"Japanese -> English" mode fails to match exact kanji

[himselfv]

Original report by me.

Originally reported on Google Code with ID 280

Reported in Issue 278, though well known before that:

In Dictionary, while "Any matches" mode successfully deflexes and matches romaji, kanji/kana
and english (at least separately), "Japanese -> English" mode only matches romaji.

The history behind this is that originally Wakan only had romaji matching for Dictionary
with the exception of "clipboard matching" where kanji was matched.

As "Any matches" was added, it was made to match either romaji or kanji/kana because
why not.

But "Japanese -> English" mode was not updated at the time.

Reported by himselfv on 2015-04-06 16:29:31

[himselfv]

Probably fixed in revision #d9c40dcd6006. I've updated the "Japanese->English" mode
to also try kanji/kana matching, just like "Any matches" does it.

There was the question of whether a separate "Kanji/kana" and "Romaji" modes are needed,
but it seems to me that there are no use cases for that.
"Jp->En" and "En->Jp" modes may cross, e.g. when you enter something like "area" which
may mean either english word or valid kana あれあ. In this case you may want to choose
to filter the results by only one of those.
But "Kanji/kana" and "romaji" modes are well separated by the characters you enter.
If it's latin/western alphabet, then you will only get romaji matches, if it's kanji/kana
then only straight ones.

Reported by himselfv on 2015-04-06 16:35:39

[himselfv]

Original comment by Anonymous.

I can confirm that the revision has fixed both issues to full satisfaction. Searching
for Kanji and via clipboard works intuitively and without any cluttering.

Reported by supermarkus420 on 2015-08-09 09:32:38

• Status changed: Verified