On Mar 18, 2008, at 03:50, Ben Schonle wrote:
> Lars Grunewaldt wrote:
>
>> Am 12.03.2008 um 17:04 schrieb Ben Schonle:
>>
>>> when importing a folder with mutiple subfolders and files I get
>>> the error:
>>>
>>> svn: Valid UTF-8 data
>>> (hex: 4b 61 69 72 6f 73 20 70)
>>> followed by invalid UTF-8 sequence
>>> (hex: f5 68 69 76)
>>>
>>> As I understand it there are some characters in the file names
>>> that are not UTF8. I was thinking not to rename the respective
>>> files / folders, but would need to know their names, locations
>>> first.
>>>
>>> How do you suggest to proceed?
>>
>> this happens mostly (for me), when entering a commit message that
>> contains German Umlaute or other non-ISO8859-1-Characters on a
>> unix terminal. Maybe that's the case for you, too?
>>
>> Otherwise, you could hex-decode the string (treat the numbers as 8-
>> Bit-ANSI-Characters), that should be a part of your filename.
>
> for me the problem is that some folders / file names contain German
> Umlaute.
>
> according to Lars I should hex-decode the string. Thus the questions:
>
> * how do I find out which folders / files I need to hexdecode?
> * how do I hexdecode them once found?
Take the hex in the error message and turn it into characters.
>>> svn: Valid UTF-8 data
>>> (hex: 4b 61 69 72 6f 73 20 70)
That's all ASCII data (all bytes are less than hex 80) so in any
character encoding that's "Kairos p"
>>> followed by invalid UTF-8 sequence
>>> (hex: f5 68 69 76)
If we assume the character encoding of these bytes is ISO-8859-1,
then this is "õhiv".
One solution is to rename the files to have no non-ASCII characters.
What you probably want to do, though, is set the LANG environment
variable correctly so that svn knows what character encoding to use
to read the file names.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: users-help_at_subversion.tigris.org
Received on 2008-03-18 21:45:49 CET