Buckets:
AnonymousSpottedToad/candidate_bucket / preliminary /mappings /python /limit_request_rate /Daylily-Informatics_bloom.json
| { | |
| "role_mappings": [ | |
| { | |
| "role": "Entity", | |
| "status": "missing", | |
| "symbol": "", | |
| "file": "", | |
| "lines": "", | |
| "notes": "Callers of the rate-limited FastAPI app are external HTTP clients (browsers, API users) outside the repo; the Enforcer keys them by scope['client'] IP at bloom_lims/api/rate_limiting.py:340-341. Searched for in-repo JS/Python clients of /api/v1; only test_bloom_endpoints.py (test script) and outbound integrations (dewey client) exist, which do not count." | |
| }, | |
| { | |
| "role": "Enforcer", | |
| "status": "present", | |
| "symbol": "__call__", | |
| "file": "bloom_lims/api/rate_limiting.py", | |
| "lines": "322", | |
| "notes": "ASGI middleware RateLimitMiddleware.__call__, registered globally via app.add_middleware(RateLimitMiddleware) at bloom_lims/app.py:266 (on by default unless BLOOM_RATE_LIMIT=no). It extracts the client IP, calls self.limiter.is_allowed(client_id) at line 343, and returns 429 with Retry-After when blocked; excludes health paths and /api/v1/external/atlas/, and is bypassed when BLOOM_DISABLE_RATE_LIMITING is set." | |
| }, | |
| { | |
| "role": "Limiter", | |
| "status": "present", | |
| "symbol": "is_allowed", | |
| "file": "bloom_lims/api/rate_limiting.py", | |
| "lines": "119", | |
| "notes": "RateLimiter.is_allowed, called by the Enforcer at bloom_lims/api/rate_limiting.py:343. Refills a per-client token bucket, resets minute/hour windows, rejects on hourly limit or minute limit with no burst tokens, otherwise consumes a token and increments counters (updating history inline)." | |
| }, | |
| { | |
| "role": "HistoryProvider", | |
| "status": "present", | |
| "symbol": "_buckets", | |
| "file": "bloom_lims/api/rate_limiting.py", | |
| "lines": "104", | |
| "notes": "In-memory dict RateLimiter._buckets (defaultdict of ClientBucket dataclasses, line 62) keyed by client id; read at line 134 and updated at lines 137-180 inside is_allowed (the Limiter). No dedicated getter/updater method exists on the traced path, so the field is cited; per-process only, not shared across workers." | |
| }, | |
| { | |
| "role": "PolicyProvider", | |
| "status": "present", | |
| "symbol": "__init__", | |
| "file": "bloom_lims/api/rate_limiting.py", | |
| "lines": "83", | |
| "notes": "RateLimiter.__init__ sets the policy (requests_per_minute, requests_per_hour, burst_size, enabled) and tries to override per-minute limit from get_settings().api.rate_limit_per_minute (line 115-117); instantiated by RateLimitMiddleware.__init__ at line 316 with defaults 60/min and 1000/h. Caveat: APISettings (bloom_lims/config.py:645) defines rate_limit_requests/rate_limit_window_seconds but not rate_limit_per_minute, so the config values are not actually used and the hard-coded defaults apply." | |
| }, | |
| { | |
| "role": "System", | |
| "status": "present", | |
| "symbol": "list_objects", | |
| "file": "bloom_lims/api/v1/objects.py", | |
| "lines": "41", | |
| "notes": "Example protected handler: GET /api/v1/objects/ (objects router included in api_v1_router, bloom_lims/api/v1/__init__.py:45, mounted at bloom_lims/app.py:272). The middleware forwards allowed requests via await self.app(scope, receive, send) at bloom_lims/api/rate_limiting.py:370; it guards all non-excluded routes, this one is representative." | |
| } | |
| ] | |
| } | |
Xet Storage Details
- Size:
- 3.31 kB
- Xet hash:
- bd71a22e29d4eebfaef4845080f49204e2da02f40415d4e937409d75b6879292
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.