--- Marcus Lundblad <ml update.uu.se> wrote:
>
[snip snip]
>
> > I can mail you sample test cases and fonts off-list if you'd like.
> > Its rather simple really - you'll need,
> 
> Yes, that would be nice.

Marcus, I mailed ya a tarball a few days ago of all the files you'd
require off-list - hope you got it.

> Yes, the first thing I did (which is in current CVS), is to combine for
> those characters were a composed glyph (equivalent to a base character and
> one or more combining characters) exists in Unicode.

Yeah saw that.
 
> The next part (which I have more or less working, need some testing), is
> to pick out those remaining composing characters (like U+064E) into a list
> of (position,value) and also reoreder those positions after BIDI is
> applied.

Sounds good.
 
> The next thing to do (the most complicated, I'm afraid), is to draw
> those composing characters on top of the "normal" characters.
> The hard thing, as it seems to me, is to calculate the position of
> each character in the ordinary string. One way is to draw each
> character separately and take care of all composing characters for
> a given character. The other might be to first draw the "base" string,
> then use the "TextLength" routine to "backtrack" the positions (in pixels)
> of each individual character.

Yeah, this is where I get lost.  I haven't looked at how other
applications do it as this might be a source of enlightenment (you
might consider doing that).  Unless I'm misunderstanding you though,
I don't see why you'd care to know the TextLength.  The way I view
this (and I could off mark here), one could simply right-align the
glyphs and draw them on top of each other (order should not matter).
The glyphs should be designed in such a manner so that superimposing
these composing glyphs remains within the realm of self-alignment (any
overlap ought to be considered a fault of the font file and not of the
superimposing process).

Regards,

 - Nadim


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com
--
Visit the official FVWM web page at <URL:http://www.fvwm.org/>.
To unsubscribe from the list, send "unsubscribe fvwm-workers" in the
body of a message to [EMAIL PROTECTED]
To report problems, send mail to [EMAIL PROTECTED]

Reply via email to