Next time, please use the GitHub issue tracker (
https://github.com/pettarin/penelope/issues) to report problems with Penelope. I do not stop by here so often, while I get an email if an issue is filed there, and in theory this thread was not about conversion problems with Penelope.
Anyway, tropoy contacted me via email, and I converted his own files just fine with Penelope v3.1.2 on my machine, and he confirmed they work.
Assuming the problem is not with the dictionary installation procedure, but actually with the output files, I guess there is a problem with your shell/Python environment. It could be any of the following:
- shell encoding must be UTF-8
- a Windows-specific issue
- bug in the MARISA Python module on your environment
- bug in Penelope that does not occur on my environment
In general, if you want to check that the input file is read correctly, you can try outputting to CSV and inspect the resulting file. If it looks ok, then the problem is in the output part. If not, then the problem is with the reading part, most likely the shell encoding and/or the input files.
Anyway, if Penelope v2.x works for you, keep using it, you do not miss much.