As of Wakan 1.87, character data is taken from KANJIDIC/Unihan. There are different
language versions at least of KANJIDIC, and potentially of Unihan too.
Currently to replace data with data from another KANJIDIC, full rebuild of character
database needs to be done. All parts need to exist, which makes it difficult to update,
say, "just Unihan" (which KANJIDIC do you take when rebuilding?).
This needs to be reworked so that:
KANJIDIC and Unihan versions are downloaded and converted completely separately.
You can have as many versions of them downloaded as you like.
There has to be a settings page where you can select active KANJIDIC and active
Unihan from all available ones. The switch needs to happen near instantly (no long
rebuilding).
If possible, it'd be nice to have several KANJIDIC active at the same time. Either:
a. Try one by one, until we find a match for this kanji
b. Same, until a match for this PROPERTY (some KANJIDICs have less info than others)
c. Same, but merge results for some properties (e.g. show Russian kanjidic description
english)
One of this, depending on how comlicated it'll make stuff. (This is optional - just
fast kanjidic switching is good enough for daily life).
Original report by me.
As of Wakan 1.87, character data is taken from KANJIDIC/Unihan. There are different language versions at least of KANJIDIC, and potentially of Unihan too.
Currently to replace data with data from another KANJIDIC, full rebuild of character database needs to be done. All parts need to exist, which makes it difficult to update, say, "just Unihan" (which KANJIDIC do you take when rebuilding?).
This needs to be reworked so that: