The new functionality in #4 introduced a new csv: player_mapping.csv. This is needed due to inconsistencies in player names across the different rankings, which influences matching the rosters (without for example techno and techno4k would not be considered the same player).
Since there were a lot of players at lower ranked teams that had this problem (namings for them can be even more inconsistent than for higher ranked teams) and I wanted to get this new feature out, I have now ignored those. Ideally, someone goes over all rosters in rosters.csv and checks whether there are obvious cases like techno/techno4k above, and then adds them to player_mapping.csv. Also, rosters.csv would then need to be cleaned up of duplicate rosters that are actually the same ones in the curr_roster and old_rosters columns.
The new functionality in #4 introduced a new csv:
player_mapping.csv. This is needed due to inconsistencies in player names across the different rankings, which influences matching the rosters (without for exampletechnoandtechno4kwould not be considered the same player).Since there were a lot of players at lower ranked teams that had this problem (namings for them can be even more inconsistent than for higher ranked teams) and I wanted to get this new feature out, I have now ignored those. Ideally, someone goes over all rosters in
rosters.csvand checks whether there are obvious cases liketechno/techno4kabove, and then adds them toplayer_mapping.csv. Also,rosters.csvwould then need to be cleaned up of duplicate rosters that are actually the same ones in thecurr_rosterandold_rosterscolumns.