Buckets:

download
raw
2.67 kB
{
"role_mappings": [
{
"role": "Entity",
"status": "present",
"symbol": "chat",
"file": "effgen/client/client.py",
"lines": "230",
"notes": "In-repo effGen client SDK; chat POSTs to the effGen server's /v1/chat/completions (client.py:256), which passes through the RateLimitMiddleware Enforcer; the client also retries/backs off on 429 (client.py:124)."
},
{
"role": "Enforcer",
"status": "present",
"symbol": "__call__",
"file": "effgen/api/middleware.py",
"lines": "293",
"notes": "__call__ of the pure-ASGI RateLimitMiddleware, registered via app.add_middleware in install_production_middleware (middleware.py:153), called from create_app (effgen/server/app.py:342). It keys on client IP, calls _check (middleware.py:302) and returns 429 with Retry-After when blocked; caveat: only installed when EFFGEN_RATE_LIMIT / --rate-limit is a positive integer (disabled by default)."
},
{
"role": "Limiter",
"status": "present",
"symbol": "_check",
"file": "effgen/api/middleware.py",
"lines": "269",
"notes": "Called by the Enforcer __call__ at middleware.py:302; implements a fixed-window per-IP counter, comparing the bucket count against self.limit and returning (allowed, remaining, reset), updating the bucket on allow."
},
{
"role": "HistoryProvider",
"status": "present",
"symbol": "_buckets",
"file": "effgen/api/middleware.py",
"lines": "254",
"notes": "In-memory dict client-ip -> [window_start, count] field of RateLimitMiddleware, read and updated inline by _check (middleware.py:269-291); no dedicated accessor method exists, so the field is cited. Per-process only."
},
{
"role": "PolicyProvider",
"status": "present",
"symbol": "_resolve_rate_limit",
"file": "effgen/api/middleware.py",
"lines": "26",
"notes": "Called in install_production_middleware (middleware.py:150); resolves requests/minute from the create_app/CLI parameter or EFFGEN_RATE_LIMIT env var and passes it to RateLimitMiddleware, stored as self.limit (middleware.py:250) and used by _check. Single global policy (with probe-path exemptions at middleware.py:213)."
},
{
"role": "System",
"status": "present",
"symbol": "chat_completions",
"file": "effgen/api/openai_compat.py",
"lines": "361",
"notes": "Allowed requests are forwarded by __call__ via self.app(...) (middleware.py:320) to the routed handlers; chat_completions is the /v1/chat/completions handler of the router mounted in effgen/server/app_runner.py:490."
}
]
}

Xet Storage Details

Size:
2.67 kB
·
Xet hash:
fc73c62ed7293054b5f2262d15ea45bb3776715bd3066d0e18b04927dcbeba41

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