Improve balldontlie query flow and dev container write stability

apps/providers/adapters/balldontlie_provider.py

@@ -1,4 +1,5 @@
 import logging
+from itertools import islice
 
 from django.conf import settings
 
@@ -13,6 +14,7 @@ from apps.providers.contracts import (
     TeamPayload,
 )
 from apps.providers.interfaces import BaseProviderAdapter
+from apps.providers.exceptions import ProviderUnauthorizedError
 from apps.providers.services.balldontlie_mappings import (
     map_competitions,
     map_player_stats,
@@ -36,6 +38,66 @@ class BalldontlieProviderAdapter(BaseProviderAdapter):
     def configured_seasons(self) -> list[int]:
         return settings.PROVIDER_BALLDONTLIE_SEASONS
 
+    @staticmethod
+    def _chunked(values: list[int], size: int):
+        iterator = iter(values)
+        while True:
+            chunk = list(islice(iterator, size))
+            if not chunk:
+                return
+            yield chunk
+
+    def _fetch_game_ids(self) -> list[int]:
+        game_ids: set[int] = set()
+        for season in self.configured_seasons:
+            rows = self.client.list_paginated(
+                "games",
+                params={"seasons[]": season},
+                per_page=settings.PROVIDER_BALLDONTLIE_GAMES_PER_PAGE,
+                page_limit=settings.PROVIDER_BALLDONTLIE_GAMES_PAGE_LIMIT,
+            )
+            for row in rows:
+                game_id = row.get("id")
+                if isinstance(game_id, int):
+                    game_ids.add(game_id)
+        return sorted(game_ids)
+
+    def _fetch_stats_rows(self) -> list[dict]:
+        game_ids = self._fetch_game_ids()
+        if not game_ids:
+            logger.info(
+                "provider_stats_skipped_no_games",
+                extra={"provider": self.namespace, "seasons": self.configured_seasons},
+            )
+            return []
+
+        all_rows: list[dict] = []
+        try:
+            # Use game_ids[] query as documented in balldontlie getting-started flow.
+            for game_id_chunk in self._chunked(game_ids, 25):
+                rows = self.client.list_paginated(
+                    "stats",
+                    params={"game_ids[]": game_id_chunk},
+                    per_page=settings.PROVIDER_BALLDONTLIE_STATS_PER_PAGE,
+                    page_limit=settings.PROVIDER_BALLDONTLIE_STATS_PAGE_LIMIT,
+                )
+                all_rows.extend(rows)
+        except ProviderUnauthorizedError as exc:
+            if settings.PROVIDER_BALLDONTLIE_STATS_STRICT:
+                raise
+            logger.warning(
+                "provider_stats_unauthorized_degraded",
+                extra={
+                    "provider": self.namespace,
+                    "path": exc.path,
+                    "status_code": exc.status_code,
+                    "detail": exc.detail,
+                },
+            )
+            return []
+
+        return all_rows
+
     def search_players(self, *, query: str = "", limit: int = 50, offset: int = 0) -> list[PlayerPayload]:
         params = {"search": query} if query else None
         rows = self.client.list_paginated(
@@ -78,30 +140,12 @@ class BalldontlieProviderAdapter(BaseProviderAdapter):
         return map_seasons(self.configured_seasons)
 
     def fetch_player_stats(self) -> list[PlayerStatsPayload]:
-        all_rows: list[dict] = []
-        for season in self.configured_seasons:
-            rows = self.client.list_paginated(
-                "stats",
-                params={"seasons[]": season},
-                per_page=settings.PROVIDER_BALLDONTLIE_STATS_PER_PAGE,
-                page_limit=settings.PROVIDER_BALLDONTLIE_STATS_PAGE_LIMIT,
-            )
-            all_rows.extend(rows)
-
+        all_rows = self._fetch_stats_rows()
         player_stats, _ = map_player_stats(all_rows, allowed_seasons=self.configured_seasons)
         return player_stats
 
     def fetch_player_careers(self) -> list[PlayerCareerPayload]:
-        all_rows: list[dict] = []
-        for season in self.configured_seasons:
-            rows = self.client.list_paginated(
-                "stats",
-                params={"seasons[]": season},
-                per_page=settings.PROVIDER_BALLDONTLIE_STATS_PER_PAGE,
-                page_limit=settings.PROVIDER_BALLDONTLIE_STATS_PAGE_LIMIT,
-            )
-            all_rows.extend(rows)
-
+        all_rows = self._fetch_stats_rows()
         _, player_careers = map_player_stats(all_rows, allowed_seasons=self.configured_seasons)
         return player_careers
 
@@ -115,16 +159,7 @@ class BalldontlieProviderAdapter(BaseProviderAdapter):
         seasons = self.fetch_seasons()
         players = self.fetch_players()
 
-        all_rows: list[dict] = []
-        for season in self.configured_seasons:
-            rows = self.client.list_paginated(
-                "stats",
-                params={"seasons[]": season},
-                per_page=settings.PROVIDER_BALLDONTLIE_STATS_PER_PAGE,
-                page_limit=settings.PROVIDER_BALLDONTLIE_STATS_PAGE_LIMIT,
-            )
-            all_rows.extend(rows)
-
+        all_rows = self._fetch_stats_rows()
         player_stats, player_careers = map_player_stats(all_rows, allowed_seasons=self.configured_seasons)
 
         logger.info(