I replaced the testcases with the desired classification of all of the current data set, provided by TILS, and then whacked the code to get it to run.
I'm not happy about the "known_state_actors" global variable, but it'll do as a short-term solution.
A long-term solution would store the "state actor?" bit in the database, and would cross-reference against the MPs table when inserting new entries, and would provide the possibility for admins to override the classification guess.
I replaced the testcases with the desired classification of all of the current data set, provided by TILS, and then whacked the code to get it to run.
I'm not happy about the "known_state_actors" global variable, but it'll do as a short-term solution.
A long-term solution would store the "state actor?" bit in the database, and would cross-reference against the MPs table when inserting new entries, and would provide the possibility for admins to override the classification guess.
Fixes #101.