[Moin-user] Re: UnicodeDecodeError'utf8' codec can't decode bytes in position 0-1: invalid data

[Ralf Gross]

moin-user-admin at lists.sourceforge.net wrote on 03.02.2005 16:52:26:

> On Thu, 3 Feb 2005 10:47:14 +0100 (CET), Ralf Gross wrote:
>
> > I just updated our old RedHat 8.0 moinmoin 1.2.4 Server to Fedora Core 3
> > with moinmoin 1.3.3.
> > Now I have one page which gaves me an error.
> >
> > UnicodeDecodeError'utf8' codec can't decode bytes in position 0-1:
> > invalid data.
> > I attached the whole error page to this mail.
>
> This error is caused by an attachement filename. This should not happen
> but there is no migration script that migrates attachments. It is planned
> by ThomasWaldmann to write one.

Ok.

> > I ran all the migration scripts, is there anything I can do to
> > reconstruct the page?
>
> In order to repair it, goto the directory ../data/pages/MtcQip/attachments
> and rename every file that has non-ascii characters in it.
> If you want to preserve the original filename, just upload it with (I
> guess) umlauts in the file name again, you will see that the filename will
> be stored differently.

Actually, this was hat I did after some minutes of meditation. I removed
the link from the current version of the page and added the attachment
with the umlauts in the file name again. This worked.
Thanks, Ralf