This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Create UML Project - Plataform Independent Model. Create new Component Diagram. Add component. Double click it. Try to type "Acentuação": appear "Acentuaç~ao". Typing "Início" shows "In´icio". Typing in properties window works fine.
This may be same as similar seen before using asian input method tools to input multibyte characters in that language - sometimes nothing can be input, other times what is input is, as mentioned below, not correct. These other bugs on this is in other bug system still. There have been other customer mentions of similar issues. ken.frank@sun.com
The problem is even worse on Linux (Ubuntu 6.06 LTS) with UTF-8 and pt_BR locale. Typing an "á" results in "HR". Typing "Gráfico" results in "GrHRfico". This occur on Documentation window too. This is a very big problem bcs is very hard to tell developers to not type accents - typing accents is almost automatic on foreign languages, like portuguese brazilian.
One point more: just happen right now. Documentation window accepted the "á" key, but when switch to another object throws ArrayIndexOutOfBoundException. I'll attach the exception.
Created attachment 33654 [details] Relevant portion of NB log when error occur.
related issues when using multibyte are 72910, 78353, 78358
*** Issue 82572 has been marked as a duplicate of this issue. ***
the reason was that in the EditControl only key_pressed were processed, and key_typed - weren't. Thus "multi-keyed" character instead of being processed as one character it was processed as a mere sequence of keycodes used to type it in. Fixed in release55 branch. Didn't see the problem with doc window.
I verified using japanese characters but could filer try latest uml using the characters they originally used in the locale they were in when they found the issue ? But I noticed perhaps another issue - that when selecting the component name and backspacing so it would delete character by character - for ascii it deletes char by char, ie if original is abcde then deleting the e leaves abcd, and so on but with multibyte, deleting the first character deletes the other characters also. will leave this one open until research more to see if separate issue needs to be filed. ken.frank@sun.com
the problem mentioned in last comment does not seem to happen now - but will leave this in resolved to give original filer a chance to verify if desired. ken.frank@sun.com
brviking, could you please verify this issue using the uml module from NB 5.5 Beta Update Center or daily build from http://uml.netbeans.org/downloads.html to see if your problem has been fixed. Thanks!
Yes, it´s fixed. If I find additional problems, I'll create a new issue. Thanks you all!
Thanks for verifying the fix!