fix(Wyscout v3): Handle unrecognized players gracefully to prevent cr…

…ashes (#358)

kloppy/exceptions.py

@@ -40,3 +40,7 @@ class UnknownEncoderError(KloppyError):
 
 class KloppyParameterError(KloppyError):
     pass
+
+
+class DeserializationWarning(Warning):
+    pass

kloppy/infra/serializers/event/wyscout/deserializer_v3.py

@@ -1,5 +1,6 @@
 import json
 import logging
+import warnings
 from dataclasses import replace
 from datetime import datetime, timedelta, timezone
 from enum import Enum
@@ -37,7 +38,7 @@
     TakeOnResult,
     Team,
 )
-from kloppy.exceptions import DeserializationError
+from kloppy.exceptions import DeserializationError, DeserializationWarning
 from kloppy.utils import performance_logging
 
 from ..deserializer import EventDataDeserializer
@@ -780,6 +781,19 @@ def deserialize(self, inputs: WyscoutInputs) -> EventDataset:
                         str(raw_event["possession"]["team"]["id"])
                     ]
 
+                if player_id == INVALID_PLAYER:
+                    player = None
+                elif player_id not in players[team_id]:
+                    player = None
+                    warnings.warn(
+                        f"Event {raw_event['id']} was performed by player {player_id} and team {team_id}, "
+                        f"but the player does not appear to be part of that team's lineup. "
+                        f"Handled by setting the event's player to None.",
+                        DeserializationWarning,
+                    )
+                else:
+                    player = players[team_id][player_id]
+
                 generic_event_args = {
                     "event_id": raw_event["id"],
                     "raw_event": raw_event,
@@ -792,11 +806,7 @@ def deserialize(self, inputs: WyscoutInputs) -> EventDataset:
                         else None
                     ),
                     "team": team,
-                    "player": (
-                        players[team_id][player_id]
-                        if player_id != INVALID_PLAYER
-                        else None
-                    ),
+                    "player": player,
                     "ball_owning_team": ball_owning_team,
                     "ball_state": None,
                     "period": periods[-1],

kloppy/tests/prs/pr_358/test_pr_358.py

@@ -0,0 +1,19 @@
+import pytest
+
+from kloppy import wyscout
+from kloppy.exceptions import DeserializationWarning
+
+
+def test_ignore_unknown_player(base_dir):
+    with pytest.warns(
+        DeserializationWarning,
+        match="the player does not appear to be part of that team's lineup",
+    ):
+        dataset = wyscout.load(
+            event_data=base_dir / "prs" / "pr_358" / "wyscout_events_v3.json",
+            coordinates="wyscout",
+        )
+
+        assert len(dataset.events) == 2
+
+        assert dataset.events[1].player is None