| Summary: | DBCS4.2 DBCS chars are garbled in folder name and file name. | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [ECD] Orion | Reporter: | Kentaroh Noji <kennoji> | ||||
| Component: | Editor | Assignee: | Ken Walker <ken_walker> | ||||
| Status: | VERIFIED FIXED | QA Contact: | |||||
| Severity: | major | ||||||
| Priority: | P3 | CC: | camle, harendra, kennoji, ken_walker, kitlo, libingw, maedera | ||||
| Version: | 0.4 | ||||||
| Target Milestone: | 0.5 RC1 | ||||||
| Hardware: | PC | ||||||
| OS: | Windows 7 | ||||||
| Whiteboard: | |||||||
| Attachments: |
|
||||||
I was able to reproduce it in my English Windows 7 with Japanese IME input. Will continue to investigate this week talked to Ken. He will continue the investigation. The data is being passed correctly it seems in the xhrDelete() from the client side Orion code. Will now have to look at what is happening on the server. Moving this out to RC1 due to time constraints This requires changes on both server and client side. For the most part the summary is that XHRs within the client side API were not setting the content type as UTF-8 and the Orion server file Servlet was not looking at the posted payload for the filename. Instead it was using the SLUG and that is always in 8859_1 I believe. Will open bugs for other DBCS issues I am seeing but will release changes for this particular defect. Client side changes: http://git.eclipse.org/c/orion/org.eclipse.orion.client.git/diff/?id=41379ca462907100f710dcf2aa15b1bd0cc77b25 Server side changes: http://git.eclipse.org/c/orion/org.eclipse.orion.server.git/diff/?id=55af3a675c655cb6b441ab8af5e98f50995c336f Fixes should be in tonights build I verified that the orion server as of 2012-06-12 fixed this problem. The server can handle Unicode character successfully. However, I have a concern. The page source file still specifies "charset=ISO-8859-1" in the meta tag as follows. <meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"> <title>Navigator</title> If possible, could you change the charset from ISO-8859-1 to UTF-8 to avoid a potential problem, please. |
Created attachment 213944 [details] DBCS chars are garbled. OS: Windows 7 SP1 Professional Japanese Edition Steps to recreate problem: 1- Login orion.eclipse.org with user id and password provided. 2- Click "Navigator". 3- Click "New folder" and enter DBCS chars such as "漢字" in new folder input field. 4- Then, submit. The DBCS chars of folder name are garbled. The same symptom happens in file name. Additional information for the problem determination: When I browse the HTML source file of Navigator, I found that the charset was ISO-8859-1, and this is for only Latin 1 characters. For globalization, using UTF-8 for charset is recommended. UTF-8 charset covers all Unicode characters including Latin 1, Japanese, Korean, Chinese, Arabic, Hebrew, Greece, Devanagari etc. Here is extraction from the HTML source: <meta http-equiv="Content-Language" content="en-us"> <meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"> This should be: <meta name="copyright" content="Copyright (c) IBM Corporation and others 2010." > <meta http-equiv="Content-Language" content="en-us"> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> When Orion's UI is translated, the content="en-us" should be changed in to each translated language id.