French blogs seem to use UTF-8 characters in the feeds. This brokes update.
We could easily get rid of this limitation migrating the application from Python 2 to Python 3: that will allow us to get UTF-8 support for free.
Alternatively, if we want to stay in Python 2, it's also possible to use an UTF-8 "codec" and prefix all the strings with u in the code.
This task means we need to fork and support Venus, or get in touch with original authors, as it hasn't been updated for 5 years and no alternative seems to be available.