Character set testing for OBF

eQualityTime / TheOpenVoiceFactory

The Open Voice Factory - open software for people with communication difficulties

http://theopenvoicefactory.org

23 stars 9 forks source link

Character set testing for OBF #126

Closed joereddington closed 3 years ago

joereddington commented 6 years ago

argument is Traceback (most recent call last): File "grab_text.py", line 543, in write_to_obf(pageset.grids, dest) File "grab_text.py", line 481, in write_to_obf create_obf_manifest(boards_names_dic,image_names_dic, dest) File "grab_text.py", line 363, in create_obf_manifest root = boards_names_dic['toppage'] KeyError: 'toppage' Josephs-Mini:openvoicefactory josephreddington$

joereddington commented 6 years ago

Spent some time on this today. Turns out it revealed a few bits that weren't working very well.

There's now a list of what needs to happen.

[ ] the other language versions should be checked manually to make sure they really do work.
[ ] add the brazil version to the test cases for regression (which will also let us run the localcreate command easily on the staging server.
[ ] regression tests need to include the obz files as well.
[ ] all ascii('ignores') should be made into ascii('replaces')

joereddington commented 6 years ago

Actually here's the plan.

Get the new file to be part of the tests.
Remove all of the encode/decode commands
Run the tests
Put in all the exception handling we need to properly understand this.
Put in the correct encode/decode commands.

joereddington commented 3 years ago

Dealt with in the June updates