[Readme] Umlauts are not rendered correctly

German umlauts in the readme file are not rendered correctly even if the readme file is correctly saved as UTF-8 and the browser recognizes the page being UTF-8:

  • ä > ä
  • ö > ö
  • ü > ü
  • ß > ß

Tested with latest releases of Firefox, Chrome, and Edge.

Somewhere there has to be another codepage. You can see a simple readme file here.

If this is renderes correctly at your site, too, we have to take a look at the installation. Do you use Mercurial? Do you have configured the code page for mercurial in the global administration (Administration - Settings - Mercurial)? Does the server have another global code page?

I just installed the SCM-Manager and imported a svn repository.
My source code view have the same issue like :

"2" Solicita��o de Compras                                            |
15|                                            "3" Cota��o     

This source code is ISO 8859-1.

Is there a way to configure it ?

Sorry for the late reply but I somehow missed your post.

  • We don’t use Mercurial; the Mercurial encoding is set to UTF-8.

I will check the server code page and report back.

Do you know where to set it ? Is it in repository or in the server config ?

For SVN I can reproduce, that files with non-UTF-8 encoding are rendered incorrectly. I will take a look whether we can fix this somehow.

would you like to have a file with those characters ? I can upload it

Though I’m using a keyboard with international layout, I know how to type spanish or german characters, too :wink:

Jokes aside, the question is how to detect different encodings in SVN files. I found something about props one can set for files, but I’m not quite sure how this works and even less sure whether svnkit (the library we use for SVN) would handle this correctly or at least help us here. Maybe this will lead to a similar approach as for Mercurial with a explicit configuration, but we would need some time for this.

Hi @ffoliveira , we checked this once more and came to the conclusion, that a setting for the enconding in SVN repositories would be very complex in SCM-Manager, because there are many places where this would have to be handled. As far as we see there is no support for switching code pages in SVNKit (the SVN library we use), so we would have to handle every single stream manually. Sorry for this, we hope you can use SCM-Manager nonetheless.

1 Like

Thank you.
It isn’t really necessary. As we edit the source code outside SCM platform.
I was just to view the source with correct page code.
But the code itself isn’t use special characters. only strings and comments.

1 Like