From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020830 Description of problem: Hardware Environment: PIII 600MHz Mem 256MB Software Environment: OpSystem: RedHat7.2 and all Japanese supported distributions Locale: ja_JP.eucJP kinput2: version 3.0 and version 3.1 beta4 Preedit characters of kinput2 (Japanese Input Method) don't follow its target window, when the main window was resized. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1.Set kinput2 to your InputMethod on Japanese locale. 1. Compile and run this sample code below on Japanese locale. Then, one window with one rectangle is shown. 2. Move mouse pointer in the rectangle. 3. Press Shift and Space key to activate kinput2 4. Input some characters and keep preedit status. Don't commit them. 5. Resize the main window Actual Results: Actual Results: The preedit characters stay the original location. Expected Results: Expected Results: The preedit characters should follow the target window. Additional info: Additional Information: It looks kinput2 doesn't work, when the target window (not the main window) was moved without its resizing. The problem occurs on kinput2 only. It doesn't on other input methods, ATOK X/htt or Wnn 7/xwnmo.
Created attachment 90222 [details] Program to reproduce the problem
Created attachment 90223 [details] program to reproduce problem.
Can someone update this issue, please? The Issue Tracker issue that originated this is about to become 1 year old, without an update.
To clarify whether this is a bug or not, I need to read the XIM spec. the most codes I know well is GTK+. GTK+ 1.2 did supports the over-the-spot like attached code. Although I'm not sure if it's kinput2 bugs, the GTK+ 1.2 applications did repositionning the preedit area, when it received the realize and size_allocate events. in other words, the applications has a way to resolve this. I'm not sure if this is MUST thing as the XIM spec at this point.
Well, I got a advice. the specification of XIM Protocol didn't mention that who should be handling the focus window, how should notify the preedit location, where should be done. we have some solution. but as a lot of apps like gtk1.2 does, most easy and no effects for other apps is the XIM client relocates the preedit line. I'm not sure how other solution affects all of XIM client. so I can't fix this soon, and this problem can be fixed in the client side so that the XIM spec is ambiguous. Set a low priority and a low severity.
Since it's deprecated in current release, and we'd recommend to use IIIMF instead of, it's hard to have a time to fix this ATM.