bisco/hoopscout
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Improve balldontlie query flow and dev container write stability

Browse Source
This commit is contained in:
Alfredo Di Stasio
2026-03-12 11:13:05 +01:00
parent e0e75cfb0c
commit c9dd10a438
10 changed files with 196 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files

95
apps/providers/adapters/balldontlie_provider.py
View File

@ -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(

39
apps/providers/clients/balldontlie.py
View File

@ -5,7 +5,7 @@ from typing import Any
import requests
from django.conf import settings
from apps.providers.exceptions import ProviderRateLimitError, ProviderTransientError
from apps.providers.exceptions import ProviderRateLimitError, ProviderTransientError, ProviderUnauthorizedError
logger = logging.getLogger(__name__)
@ -89,9 +89,14 @@ class BalldontlieClient:
if status >= 400:
body_preview = response.text[:240]
raise ProviderTransientError(
f"balldontlie client error status={status} path={path} body={body_preview}"
)
if status == 401:
raise ProviderUnauthorizedError(
provider="balldontlie",
path=path,
status_code=status,
detail=body_preview,
)
raise ProviderTransientError(f"balldontlie client error status={status} path={path} body={body_preview}")
try:
return response.json()
@ -109,20 +114,36 @@ class BalldontlieClient:
page_limit: int = 1,
) -> list[dict[str, Any]]:
page = 1
cursor = None
rows: list[dict[str, Any]] = []
query = dict(params or {})
while page <= page_limit:
query.update({"page": page, "per_page": per_page})
payload = self.get_json(path, params=query)
request_query = dict(query)
request_query["per_page"] = per_page
if cursor is not None:
request_query["cursor"] = cursor
else:
# Keep backwards compatibility for endpoints still supporting page-based pagination.
request_query["page"] = page
payload = self.get_json(path, params=request_query)
data = payload.get("data") or []
if isinstance(data, list):
rows.extend(data)
meta = payload.get("meta") or {}
next_cursor = meta.get("next_cursor")
if next_cursor:
cursor = next_cursor
page += 1
continue
next_page = meta.get("next_page")
if not next_page:
break
page = int(next_page)
if next_page:
page = int(next_page)
continue
break
return rows

11
apps/providers/exceptions.py
View File

@ -6,6 +6,17 @@ class ProviderTransientError(ProviderError):
"""Temporary provider failure that can be retried."""
class ProviderUnauthorizedError(ProviderError):
"""Raised when provider credentials are valid format but not authorized for an endpoint."""
def __init__(self, *, provider: str, path: str, status_code: int, detail: str = ""):
super().__init__(f"{provider} unauthorized status={status_code} path={path} detail={detail}")
self.provider = provider
self.path = path
self.status_code = status_code
self.detail = detail
class ProviderRateLimitError(ProviderTransientError):
"""Raised when provider rate limit is hit."""