Sorry for not answering but i was a little bit busy.
I finished the work on the Version 0.0.5 of the plugin (yes i started to give version numbers... *G*)
1) I added a settings - page. You can now choose your videoquality. If you wanted quality isn't available, the nearest lower will be chosen. If there is now lower quality, the next higher will be taken.
2) After suffering extreme "performance"-lags with my regular-expression while parsing the ZDF-pages, some optimizations going in this point.
3) After still suffering long loading time on the ZDF-Mainpage i try to introduce multi-threading. Cause "insight" XBMC multi-threading is very painfully an doesn't offer a big advantage i write a background script. This script crawls the whole pages and transform all needed information into single (small) xml-files. In this case multithreading cut down "load-times" from over 40 minutes down to 4 minutes.
4) fixed some bugs with str and unicode.
you will find the plugin here: http://blog.raptor2101.de/wp-content/upl...iathek.zip
to use the background thread, you have to do some manual work
after installing the plugin, choose the settings site. Set the AccessMode to "local Cache".
create a directory named "archives" insight "<XBMC-userfolder>userdata/addon_data/plugin.video.Mediathek/"
create a cron-job (linux) or task (windows) and run the fetchOfflineData.py - Script located in the addon-directory. The script takes two parameters. First the full path to the archive-dir (eg "/home/<user>/.xbmc/userdata/addon_data/plugin.video.Mediathek/archives/"). Second a number to define how many threads could be used(anything less/equal 1 means no threading).
On windows you may need a separate python install.
if you got errors please check your XBMC.log (userdata/temp) and post the corresponding errormessage/stacktrace