Fix the working of Ctrl-6

[ichizok]

fixes #785

• Restrict the condition not calling interpretKeyEvents: ; only Ctrl-O and Ctrl-U without no marked text

[ychin]

Wow… these are some really unfortunately hairy special cases. A couple comments for @ichizok:

• Has this bug been reported to Apple before? I played around with the Japanese IME with "Windows-like shortcuts" turned on, and seemed to me Ctrl-U/I/O would toggle hiragana/katakana/romanji when marked texts are visible, but when no marked texts are up I couldn't figure out what Ctrl-U/O are even supposed to do. Is this simply a bug where the IME is eating keys for no good reason?

• The condition for checking whether we call interpretKeyEvents is getting quite long. Can you move the "is this Ctrl-U or Ctrl-O" logic to a separate variable to make it clearer?

• I'm wondering if we should make sure to only enable this special case for the Japanese IME, since I don't know if there are other IMEs there could for some reason use Ctrl-U and broken by MacVim's special handling. It would also make it more explicit that this is a special hack for one case instead of a generic thing. I think something like this would work:

  BOOL isJapaneseIME = [[[NSTextInputContext currentInputContext] selectedKeyboardInputSource] hasPrefix:@"com.apple.inputmethod.Kotoeri"];
  

[ychin]

I made those changes and it's merged now.

[ichizok]

@ychin Sorry for the late reply.. and Thank you for improving 😄

[ychin]

Sure np. I filed a bug with Apple, so hopefully they can address this (not going to share the radar number since it's internal to Apple anyway).