+ ~j aRWth0tRt^RIt^RIt^RIt^RIt^RIt^RIt^RIt^RI t ^RI H t H t H t ^RIHt^RIHt^RIHt^RIHtHt^RIHt^RIt^RIHt^R IHt^R IHtH t H!t!H"t"H#t#H$t$H%t%H&t&^R IH't'^R I(H)t)^R I*H+t+]'PX!]-4t.]!RR7!RR44t/]P`!R4t1]P`!R4t2RRlt3Rt4Rt5Rt6]P`!R]PnR7t8]P`!R]PnR7t9]P`!R4t:]P`!R4t;0RXmtR"R#lt?R$R%lt@R&R'ltAR(R)ltBR*R+ltC] .]P3,tE] .]P3,tG]P!4tI]BsJ]^k]CsK]^kRsL]^kR,R-ltMR.R/ltNR0R1ltOR2R3ltPR4R5ltQ]P!]Q4]S!]R64'd]P!]QR77]P]P3tW]^kRYtX]^kR8^R9^R:^R;]WR<]XR=R>/R?R@lltYR8^R9^R:^R;]WR<]X/RARBlltZ]R8^R9^R:^R;]WR<]X/RCRDll4t[RER>/RFRGllt\RHRIlt]RZRJRKllt^]_!4t`RLRMlta]!RN]#RO7tbRPRQltcRRRSltd]P`!RT]P4tfRURVltgR#)[z>Contains utilities to handle HTTP requests in huggingface_hub.N)Callable GeneratorMapping)contextmanager) dataclass)quote)AnyTypeVar)urlparse)OfflineModeIsEnabled) constants)BadRequestErrorBucketNotFoundErrorDisabledRepoErrorGatedRepoErrorHfHubHTTPErrorRemoteEntryNotFoundErrorRepositoryNotFoundErrorRevisionNotFoundError)logging) SliceFileObj) HTTP_METHOD_TT)frozenc8a]tRt^6toRtRtRtV3RltRtVt R#) RateLimitInfoa Parsed rate limit information from HTTP response headers. Attributes: resource_type (`str`): The type of resource being rate limited. remaining (`int`): The number of requests remaining in the current window. reset_in_seconds (`int`): The number of seconds until the rate limit resets. limit (`int`, *optional*): The maximum number of requests allowed in the current window. window_seconds (`int`, *optional*): The number of seconds in the current window. Ncr<V^8dQh/S[;R&S[;R&S[;R&S[R,;R&S[R,;R&#) resource_type remainingreset_in_secondsNlimitwindow_secondsstrint)format __classdict__s"p/Users/mitch_tango/dev/rabbit-r1-livekit/agent/.venv/lib/python3.14/site-packages/huggingface_hub/utils/_http.py __annotate__RateLimitInfo.__annotate__6sMN !" :#$$J%%) __name__ __module__ __qualname____firstlineno____doc__r r!__annotate_func____static_attributes____classdictcell__)r&s@r'rr6s E!%N%r*rzL\"(?P\w+)\"\s*;\s*r\s*=\s*(?P\d+)\s*;\s*t\s*=\s*(?P\d+)z'q\s*=\s*(?P\d+).*?w\s*=\s*(?P\d+)c`V^8dQhR\\\3,R\R,/#)rheadersreturnN)rr#r)r%s"r'r(r(Rs(88WS#X%68=4;O8r*c6RpRpVF5pVP4pVR8Xd W,pK$VR8XgK-W,pK7 V'gR#\PV4pV'gR#VPR4p\ VPR44p\ VPR44pRp Rp V'dR\ PV4p V 'd5\ V PR44p \ V PR44p \ VVVV V R 7#) aParse rate limit information from HTTP response headers. Follows IETF draft: https://www.ietf.org/archive/id/draft-ietf-httpapi-ratelimit-headers-09.html Only a subset is implemented. Example: ```python >>> from huggingface_hub.utils import parse_ratelimit_headers >>> headers = { ... "ratelimit": '"api";r=0;t=55', ... "ratelimit-policy": '"fixed window";"api";q=500;w=300', ... } >>> info = parse_ratelimit_headers(headers) >>> info.remaining 0 >>> info.reset_in_seconds 55 ``` N ratelimitzratelimit-policyrrtqw)rrrr r!)lower_RATELIMIT_REGEXsearchgroupr$_RATELIMIT_POLICY_REGEXr) r5r8policykey lower_keymatchrrrr r! policy_matchs & r'parse_ratelimit_headersrGRs*!IFIIK  # I , ,\F    # #I .E KK0MEKK$%I5;;s+,E!%N .55f=  **3/0E !3!3C!89N #)%  r*z x-request-idzX-Amzn-Trace-Idz x-amz-cf-ida  # staging or production endpoint ^https://[^/]+ ( # on /api/repo_type/repo_id /api/(models|datasets|spaces)/(.+) | # or /repo_id/resolve/revision/... /(.+)/resolve/(.