Re: Entering Unicode characters

[Eli Zaretskii]

> From: Helmut Eller <address@hidden>
> Date: Sat, 06 Feb 2016 12:56:14 +0100
> 
> On Tue, Jan 26 2016, Richard Stallman wrote:
> 
> > Would someone please implement this?
> 
> Here is my shot at this.  Load the attached file and start the browser
> with "M-x ubb-browse". That displays the characters of the "Basic Latin"
> block.  From there it's probably the easiest to use the menu to explore
> the available commands.
> 
> I defined a sets of characters for some languages by looking at
> Wikipedia.  This would obviously need more work by people who are
> familiar with those languages.
> 
> The file is also available at: https://github.com/ellerh/ubb

Thanks, I think this is very good.

One aspect where some improvement is probably needed is when
displaying large ranges of characters.  For example, select
"Scripts->symbol" from the "Character-Browser" menu -- this displays
all the punctuation characters defined by Unicode.  You get 160 lines
that display something like 4K characters, which takes a long time to
display (because Emacs looks for a suitable font), and somewhat hard
to view.  Maybe we should display with those +/- widgets that allow
showing and hiding portions of the characters, so that each 256
characters could be shown or hidden separately?

Also, I think unassigned codepoints should not be shown, and the same
for Surrogates and Private-Area codepoints.  They will never yield
anything but those rectangles with a codepoint in hex inside them.

Finally, I think a command to insert the character at point into
another buffer would be a good addition.

Thanks again for working on this.