[digiKam-users] DigiKam and multiple computers -- best practices

[karsten_df]

Simon Cropper wrote
> It is usually things like curly brackets and em-dash characters that 
> make it into paths.
> If your European characters are coming through your don't had CP-1252, 
> probably using CP-1251 (see potential alternatives here 
> https://en.wikipedia.org/wiki/Windows_code_page).
> Here is a thread that can help you determine the code page your system 
> is using 
> https://serverfault.com/questions/80635/how-can-i-manually-determine-the-codepage-and-locale-of-the-current-os


Warning - somehow lengthy text below - should discussion be moved to
different place?

let's start with the WIN10 machine
- chcp on the WIN10 machine says codepage 850 (locale is German)

using a shared folder between WIN10 and a Linux Virtiual Machine (file
system is vboxsf):
- when I create a text file in WIN10 using Notepad, special characters
(µÖÄÜß{}~€) are shown correctly on the linux side
  (only) if I use CP-1252 or 1250 to display the text.
- in the opposite direction, these characters, when entered as utf-8 in
Linux, show correctly at the Windows side in Notepad.  
  A check with a more versatile editor reveals that the encoding actually
still is utf-8 in WIN10.
- all filenames using characters µÖÄÜß{}~€ are transferred correctly between
systems *(!)*

now looking at the target scenario, a linux server, running nfs and samba
servers for one linux & one windows clients resp.
- exactly the same behavior as above

=> my takeaways:
- in my (default) configuration, there are no obvious problem with
filenames, although WIN10 should use utf-16 or 'OEM',
  depending on the file system type, and Linux should use utf-8.
  Some transcoding of filenames seems to be done between the systems.
- on the application level, behavior on WIN side is app specific -
  some programs create text in CP-1252, while others use utf-8.
  Also at least some do interpret utf-8 correctly

==>> so the interesting question is:
Does digikam always (Linux / Windows) use utf-8 encoding for all text that
goes into picture metadata?
If so, any issues would be restricted to the filenames (unfortunately also
the most critical part ;-),
which, at least in my setup, seem to be handled correctly across WIN10 and
Linux for reasons not yet fully understood..



--
Sent from: http://digikam.1695700.n4.nabble.com/digikam-users-f1735189.html