tPj UdZddlmZddlZddlmZddlmZddlm Z dZ ee d<d e dzd e e e ffd Zd e d e e e fdzfd Zd ee zdedzd efdZy)z2 Shared utility functions for rate limiter hooks. )FinalN)verbose_proxy_logger)ModelGroupInfo)PriorityReservationDict litellm_proxyPROXY_LLM_PROVIDER_FALLBACKmodelreturnc |sdtfS tj|\}}}}|xs||xstfS#t$rK}t |}||cYd}~St j d|tt||tfcYd}~Sd}~wwxYw)uw Resolve ``(model, llm_provider)`` for a request being rejected by an internal proxy-side rate-limit hook. These hooks fire from ``async_pre_call_hook`` — well before :func:`litellm.get_llm_provider` is invoked anywhere else in the request lifecycle — so the raised 429 would otherwise have an empty ``llm_provider`` field, making the resulting Prometheus ``litellm_proxy_failed_requests_metric`` show up with ``exception_class="RateLimitError"`` and no provider attribution. Resolution order: 1. ``litellm.get_llm_provider(model)`` — covers raw provider/model strings the SDK already understands (``"gpt-4o-mini"``, ``"anthropic/claude-3-5-sonnet"``, ``"bedrock/..."`` etc.). 2. **Router alias fallback** — nearly every real proxy deployment routes through a router ``model_name`` alias (e.g. ``"tpm-locked"`` → ``litellm_params.model: openai/gpt-4o-mini``). ``get_llm_provider`` doesn't know router aliases, so without this step every alias call ended up labeled ``"litellm_proxy"``, defeating the field's purpose for the most common case. 3. Defensive fallback to ``("", "litellm_proxy")`` — used only when ``model`` is missing, malformed, or both lookups fail. We never let a secondary exception escape and mask the rate-limit error we're trying to surface. r Nz{rate_limiter_utils.resolve_llm_provider_for_rate_limit: could not resolve provider for model=%s, falling back to %s. err=%s)rlitellmget_llm_provider Exception#_resolve_provider_from_router_aliasrdebugstr)r resolved_modelcustom_llm_provider_ealias_resolutions k/home/oem/CyberBrain26-unified/.venv/lib/python3.12/site-packages/litellm/proxy/hooks/rate_limiter_utils.py#resolve_llm_provider_for_rate_limitrs< ...24;4L4L5 1+Q  #e  >#>    2"Ee"L  '# #"" R  ' F  111 2s'*7 B BB ,BB B c ddlm}|y t|dd}|sy|D]}t |t s|j d|k7r)|j d}t |t sK|j d}t |tr|so tj|\}}}}|s|xs||fcSy#t$rYywxYw#t$rYwxYw#t$rYywxYw) u Resolve a router ``model_name`` alias to ``(underlying_model, provider)`` by scanning the active router's ``model_list``. Returns ``None`` if the router isn't initialized, the alias isn't registered, the deployment has no usable ``litellm_params.model``, or any underlying lookup raises. Callers fall through to the defensive ``litellm_proxy`` fallback in that case — never raising secondary exceptions out of the rate-limit raise path. r) llm_routerN model_list model_namelitellm_paramsr r ) litellm.proxy.proxy_serverrrgetattr isinstancedictgetrrr) r rr deploymentparamsunderlying_modelrrrs rrrEs9 #J dC $ Jj$/~~l+u4^^$45Ffd+%zz'2 .4 0.9 (90%) 2. Dict with percent: {"type": "percent", "value": 0.9} -> 0.9 3. Dict with rpm: {"type": "rpm", "value": 900} -> 900/model_rpm 4. Dict with tpm: {"type": "tpm", "value": 900000} -> 900000/model_tpm Args: value: Priority value as float or dict with type/value keys model_info: Model configuration containing rpm/tpm limits Returns: float: Percentage value between 0.0 and 1.0 typepercentr(g?rpmrtpmN)r"intfloatr#r$r-r.)r(r)val_typeval_nums rconvert_priority_to_percentr3{s"%#u&U|%))FI67C0 y > !  :*..Z^^VWEW>JNN2 2  :*..Z^^VWEW>JNN2 2W~)__doc__typingrrlitellm._loggingrlitellm.types.routerrlitellm.types.utilsrr__annotations__rtuplerrr0r3r4rr=s1/7%4U432 :32 38_32l3 3 38_t3l u/F'F TbeiTi ns r4