[Freefont-bugs] Re: Cyrillic in Serif

[Steve White]

Hi again,

I have made a bunch of commits to CVS concerning Cyrillic.

I changed my mind about what to do.  For the time being, I just raised
the few letters in the Russian alphabet so they are consisitent among
themselves.  They are sitll inconsistent with Latin and with the other
faces, and a close look shows a variety of weights.  It is clear from
the record that the glyphs came from several different fonts.

This doesn't occur in Sans or Mono.  My thinking is, the Serif set
should be replaced all in one go.

But there were lots of problems, including

* horizontal spacing in italic styles was total chaos.  Did my best to
smooth it out
* especially in italic, but throughout the faces, there were incorrect
glyphs, mostly consisting of a glyph from the font that resembled the
correct glyph.  I deleted these--several dozen altogether.
* effort to unify look and spacing of accent marks
* added an italic yat (bug fix)

If anybody is interested, I can whip up a TrueType version for
testing.  I would really appreciate any comments!

Cheers!

On Thu, Mar 27, 2008 at 8:09 PM, Steve White
<address@hidden> wrote:
> Hi,
>
>  A bug was filed recently a about the Cyrillic letter i being too low.
>     https://savannah.gnu.org/bugs/index.php?22289
>  It was indeed, strangely a bit lower than other flat-topped Cyrillic letters.
>
>  I assumed this was just a glitch, maybe a slip of the mouse; it is
>  fixed in the latest release.
>
>  However, I revisited Cyrillic recently, and in Serif anyway, found
>  that this inconsistency happens in many other letters, as though the
>  range were composed from more than one font.  Indeed this was the
>  case.  (A quick check of Sans and Mono didn't turn up the same sort of
>  problem.)
>
>  Also, most of the letters in the Russian alphabet are substantially
>  taller and heavier than their Latin counterparts, by a good 6%.  Just
>  enough to be visible at normal text sizes.  That's poor.
>
>  Before I launch into a repair by hand, I thought it might be best to
>  ask the list for advice.
>
>  Is there a preferable way to fix this problem?
>