lilypond-devel
[Top][All Lists]
Advanced

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

Re: How to report `FontForge` problems while creating LilyPond fonts


From: Torsten Hämmerle
Subject: Re: How to report `FontForge` problems while creating LilyPond fonts
Date: Thu, 27 Feb 2020 02:40:18 -0700 (MST)

Werner LEMBERG wrote
> Yes, and avoiding overlaps everywhere would lead to unnecessarily
> complicated code.  Tools like FontForge should be used to fix this in
> a post-processing stage, exactly as we do.

I totally agree. I'm going to fix the cases giving an error.
"Don't fix it if it ain't broke".



Werner LEMBERG wrote
> Great.  However, as mentioned above, I would not be happy with a
> serious decrease of readability of the already quite complicated
> METAFONT code...

You'll be surprised that there are no serious changes, it' only minimally
invasive surgery.
I'll open up a ticken and upload a patch as soon as it's finished (it
already works, but I want a proper solution).
You can review it then and Rietveld will show all the changes and if you're
not happy with them, we can junk them.  But I promise it won't get worse,
just a bit different.



Werner LEMBERG wrote
> Your ideas look nice, thanks for working on that!  Please note that
> for testing and inspecting outlines you should process the METAFONT
> sources with
> 
>   FONTFORGE=foo mf2pt1 ...
> 
> to really see the outline overlaps (if viewed with FontForge, say).
> The output of `gftodvi` doesn't help here.

Yes, I'm aware of that and I've always tested my changes doing a full make
in my development box, observing all the messages and looking at the final
result (PFB/OTF) using FontForge.

But gftodvi is a great help to see what mf actually does (as you can see,
I've even temporarily switched it to draw outlines only. That way, it can
easily be seen what overlaps, loops and curls FontForge has to cope with). 
In the end, gftodvi shows what is handed over to FontForge - and if that is
quirky in some way, FontForge will have a much harder job to do.
I'm trying to correct the cause rather than to alleviate the symptoms.

*By the way:* how about TABs? Obviously, gitk, Rietveld et al. seem to use 8
spaces per tab.
I think I'd better continue using tabs for the time being so that my changes
clearly stick out.


Cheers,
Torsten



--
Sent from: http://lilypond.1069038.n5.nabble.com/Dev-f88644.html



reply via email to

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