>> +27E9
>> which are for mathematical use and Western languages.  The former
>> characters should not be used; they are troublesome because of their
>> canonical equivalence to CJK punctuation, and are frequently rendered
>> (correctly) as having the full width of a CJK character.
>
> Sounds OK.  BTW why not 3008 and 3009 instead?

U+3008 and U+3009 are the CJK brackets that U+2329 and U+232A are  
canonically equivalent to. They are full-width, so they don't go well  
typographically with Western languages or formal notation, where use  
of \rangle and \langle is intended. There could also be other  
differences in appearance (stroke width, height and so on).

>> These definitions are in leim/quail/latin-ltx.el. The same file also
>> produces the deprecated angle brackets from \leftparengtr and
>> \rightparengtr, but this makes even less sense, since those names  
>> should
>> properly be used for U+2993 and U+2994, unless I'm completely  
>> mistaken.
>
> This is less clear.  I can't find any of \leftparengtr or  
> \rightparengtr
> in any TeX package on my system.  Searching for them on the web, leads
> to more confusion.  Basically, I think that 2994 for \rightparengtr
> would be OK (tho it seems that there's a TeX package that provides it
> under the name \rparengtr), but for \leftparengtr
> http://www.w3.org/Math/characters/iso8879/isoamsc.html seems to  
> suggest
> that 29a0 is a better choice (after all, 2993 is a left paren with
> a "less-than" rather than "greater-than" in it).

Yes, you are probably right, but it's not crystal clear. U+2994 and U 
+29A0 seem accurate at first blush, but the Unicode reference chart (http://www.unicode.org/charts/PDF/U2980.pdf 
) does not render them in a way that suggests any relation between them.

U+2993 and U+2994 are more obviously rendered as a pair, but as you  
say the name \leftparengtr would not make sense for U+2993. By the  
way, the mirror of U+29A0 is U+2222, which in TeX is \sphericalangle.  
In addition, the Unicode metadata marks U+2994 as a closing paired  
delimiter and U+2993 as an opening one. U+29A0 and U+2222 are just  
marked as symbols.

If I were emperor of Emacs, I would delete them from latin-ltx.el and  
ask anyone who complains to give references, but if you prefer to keep  
them but changed to U+2994 and U+29A0 (which are the only characters  
that make sense after all), then please do so. The current definitions  
are clearly wrong.

> WDYT?

Sorry, I didn't understand that.





This bug report was last modified 12 years and 169 days ago.

Previous Next


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