bug#13399: 24.3.50; Word-wrap can't wrap at zero-width space U-200B

[Eli Zaretskii]

> Date: Fri, 11 Jan 2013 11:29:46 +0100
> From: martin rudalics <rudalics@gmx.at>
> CC: 13399@debbugs.gnu.org
> 
>  >>  > You mean, not wrapped at all.  Witness the continuation bitmaps in the
>  >>  > fringes, which shouldn't appear when a line is wrapped.
>  >>
>  >> I thought these bitmaps appear when a line is wrapped.
>  >
>  > Not by default.  Not unless you customize visual-line-fringe-indicators.
> 
> With emacs -Q I see curly arrows in the fringes regardless of whether I
> set `visual-line-fringe-indicators' or not.  What am I missing?

If this is still with u+200B?  You need to try with regular spaces.
Then the indicators should disappear on wrapped lines.

>  >> The doc-string of `word-wrap' says
>  >>
>  >>    When word-wrapping is on, continuation lines are wrapped at the space
>  >>    or tab character nearest to the right window edge
>  >>
>  >> Since U-200B is a space character the line should wrap at it.
>  >
>  > No, it means literally "the space character", U+0020.
> 
> So `word-wrap' is ASCII-only?

Yes.

> The doc-string should say so.

Well, I personally find it hard to imagine that "the space character"
could be interpreted as something other than U+0020.  But I see what
you mean.

>  >> and Emacs apparently does handle it specially since it reserves a few
>  >> pixels when drawing it.
>  >
>  > See glyphless-char-display and glyphless-char-display-control for why.
> 
> IIUC it has a `thin-space' display method entry and I could set this to
> `zero-width' (the doc-string of `glyphless-char-display' is ambiguous
> about that)?

Yes.

>  Does this also mean that I can separate text properties of
> adjacent words by inserting a zero-width space between them?

Yes, I think so (if I understand correctly what you mean).

>  >>  > If we want to add more characters to the set, we should probably
>  >>  > arrange a special char-table for this, and have it exposed to Lisp, so
>  >>  > it could be customized.  Patches are welcome.
>  >>
>  >> IIUC all breakable spaces are between U-2000 and U-200B so maybe a
>  >> character table is not needed.
>  >
>  > Who said we want only break at breakable space characters?  Who said
>  > Unicode will never add more such characters in another block?  And
>  > what about low-ASCII characters, which are already in a different
>  > block?
> 
> But implementing a character table and working with it is harder.

I don't think it's harder, it's actually very simple.  You have a
simple API for setting values in the table and a simple API for
accessing a property of a character.

>  > In any case, even if you are right, a char-table is a way to store
>  > character properties efficiently.  In particular, it will waste very
>  > little storage to mark a contiguous range of characters with the same
>  > property.  The advantage of using a char-table is that it will
>  > dynamically expand as needed if more characters are added to the set.
> 
> Is it useful to make a _separate_ table for line-break properties?

Why not?  What existing table would you reuse for that?

>  > What did _you_ want exposed to Lisp?
> 
> Two functions: One to get the width of some arbitrary buffer text in
> pixels and one to get the full height of a buffer text line in pixels.
> The former would be used for doing word-wrapping variants in Lisp, the
> latter for fitting windows to their buffers.

The latter already exists as window-line-height, doesn't it?

Anyway, how would you word-wrap in Lisp, except by adding display
strings with newlines (which AFAIR features like longlines
etc. already do)?