Tim Armes wrote:
> I use TortoiseSVN on Windows to commit my changes. The message logs that I
> use often include accented characters such as the acute e. When I use the
> Command prompt to retrieve the log, everthing is as expected. There is,
> therefore some sort of consistancy of character encodings.
>
> Now, when I try to view the repository using ViewCVS, my accented characters
> don't come out correctly at all. Looking in to the problem a little further,
> I discover that the character code returned for the acute e is 0x82, which
> is in fact a "low quote" under the Windows character encoding.
>
> This is strange since the Windows encodings place this character at 0xE9.
> Indeed, had it been E9, ViewCVS would have displayed it correctly since the
> acute e is also at E9 under ISO 8859-1.
>
> Can anyone explain what's going on.
On checkin, Subversion converts from your local encoding to UTF-8. On
checkout, it does the reverse.
ViewCVS is displaying the file as UTF-8, rather than translating it to
your local encoding like the Subversion client does.
Unfortunately, your browser is not detecting that the file is UTF-8, and
is trying to display it using the wrong encoding. Plain ASCII
characters work anyway because both encodings use the same values to
represent them; accented characters do not.
You can work around this by telling your browser to use the UTF-8
encoding before viewing a Subversion repository with ViewCVS. If that's
not good enough, and you control the repository, you can probably coerce
either your web server or ViewCVS to set the content-type to "text/html;
charset=utf-8" instead of just "text/html"
--
David Ripton dripton@ripton.net
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Tue Feb 3 05:49:06 2004