Jens Lautenbacher <jtl(a)schlund.de> writes:
"Stephen J. Turnbull" <turnbull(a)sk.tsukuba.ac.jp>
writes:
> >>>>> "Hrvoje" == Hrvoje Niksic <hniksic(a)iskon.hr>
writes:
>
> >> Well, is the only reason for motion events while over the
> >> toolbar for tooltips? Those are handled natively by the GTK
> >> toolbar.
>
> Hrvoje> Can you embed images in such tooltips? Do they do the
> Hrvoje> right thing with Mule-coded text (hi Stephen)? Can you
>
> Probably, but who knows? GTK I18N docs are nonexistent, the
> ChangeLogs and News files are discouraging, containing comments about
> "the internal representation is in flux". I don't have time to
follow
> another ML or read the source, and nobody else on XEmacs-beta seems to
> know.
There is an effort to implement proper i18n (complete with right/left
handling of text input, even in mixed mode, even in a singel dialog
entry box, and mirroring of dialog layouts based on reading direction)
in gtk via the pango project.
You may want to look at
www.pango.org
I'm not sure how well Pango will play with XEmacs. At least not in the
right-to-left text area. It looks like pango rendering operates on a
PangoLayout structure, which according to the docs 'represents an entire
paragraph of text'. That's how it handles proper wrapping of text in r2l
languages.
If XEmacs handled different direction text itself, pango could probably be
used to render the actual glyphs pretty easily though. You would basically
need to keep a PangoContext in the device record, gotten via
pango_x_get_context(), set up some routines to get the appropriate GCs with
pango_x_context_set_funcs(), and draw glyphs with pango_x_render() instead
of gdk_draw_text().
Hmmm... :) I think it is a little premature to start trying to use the
pango stuff though. Could be cool to experiment with once the unicode guys
are ready to test it out though. Might be easier than trying to munge up
a huge list of fonts for dealing with it under standard X.
-Bill P.