>     The problem is that text-properties are not really adapted for such uses.
>     After all, they apply to chars, not to text.

> They apply to text.  Text is made up of characters.

We're in violent agreement.

> Perhaps what you mean is that text properties belong to the text
> rather than to defined extents in the text.  This is intentional and
> necessary; it is the only way to get consistent behavior as text
> is killed, copied, and yanked.

Agreed again.

>     So by convention we consider that contiguous properties that are `eq'
>     make up a region/area/extent, which works 99%.

> Again, it is the only way to get consistent behavior as text is
> killed, copied, and yanked.  But you have to know the convention
> and arrange to follow it.

Agreed of course, once more.

>     Another alternative is to use overlays.  Except that overlays are not
>     duplicable and don't apply to strings.  That's where XEmacs's extents
>     make sense.

> They only appear to make sense, in that they prevent this
> easily-avoided pitfall.

No, what I'm saying is that in this particular case, XEmacs's duplicable
extents make sense, whereas when doing partial kills&yanks they indeed don't
make nearly as much sense as text properties.  You can't win both ways.


        Stefan


_______________________________________________
Emacs-pretest-bug mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/emacs-pretest-bug

Reply via email to