きんゆう in romaji should be typed "kinnyuu"

[mvysny]

きんゆう in romaji should be typed "kinnyuu" as every other IMEs do, instead of having to type "kin'yuu".

---

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

[mvysny]

Hmm, implementing this feature would no longer find words like 侵入/しんにゅう since searching for shinnyuu would search for しんゆう instead. What would you suggest in this situation?

[K-410]

Thanks for getting back!

侵入 would require three consecutive "n"s. The input would be "shinnnyuu". It might come off as illogical at first, but IMEs usually interprets ん based off double keystrokes, meaning two "n"s equals ん and a third "n" would start on the next syllable (にゅ in this case).

[mvysny]

Currently Aedict also recognizes "x" as a modifier; so for example xn always produces n, so you can search for shixnyuu to search for しんゆう. There are also other possible combinations: xa will input small a, xya will input small ya, etc. Just a sidenote: I noticed a weird thing. On my phone, even if you do search for shinyuu (and the phone will say it will search for しんにゅう), it will still find しんゆう. The same for きんゆう. Would it be acceptible for you to use x in your searches to enforce ん? Since using kinnyuu->きんゆう creates this duality in which きんゆう is further converted to romaji kinyuu, which would then be converted back to きにゅう...

[K-410]

To be honest it's not much of a deal. Up until now I've been typing "kin'yuu" to get きんゆう, and just found out a while back that "kin.yuu" or even "kin,yuu" works. I just wanted to point out that aedict doesn't follow all the IME input standards in regards of romaji > kana in search.

I was looking for an example using a different app, but it seems that that one changes input to kana before search so it doesn't have to deal with romanization issues in the first place.

Anyway, thanks for taking the time.

[mvysny]

Sure, no problem :) I'll install scim (or whichever ime works on Linux) and I'll test for myself.