Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 14524

Summary: [CVS CVSNT] DBCS: need '-ku' keyword for Unicode file
Product: [Eclipse Project] Platform Reporter: Masayuki Fuse <fuse>
Component: TeamAssignee: Platform-VCM-Inbox <platform-vcm-inbox>
Status: CLOSED WONTFIX QA Contact:
Severity: normal    
Priority: P5 Keywords: helpwanted
Version: 2.0   
Target Milestone: ---   
Hardware: PC   
OS: Windows All   
Whiteboard:

Description Masayuki Fuse CLA 2002-04-24 11:18:58 EDT 
Windows cvs server has Unicode file translation when the file maked with '-ku' 
flag.
I'd recommend that "Set keyword SubstitutionMode" dialog supports the '-ku' to 
set.
Comment 1 Kevin McGuire CLA 2002-04-24 15:58:52 EDT 
CVSNT specific
Comment 2 Michael Valenta CLA 2002-09-09 16:21:25 EDT 
Reopening
Comment 3 Brice Ruth CLA 2004-01-22 22:44:04 EST 
Is this truly CVSNT specific? Are files committed to Linux/UNIX CVS as unicode
files? The only options I see in Eclipse (2.1.x & 3.0Mx) are "ASCII". I've
noticed that any project files containing "native" non-Latin1 characters are
corrupted upon commit/check-out.

So, does Eclipse support unicode and/or utf-8 for CVS or no?
Comment 4 Michael Valenta CLA 2004-11-25 06:24:08 EST 
As previously stated, -ku is CVSNT specific and will not be handled. My 
understanding is that CVSNT has modified the way that keyword substituation 
modes are handled anyway. 

As to the question in the last comment, the cvshome.org version of CVS does 
not appear to have additional support for storing UTF in non-binary files. 
However, my impression is that UTF-8 is OK to store as ASCII (-kkv, -ko, etc) 
but UTF-16 must be stored as binary (-kb).
Comment 5 Masayuki Fuse CLA 2004-11-29 22:49:20 EST 
I'm closing this pr, since the 3.0's encoding support doesn't need to address 
the -ku CVSNT specific support. We can convert on Eclipse if needed.