freetype-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Devel] Re: address@hidden: Bug#149575: freetype1-tools: Bad computation


From: Jean-Christophe Dubacq
Subject: [Devel] Re: address@hidden: Bug#149575: freetype1-tools: Bad computation of heights in vpl files with ttf2tfm]
Date: Fri, 21 Jun 2002 10:33:19 +0200
User-agent: Mutt/1.3.28i

On Thu, Jun 20, 2002 at 07:08:40PM +0200, Werner LEMBERG wrote:
> [About ttf2tfm which incorrectly assumes that the x-height of TTFs is
> always 400 PostScript units.]
> Indeed, this is a bug which hasn't been noticed by any user before.
> The patch below (which I'll commit to the CVS soon) adds handling of
> the TTF OS/2 table version 2 to FreeType 1 together with a proper fix
> for ttf2tfm.  Please test.

I will test this this afternoon. However, the 'ugly' handling of
texheight (in vplaux.c) also has to be removed. The bounding boxes
should be correct for good operation of TeX (the present way is to
reduce the bounding box of some of the capital if they may combine with
accents that have to be constructed; however, when there are no accents,
which is the most frequent case, the height is not correct).


> > The best way to attack this is probably to compute x-height from the
> > size of 'x'.
> Exactly.  This will be used if the sxHeight field of the OS/2 table
> is zero.  The value `400' is now the last resort if there isn't an
> `x' glyph in the font.

I also would like a patch to allow for user-choice of xheight. I will
return to this this afternoon (european time). Sometimes, x has a
strange shape, and an average of u, v, w, x, y, z should be used (other
low chars have curves and are often a bit higher).

-- 
Jean-Christophe Dubacq



reply via email to

[Prev in Thread] Current Thread [Next in Thread]