GrammaticalFramework/gf-wordnet

In Python, German and French downloads fail

stephanemagnenat opened this issue · 2 comments

Following the instruction of the Python README, the following works for English:

>>> wordnet.download(['eng'])
Download and boot the grammar 355MB (Expanded to 2719MB)
Download the semantics database 2814MB done
Reload wordnet

but fails for French and German:

wordnet.download(['fra'])
Download and boot the grammar 138MBTraceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/steph/.local/lib/python3.10/site-packages/wordnet/__init__.py", line 54, in download
    pgf.bootNGF(readinto, path+"Parse.ngf")
pgf.PGFError: reached end of file while reading the grammar

I am using Python 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0] on linux, on Ubuntu 22.04.1, with wordnet installed using pip3 as explained in the README.

We use 'fre' for French and 'ger' for German

Thank you for the answer! I got confused because the README talks about using "ISO 639‑2" and I did not realize these languages had two codes and here the "bibliographic" is used. Maybe in term of user experience, a feature wish could be that the user-facing API could accept both?