Community
Participate
Working Groups
Text compare for .project xml file display bogus for DBCS name. The compare function will not manage the file encoding to display. STEPS 1) create a project with DBCS name 2) right crick and select Share Project, then select proper cvs 3) Workspace file: .profile will be displayed 4) The DBCS name referred by the name tag displays bogus. UTF-8 encoding should be considered properly to display the DBCS.
Eclipse editors in general don't interpret the encoding tag of xml files. As a consequence double byte strings are not shown correctly. Moving to UI.
As part of the encoding work, we are looking at how to properly handle UTF encoded files, in addition to having a global encoding preference. XML files are another case where the encoding is indicated by the file itself. All components in Eclipse should process text files similarly, not just the editors.
Although flat file encoding depends on platform locale, the set of file should be portable in eclipse family. There is a encoding option in JBuilder when we create a project. I assume that it might be used runtime locale encoding for the default and we can set other encoding for the project if we need. For example we can use set of files on Windows that were created on Linux at the project by setting encoding properly. CVS does not have encoding conversion at any operations. We need an absorbed function in eclipse.
The current plan is to have a single global preference for the encoding, rather than a per-project preference.
prioritised must fix for 2.0 release
Please clarify whether the issue here is with the file name or the file contents. The initial description refers to the name. The rest of the comments refer to the contents. In general, editors which know nothing about XML cannot parse it to determine the encoding. If there is an issue with the file name, that is more critical and may involve Core behaviour.
It's about contents. The initial description is about DBCS project name in a <name> tag in .profile, it's about contents of xml file encoded by UTF-8.
Will the recent encoding changes address this problem?
The problem is addressed only partially. Compare will respect the workbench encoding preference. Thus, files written with this encoding will correctly be presented. However, this does not solve the issue in general. Still open: - dealing with xml files that are self describing in respect to encoding - support for file specific encodings when saving - encoding display option for the compare view The general issue can not be solved for 2.0 as it requires Core to open up for the encoding issue. Reassign Compare to ensure that all files are read honoring the workbench encoding preference.
Fixed for 20020606 The Compare plugin now respects the workbench encoding preference.
Fix reviewed by Dirk Baeumer
*** Bug 12855 has been marked as a duplicate of this bug. ***
I've just download the integration build 20020606, but it didn't fix at all. Could you tell me the fixed build to verify? or anything I have to configure to display correctly?
Did you set the 'Text File Encoding' in Preferences/Workbench/Editors to UTF-8? As mentioned before the platform does not yet look into xml-files for determining the encoding.
Once changing the editor encoding to UTF-8, all of files even java program file will be treated as UTF-8 file. This will lose interoperability. In general, java program file encoding should be platform encoding. I don't accept the current fix.
*** This bug has been marked as a duplicate of 5399 ***