AttributeError when trying to load replay

I get this error whenever I try to load a Replay. I tried reinstalling and installing directly from GitHub but the error persists. Any ideas on what could be causing this?

  File "<stdin>", line 1, in <module>
  File "/home/stefan/Documents/ML/env/lib/python3.6/site-packages/sc2reader/factories/sc2factory.py", line 85, in load_replay
    return self.load(Replay, source, options, **new_options)
  File "/home/stefan/Documents/ML/env/lib/python3.6/site-packages/sc2reader/factories/sc2factory.py", line 137, in load
    return self._load(cls, resource, filename=filename, options=options)
  File "/home/stefan/Documents/ML/env/lib/python3.6/site-packages/sc2reader/factories/sc2factory.py", line 146, in _load
    obj = cls(resource, filename=filename, factory=self, **options)
  File "/home/stefan/Documents/ML/env/lib/python3.6/site-packages/sc2reader/resources.py", line 312, in __init__
    engine.run(self)
  File "/home/stefan/Documents/ML/env/lib/python3.6/site-packages/sc2reader/engine/engine.py", line 177, in run
    for new_event in (event_handler(event, replay) or []):
  File "/home/stefan/Documents/ML/env/lib/python3.6/site-packages/sc2reader/engine/plugins/context.py", line 57, in handleTargetUnitCommandEvent
    self.last_target_ability_event[event.player.pid] = event
AttributeError: 'NoneType' object has no attribute 'pid'

[lfricken]

So I wanted to try solving this myself (I really want to use this library for those replays by blizzard). I had resources.py use replay.details.backup and replay.initData.backup and it successfully loaded some stuff, including players. But there were some problems.

replay.tracker.events doesn’t exist, so I can’t seem to find any UnitBornEvent’s

[StoicLoofah]

@Gusgus01 thanks for looking more deeply into this! My short answer is that I would love to have a good solution for this but don’t have a strong opinion on how to do it.

In the general case, it sounds like we won’t stumble across replays with this problem in the wild: they just happen to be how Blizzard treated this particular data set, but since they are official, that makes them pretty relevant.

As I mentioned above, I really don’t understand how the archive reading logic works: most of my tweaks to sc2reader start assuming that it was unpacked properly, then working to massage that into the python class structure. I would be happy to look at any PRs that manage to avoid crashes. From there, we can talk more about how we want to fill in details e.g. using pids or placeholder data.