Error storing file names containing unicode characters in database about viper HOT 4 CLOSED

I've been experimenting with this and still haven't found a proper solution.

from viper.

Whilst not ideal for retaining the original name you could use the string_clean function. This would at least enable it to be stored. then warn that a file was renamed

from viper.

I have added a pull-request that should fix this issue.

Some more info given in the pull-request:
I had a look at this as this is affecting me too.
The main issue is that the __sessions__.current.file.name is used in different ways around in the code. (str, unicode). I would use unicode everywhere and only convert when legacy functions (like raw_imput) can't handle it. Unfortunately it is a massive and unrealistic job to fix this everywhere.

To resolve this I have added the verification and conversion steps in the most efficient locations.
Going from std_in, the storing of files and the database function.

I have tested opening files (with and without special chars), storing them, loading them from the db, and it seems to work. As a test I have also ran multiple modules with no issue.

I expect some issues to occur here and there, but only with files containing special chars. So this patch should not break existing functionality, but instead extend it.
Mostly solution for the encountered issues will be to encode the unicode variable as can be seen below.

__sessions__.current.file.name.encode('utf-8')

from viper.

Must be fixed since the py3 branch is merged.

from viper.