dhSrSSKrSSKJrJrJrJrJr SSKJ r SSK J r \R"\ 5rSSKJr "SS\5r"S S \ 5rg!\a SrN!f=f) z0Ratelimiting Handler to limit requests or tokensN)AnyDictListLiteralOptional)BaseCallbackHandler) LLMResult) Ratelimitc Z^\rSrSrSrS S\S\SS\\S\\ 4U4Sjjjr S r U=r $) UpstashRatelimitErrorz] Upstash Ratelimit Error Raised when the rate limit is reached in `UpstashRatelimitHandler` messagetype)tokenrequestlimitresetcH>[TU]U5 X lX0lX@lg)aZ Args: message (str): error message type (str): The kind of the limit which was reached. One of "token" or "request" limit (Optional[int]): The limit which was reached. Passed when type is request reset (Optional[int]): unix timestamp in milliseconds when the limits are reset. Passed when type is request N)super__init__rrr)selfrrrr __class__s p/var/www/html/shao/venv/lib/python3.13/site-packages/langchain_community/callbacks/upstash_ratelimit_callback.pyrUpstashRatelimitError.__init__s!$ !   )rrr)NN) __name__ __module__ __qualname____firstlineno____doc__strrrintfloatr__static_attributes__ __classcell__)rs@rr r sN $!% ()}   rr c \rSrSr%SrSr\\S'Sr\\S'SSSS.S \ S \ \ S \ \ S \4S jjr S\ \ \4S\ \ \4S\S\4SjrS\ \ \4S\\ S\SS4SjrS\S\SS4SjrSS \ \ SS4SjjrSrg)UpstashRatelimitHandler/a Callback to handle rate limiting based on the number of requests or the number of tokens in the input. It uses Upstash Ratelimit to track the ratelimit which utilizes Upstash Redis to track the state. Should not be passed to the chain when initialising the chain. This is because the handler has a state which should be fresh every time invoke is called. Instead, initialise and pass a handler every time you invoke. T raise_errorF_checkedN)token_ratelimitrequest_ratelimitinclude_output_tokens identifierr+r,r-cl[X#/5(d [S5eXlX lX0lX@lg)a Creates UpstashRatelimitHandler. Must be passed an identifier to ratelimit like a user id or an ip address. Additionally, it must be passed at least one of token_ratelimit or request_ratelimit parameters. Args: identifier Union[int, str]: the identifier token_ratelimit Optional[Ratelimit]: Ratelimit to limit the number of tokens. Only works with OpenAI models since only these models provide the number of tokens as information in their output. request_ratelimit Optional[Ratelimit]: Ratelimit to limit the number of requests include_output_tokens bool: Whether to count output tokens when rate limiting based on number of tokens. Only used when `token_ratelimit` is passed. False by default. Example: .. code-block:: python from upstash_redis import Redis from upstash_ratelimit import Ratelimit, FixedWindow redis = Redis.from_env() ratelimit = Ratelimit( redis=redis, # fixed window to allow 10 requests every 10 seconds: limiter=FixedWindow(max_requests=10, window=10), ) user_id = "foo" handler = UpstashRatelimitHandler( identifier=user_id, request_ratelimit=ratelimit ) # Initialize a simple runnable to test chain = RunnableLambda(str) # pass handler as callback: output = chain.invoke( "input", config={ "callbacks": [handler] } ) zhYou must pass at least one of input_token_ratelimit or request_ratelimit parameters for handler to work.N)any ValueErrorr.r+r,r-)rr.r+r,r-s rr UpstashRatelimitHandler.__init__@s?tO788E  %.!2%:"r serializedinputskwargsreturnc UR(arUR(d`URRUR5nUR(d"[ SSURUR 5eSUlggg)a, Run when chain starts running. on_chain_start runs multiple times during a chain execution. To make sure that it's only called once, we keep a bool state `_checked`. If not `self._checked`, we call limit with `request_ratelimit` and raise `UpstashRatelimitError` if the identifier is rate limited. zRequest limit reached!rTN)r,r*rr.allowedr r)rr3r4r5responses ron_chain_start&UpstashRatelimitHandler.on_chain_startse  ! !$----33DOODH##+,i!DM +8 !rpromptsc UR(a8URRUR5nUS::a [SS5egg)z Run when LLM starts running rzToken limit reached!rN)r+ get_remainingr.r )rr3r<r5 remainings r on_llm_start$UpstashRatelimitHandler.on_llm_startsF   ,,::4??KIA~+,BGLL rr9c UR(a[UR=(d 0nUSnUR(aUSOUSnURR UR US9 gg![a [ S5ef=f)z Run when LLM ends running If the `include_output_tokens` is set to True, number of tokens in LLM completion are counted for rate limiting token_usage total_tokens prompt_tokenszLLM response doesn't include `token_usage: {total_tokens: int, prompt_tokens: int}` field. To use UpstashRatelimitHandler with token_ratelimit, either use a model which returns token_usage (like OpenAI models) or rate limit only with request_ratelimit.)rateN)r+ llm_outputr-KeyErrorr1rr.)rr9r5rGrC token_counts r on_llm_end"UpstashRatelimitHandler.on_llm_ends    %006B (7 11 /$_5"  & &t[ & I+   Q s 5A..Bc|[U=(d URURURURS9$)z Creates a new UpstashRatelimitHandler object with the same ratelimit configurations but with a new identifier if it's provided. Also resets the state of the handler. )r.r+r,r-)r'r.r+r,r-)rr.s rrUpstashRatelimitHandler.resets8'!4T__ 00"44"&"<"<   r)r*r.r-r,r+)N)rrrrr r)bool__annotations__r*r!rr rrrr:rr@r rJrr$rrr'r'/s  KHd 0415&+ C;C;"), C; $I. C; $ C;J!sCx.!26sCx.!LO! !& MsCx. M379 MHK M  MJ9JJJ<    9R   rr')r loggingtypingrrrrrlangchain_core.callbacksrlangchain_core.outputsr getLoggerrloggerupstash_ratelimitr ImportError Exceptionr r'rPrrrZsa6558,   8 $+ I>_ 1_ GIsAAA