Re: Charis SIL font - Z-a issue

2012-09-11 Thread Patrick McCarty
On Tue, Sep 11, 2012 at 11:30 AM, Jakub Pavlík wrote: > 2012/9/11 Werner LEMBERG >> >> Please report this (with examples) to the ghostscript people so that >> they can analyze and fix it. > > Patrick, will you report it to the GS devs? I don't understand PostScript > (and thus also what you found

Re: Charis SIL font - Z-a issue

2012-09-11 Thread Jakub Pavlík
2012/9/11 Werner LEMBERG > Please report this (with examples) to the ghostscript people so that > they can analyze and fix it. > > > Werner > Patrick, will you report it to the GS devs? I don't understand PostScript (and thus also what you found about the issue) enough... Thanks, Jakub

Re: Charis SIL font - Z-a issue

2012-09-10 Thread Werner LEMBERG
> With LilyPond 2.17.0, I ran a few short tests, analyzing the PS > output and corresponding PDF output after every run. The strange > thing is that the PS output always looks correct! For example, here > is the glyph string from a failure case, which is output as "azaz" > in the PDF: > > 5.6906

Re: Charis SIL font - Z-a issue

2012-09-10 Thread Patrick McCarty
On Mon, Sep 10, 2012 at 6:33 AM, Jakub Pavlík wrote: > Hi, > > I have a very funny issue using the Charis SIL font: > under some special circumstances instead of the "Z" "a" is set in lyrics as > well as in markups. Interesting issue! > I tried to isolate the issue, but haven't been really succe

Re: Charis SIL font - Z-a issue

2012-09-10 Thread Janek Warchoł
On Mon, Sep 10, 2012 at 3:33 PM, Jakub Pavlík wrote: > Hi, > > I have a very funny issue using the Charis SIL font: > under some special circumstances instead of the "Z" "a" is set in lyrics as > well as in markups. Do you encode all your source files in UTF-8? Janek ___

Charis SIL font - Z-a issue

2012-09-10 Thread Jakub Pavlík
Hi, I have a very funny issue using the Charis SIL font: under some special circumstances instead of the "Z" "a" is set in lyrics as well as in markups. I tried to isolate the issue, but haven't been really successfull. What I can see about the code below is that 1. the issue only occurs with th