The big news for version 0.5 is the introduction of the Jieubsida series of
fonts, covering Korean hangul. I decided to build these even though I don't
actually know the language, because the writing system seemed simple and I
thought it would be easy to add, providing a lot of bang for the buck in
terms of increasing the potential user base for Tsukuimashou. I was wrong
about it being easy to implement hangul; but doing the work had several
important side benefits. In particular, because of creating some
drastically larger fonts than this project has generated before, the Korean
fonts gave me the opportunity to find out about some problems in third-party
software that would if unfixed eventually cause big trouble for the Japanese
The Jieubsida fonts should be of usable quality (comparable to the Latin
range) already. All the 11172 Unicode precomposed syllables are included,
as well as all the Unicode "hangul compatibility jamo" and "hangul jamo"
code points - but only the main sequence, not the "extended A and B"
ranges. (The extensions may arrive in a later version.) There are extensive
OpenType features that should, if used correctly, allow all the
Unicode-defined combining behaviour as well as all of Microsoft's.
One of the side effects of implementing the OpenType features for the Korean
fonts was that I discovered a pretty serious bug in XeTeX's implementation
of glyph substitution; and, unfortunately, earlier versions of the
Tsukurimashou vertical fractions feature were actually written to depend on
the bug. In this version those tables have been rewritten in such a way
that they should work both in XeTeX and in any correct implementations of
the standard. The bug has been reported to the appropriate authorities.
There are also a number of behind-the-scenes changes to the infrastructure.
The build system has been rearranged to avoid doing in-place modification of
files wherever possible, because that was making it hard for make to
properly track changes. The kerning recipes have been split into smaller
chunks, the better to debug FontForge bugs that turn up during kerning.
It's not included in the distribution because by nature it depends on
outside fonts, but I now have a "missing kanji" maintainer tool that makes
a chart of kanji used in the documentation and not yet included in the font,
so as to make planning easier. And the documentation itself has expanded