Eclipse Community Forums: Newcomers » Subclipse UTF-8 Question

Help

Home

Home » Newcomers » Newcomers » Subclipse UTF-8 Question

Show: Today's Messages :: Show Polls :: Message Navigator

Subclipse UTF-8 Question [message #270492]

Thu, 23 April 2009 20:59

Mark Dexter

Messages: 324
Registered: July 2009

Senior Member

Hi. I'm hoping someone can direct me to the correct forum to ask this.

I'm using Subversion Client Adapter 1.6.0.2 with Eclipse 3.4.2 on Windows
Vista. I have a problem creating and applying patches for text files with
UTF-8 characters. The Eclipse text editor seems to handle UTF-8 correctly,
but when I create a patch it seems to lose the UTF-8 encoding and put in
weird characters in the patch file.

My default encoding (Preferences/Workspace/Encoding) for Eclipse is UTF-8.
I have tried playing with the Preferences/General/Content Types and
haven't had any luck.

I tried an older version of Subversion (1.5) and it seems to create the
patch ok but, when that patch is applied, the character encoding of the
patched file gets messed up.

If anyone can point me to the right place to get help on this, I would be
most grateful. Thanks!

Report message to a moderator

Re: Subclipse UTF-8 Question [message #499747 is a reply to message #270492]

Mon, 23 November 2009 22:27

Christian Cooper

Messages: 1
Registered: November 2009

Junior Member

Hi Mark, I don't suppose you managed to get a solution to this?

If not, anyone...?

I have the same problem, the default character encoding in Eclipse is UTF-8, I'm working on a Java project, and the files (but not the file names) contain extended characters in French and German words, the usual things are present, accents, circumflexes, cedillas and umlauts.

When I commit the code to the repository, all appears well, but when anyone else does an update, they get '?' instead of accented chars, and if anyone else commits changes and I do an update, I get the same '?' chars in the code.

I'm running on Windows 7, Eclipse 3.5.1, Subclipse 3.0.0, Subversion Client Adapter 1.6.4.1 and JavaHL 1.6.5 (r38866).

As I understand it Subversion should be completely agnostic about the content of the files, so either the commit of the file is causing the content to be damaged prior to transmission, or the file is fine but Eclipse is trying to display it incorrectly when it comes back.

I am inclined to think that the file is being damaged prior to transmission, simply because other tools that support UTF-8 (E.g.: TextPad 5), concur with Eclipse, that the file contains odd characters post update.

Any help would be gratefully received.

Thanks in advance,

Chris.

Report message to a moderator

Re: Subclipse UTF-8 Question [message #520267 is a reply to message #499747]

Thu, 11 March 2010 18:43

Mark Dexter

Messages: 324
Registered: July 2009

Senior Member

The only solution I found was to use TortoiseSVN to apply patches with UTF-8 encoding. That works perfectly.

The same problem has also been encountered by a colleague on Mac OSX.

I don't know what causes it. If it was a windows issue, you would think TortoiseSVN would have the same problem. Weird.

This just in. I got it to work in Eclipse by copying the patch file to the clipboard and using the Clipboard option in the Patch Input Specification dialog. When you preview the patch, all of the UTF-8 characters show as "???", but the patch applies correctly and you can compare in Eclipse after the patch and see the changes correctly. Hope this helps someone.

[Updated on: Thu, 11 March 2010 18:51]

Report message to a moderator

Previous Topic:	eclipse broke (can't open editor)
Next Topic:	Force.com project deployment updated profiles!

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

]

Current Time: Fri Apr 26 21:09:47 GMT 2024

.:: Contact :: Home ::.

Breadcrumbs

Sign up to our Newsletter