[Devel] Re: FIXED: Problems with oblique fonts

[Tom Kacvinsky]

I have a new fix ready for commiting, but I can't access the cvs server.
Anyway, I now understand in more depth as to what is going on, and have a fairly
good idea why the patch works (I am still hung up on applying the inverse of the
font matrix to the translation vector).

Tom

On Sat, 16 Dec 2000, Tom Kacvinsky wrote:

> Attached is patch that fixes the problems with oblique fonts and the 
> autohinter.
> As far as I can tell, the problem lies with the computation of hinting 
> globals.
> The reason why is that when the autohinting glyph loader is called for the 
> first
> time (with the first glyph in a string, as Keith pointed out), the global 
> hints
> are calculated, and that throws off the first glyph.  When the second glyph is
> loaded, the global hints are already in place (so they are not calculated), 
> and
> everything is peachy.
>
> My simplistic solution was to immediately call the hinting glyph loader after
> returning from the global hints calculation routine(s).  The only time this
> will happen is when the global hints are not already in calculated.
>
> I also fixed a "problem" where the glyph transform was applied before the call
> to ah_hinter_load.
>
> Anyway, I would appreciate it if David could go over this patch and make sure 
> it
> is OK.  I wills ay that it fixes the problems I was having with ftstring, and 
> it
> works with all of the Type 1 fonts I have...
>
> Question:
>
> David, why is the transformation matrix inverted before the transformation 
> takes
> place?
>
> Regards,
>
> Tom
>
>