Buckets:

download
raw
2.99 kB
{
"role_mappings": [
{
"role": "Entity",
"status": "present",
"symbol": "_request",
"file": "python-client/neurograph/client.py",
"lines": "102",
"notes": "In-repo SDK NeuroGraphClient._request sends HTTP requests to the NeuroGraph API (e.g. TokensAPI.create -> POST /api/v1/tokens, client.py:224) which pass through the Enforcer dispatch (src/api/middlewares/rate_limit.py:178); _handle_error_response (client.py:174) maps the Enforcer's 429 to RateLimitError."
},
{
"role": "Enforcer",
"status": "present",
"symbol": "dispatch",
"file": "src/api/middlewares/rate_limit.py",
"lines": "178",
"notes": "RateLimitMiddleware.dispatch, registered via app.add_middleware(RateLimitMiddleware, ...) at src/api/main.py:114, intercepts every incoming HTTP request (health exempt), derives the id via get_user_id (rate_limit.py:110: user/API-key prefix/IP), asks the bucket's consume (line 191) and returns 429 with Retry-After when blocked, else call_next (line 225)."
},
{
"role": "Limiter",
"status": "present",
"symbol": "consume",
"file": "src/api/middlewares/rate_limit.py",
"lines": "37",
"notes": "Called by Enforcer dispatch at rate_limit.py:191. TokenBucket.consume refills tokens by elapsed time up to capacity and returns True (allow) if a token is available, False (block) otherwise; it also updates the bucket state (tokens/last_update)."
},
{
"role": "HistoryProvider",
"status": "present",
"symbol": "get_bucket",
"file": "src/api/middlewares/rate_limit.py",
"lines": "133",
"notes": "Called by Enforcer dispatch at rate_limit.py:188 to fetch the per-id TokenBucket from the in-memory dict self.buckets (declared line 105) and record last_access; the bucket's tokens/last_update are the request history consumed and updated by Limiter consume (line 37). Stale buckets pruned by cleanup_old_buckets (line 155)."
},
{
"role": "PolicyProvider",
"status": "present",
"symbol": "get_bucket",
"file": "src/api/middlewares/rate_limit.py",
"lines": "133",
"notes": "Same method as HistoryProvider: when creating a bucket, get_bucket (reached from dispatch at rate_limit.py:188) selects the policy per id type (ip_ -> ip_rate_limit, apikey_/user -> default_rate_limit, lines 138-147), which becomes the bucket capacity used by consume. Policy values configured at src/api/main.py:116-117 (100 / 30)."
},
{
"role": "System",
"status": "present",
"symbol": "create_token",
"file": "src/api/routers/tokens.py",
"lines": "138",
"notes": "Representative protected handler reached via call_next (rate_limit.py:225) after the Enforcer allows the request; POST /api/v1/tokens (router included at src/api/main.py:147) is the endpoint the SDK Entity calls via TokensAPI.create. All other /api/v1 routers are guarded the same way."
}
]
}

Xet Storage Details

Size:
2.99 kB
·
Xet hash:
8487dd1954d6f8c8faf5fc91e6b91fe51b3b14548b702fa42bebe07fa4ffa173

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.