French translation (release 1.2)

[olange@…]

[jira2trac import : issue created on January 27, 2005 7:09:38 PM CET ​http://issues.cocoondev.org/browse/DSY-109 ]

French translation of Daisy messages, with the new texts for release 1.2 according to the email from Bruno (​http://lists.cocoondev.org/pipermail/daisy/2005-January/000729.html)

[paul]

daisy-i18n-fr.patch

[paul]

[jira2trac import : comment created by olange on January 27, 2005 7:18:30 PM CET]

The french translated messages files in a Zip archive with the correct folder structure.

All files UTF-8 encoded and with UNIX-style end of lines.

I tried to make a patch with SVN, but all extended characters appeared scrambled! It seems that SVN does not support UTF-8? Anyway, the existing files may just be replaced.

I translated against revision 1698 of Daisy (made an update this evening before starting translating).

[paul]

[jira2trac import : comment created by olange on January 31, 2005 4:59:40 AM CET]

I received this morning a bunch of delayed messages sent to daisy@… these last weeks (my email provider had some trouble with its storage).

I read that Bruno asked on 17.01.2005 to receive the updates as a SVN DIFF rather than plain files:
​http://lists.cocoondev.org/pipermail/daisy/2005-January/000696.html

But does SVN DIFF work with UTF-8 files? How should I ask SVN DIFF to handle UTF-8 encoded files?

I am confused. I rather sent a ZIP file, because when I tried:

svn diff > daisy-i18n-fr.patch

and examined the daisy-i18n-fr.patch file (attached), I noticed the diff contains scrambled characters in place of the extended latin characters (with accents):

[...]
Index: repository/api/src/java/org/outerj/daisy/repository/messages_fr.properties
===================================================================
--- repository/api/src/java/org/outerj/daisy/repository/messages_fr.properties (revision 1698)
+++ repository/api/src/java/org/outerj/daisy/repository/messages_fr.properties (working copy)
@@ -1,2 +1,3 @@
document-not-found-exception=Il n'existe pas de document avec l'id {0}.
-authentication-failed=L'authentification a échoué (nom d'utilisateur inconnu ou mot de passe invalide).
\ No newline at end of file
+authentication-failed=L'authentification a échoué avec le nom d'utilisateur "{0}" (nom d'utilisateur inconnu, mot de passe invalide ou enregistrement non confirmé).
+duplicate-login=Le nom d'utilisateur "{0}" est déjà utilisé.
\ No newline at end of file
[...]

I am of course trying to open the file with an UTF-8-aware text editor and explicitely tryied to set the UTF-8 encoding when opening the file.

Am I misunderstanding SVN DIFF's output? It seems to me that SVN does not decode the UTF-8 files when it opens the file.

I would have expected the following output:

+duplicate-login=Le nom d'utilisateur "{0}" est déja utilisé.

Thomas and I had an off-list discussion about that, searched the SVN Book and svn list archive, send a message to the users@… list, but with no success till now.

[paul]

[jira2trac import : comment created by olange on January 31, 2005 5:00:22 AM CET]

(see previous comment)

[paul]

[jira2trac import : comment created by bruno on February 7, 2005 4:24:11 AM CET]

merci beaucoup!

I used your zip variant, I still need to look at the encoding problem with svn diff.

Note that the .property files should be ISO-8859-1 encoded rather then UTF-8, so I've recoded them.