bug#33688: UTF symbols in 25.2.2 under Ubuntu

[Eli Zaretskii]

> From: Sergei Meshveliani <mechvel@botik.ru>
> Cc: 33688@debbugs.gnu.org
> Date: Mon, 10 Dec 2018 23:43:54 +0300
> 
> > IOW, this doesn't seem to be an Emacs problem, but a problem with
> > terminal emulator and/or fonts used by the terminal emulator.
> > 
> > Thanks.
> > 
> 
> 
> I am sorry, I have forgotten of how do I call emacs!
> I called it via alias as
>                           emacs -nw
> 
> I Debian + emacs 23..  it works fine, 
> but not in Ubuntu 18.04 + emacs 25.2.
> 
> Now, if I call it simply by
>                             emacs 
> 
> then it is almost all right.
> Only
>  ∷   (\::)   is shown considerably larger than, for example, capital O,
>  and ∣  ( \| )  is separated with half-blank, so that  ∣x  
>  is drawn close to  ∣ x
> 
> (while  Debian + (emacs-23 -nw)   draws everything all right).
> 
> Ubuntu 18.04 + emacs 25.2  shows the following description:
> 
> ******************************************************************
> \::
> 
>  character: ∷ (displayed as ∷) (codepoint 8759, #o21067, #x2237)
>     preferred charset: unicode (Unicode (ISO10646))
> code point in charset: 0x2237
>                script: symbol
>                syntax: w        which means: word
>              category: .:Base, c:Chinese
>              to input: type "\::" with Agda input method
>           buffer code: #xE2 #x88 #xB7
>       file code: #xE2 #x88 #xB7 (encoded by coding system utf-8-unix)
>               display: by this font (glyph code)
>     xft:-GOOG-Noto Sans Mono CJK
>     SC-normal-normal-normal-*-17-*-*-*-*-0-iso10646-1 (#x34B)
> 
> Character code properties: customize what to show
>   name: PROPORTION
>   general-category: Sm (Symbol, Math)
>   decomposition: (8759) ('∷')
> 
> **********************************************************************
> \|
> 
> character: ∣ (displayed as ∣) (codepoint 8739, #o21043, #x2223)
>     preferred charset: unicode (Unicode (ISO10646))
> code point in charset: 0x2223
>                script: symbol
>                syntax: w        which means: word
>              category: .:Base, c:Chinese
>     to input: type "\|" or "\shortmid" or "\mid" with Agda input method
>     buffer code: #xE2 #x88 #xA3
>     file code: #xE2 #x88 #xA3 (encoded by coding system utf-8-unix)
>               display: by this font (glyph code)
>     xft:-GOOG-Noto Sans Mono CJK
>       SC-normal-normal-normal-*-17-*-*-*-*-0-iso10646-1 (#x33D)
> 
> Character code properties: customize what to show
>   name: DIVIDES
>   general-category: Sm (Symbol, Math)
>   decomposition: (8739) ('∣')
> ***********************************************************************

My only advice is to try using a different font for these characters,
not Noto Sans Mono CJK.

I don't think there's an Emacs bug here.