In Chinese (C/J/K) input method thereās a āOn the Spotā composition mode, which means when inputting text and launched the input method, the composition area and the cursor were treated as a part of the text input frame. In the CODE (both 2.x and 3.0) when using this mode, we couldnāt see the composition area hence couldnāt know what we typed.
In the āOver the Spotā mode the composition area was an independent window hence it was okay to see what we typed and the result of the composition was sent to the text input frame successfully.