+) ) )flagszz # staging or production endpoint ^https?://[^/]+ # on /api/buckets/... /api/buckets/ zA^https?://[^/]+/api/(models|datasets|spaces)/([^/]+)(?:/([^/]+))?z)^https?://[^/]+/api/buckets/([^/]+/[^/]+)cnV^8dQhR\R\\R,\R,3,/#rurlr6N)r#tuple)r%s"r'r(r(s,35tS4Z1G+Hr*c"\PV4pV'gR#\PP VP ^44pVP ^4VP ^4rCV'dV\ 9d V RV 2pW%3#TpW%3#)aExtract (repo_type, repo_id) from an API URL. Returns canonical repo_type values: "model", "dataset", "space" (or None). Examples: >>> _parse_repo_info_from_url("https://huggingface.co/api/models/user/repo") ("model", "user/repo") >>> _parse_repo_info_from_url("https://huggingface.co/api/datasets/user/repo/resolve/main/data.csv") ("dataset", "user/repo") >>> _parse_repo_info_from_url("https://huggingface.co/api/models/bert-base-cased/resolve/main/config.json") ("model", "bert-base-cased") /NN)_REPO_ID_FROM_URL_REGEXr?r REPO_TYPES_MAPPINGgetr@_REPO_URL_SUBPATHS)rKrE repo_typefirstsecondrepo_ids& r'_parse_repo_info_from_urlrXs $ * *3 /E ,,00Q@IKKNEKKN6 & 22G1VH%    r*c>V^8dQhR\R\R,/#rJr#)r%s"r'r(r(s--3-3:-r*cb\PV4pV'dVP^4#R#)z9Extract bucket_id (namespace/name) from a bucket API URL.N)_BUCKET_ID_FROM_URL_REGEXr?r@)rKrEs& r'_parse_bucket_id_from_urlr]s' % , ,S 1E"5;;q>,,r*c<V^8dQhR\PRR/#rrequestr6NhttpxRequest)r%s"r'r(r(s5==Tr*c |\P!4'd\RVP R24h\VP 9dWVP P \4;'g\\P!44VP \&VP P \4p\PRVVPVPVP P R4RJ4\P'd \PR\V44V#)z Event hook that will be used to make HTTP requests to the Hugging Face Hub. What it does: - Block requests if offline mode is enabled - Add a request ID to the request headers - Log the request if debug mode is enabled z Cannot reach za: offline mode is enabled. To disable it, please unset the `HF_HUB_OFFLINE` environment variable.z%Request %s: %s %s (authenticated: %s) authorizationNzSend: %s)r is_offline_moder rKX_AMZN_TRACE_IDr5rR X_REQUEST_IDr#uuiduuid4loggerdebugmethodHF_DEBUG_curlify)r` request_ids& r'hf_request_event_hookrqs  """GKK=)J K  goo-+2??+>+>|+L+a+aPSTXT^T^T`Pa($$_5J LL/ O,D8  Z'!23 r*c<V^8dQhR\PRR/#r_ra)r%s"r'r(r(s**u}}**r*c "\V4#5i)z+ Async version of `hf_request_event_hook`. )rq)r`s&r'async_hf_request_event_hookrts ! ))s c<V^8dQhR\PRR/#rresponser6NrbResponse)r%s"r'r(r(s ' ' 'D 'r*c"VPR8dTRVP9dA\VPR,4pTR8dTP 4GRjxL R#R#R#R# \dR#i;iL5i)zContent-lengthNi@B) status_coder5r$ ValueErroraread)rwlengths& r'async_hf_response_event_hookrss" x// / X--.>?@ !nn&&&" 0#   's4#A=A)A=A; A=) A84A=7A88A=c8V^8dQhR\P/#rr6rbClient)r%s"r'r(r( s r*cB\P!R\./RRR7#)zI Factory function to create a `httpx.Client` with the default transport. r`TN event_hooksfollow_redirectstimeout)rbrrqr+r*r'default_client_factoryr s' <<!6 78 r*c8V^8dQhR\P/#rrb AsyncClient)r%s"r'r(r(se&7&7r*cP\P!R\.R\./RRR7#)zN Factory function to create a `httpx.AsyncClient` with the default transport. r`rwTNr)rbrrtrr+r*r'default_async_client_factoryrs1   !< =zLhKij r*c(V^8dQhR\RR/#)rclient_factoryr6N)CLIENT_FACTORY_T)r%s"r'r(r(+s00'70D0r*ct\;_uu_4\4VsRRR4R# +'giR#;i)a Set the HTTP client factory to be used by `huggingface_hub`. The client factory is a method that returns a `httpx.Client` object. On the first call to [`get_session`] the client factory will be used to create a new `httpx.Client` object that will be shared between all calls made by `huggingface_hub`. This can be useful if you are running your scripts in a specific environment requiring custom configuration (e.g. custom proxy or certifications). Use [`get_session`] to get a correctly configured `httpx.Client`. N) _CLIENT_LOCK close_session_GLOBAL_CLIENT_FACTORY)rs&r'set_client_factoryr+s  !/ s & 7 c(V^8dQhR\RR/#)rasync_client_factoryr6N)ASYNC_CLIENT_FACTORY_T)r%s"r'r(r(<s883I8d8r*c VsR#)a Set the HTTP async client factory to be used by `huggingface_hub`. The async client factory is a method that returns a `httpx.AsyncClient` object. This can be useful if you are running your scripts in a specific environment requiring custom configuration (e.g. custom proxy or certifications). Use [`get_async_client`] to get a correctly configured `httpx.AsyncClient`. Contrary to the `httpx.Client` that is shared between all calls made by `huggingface_hub`, the `httpx.AsyncClient` is not shared. It is recommended to use an async context manager to ensure the client is properly closed when the context is exited. N_GLOBAL_ASYNC_CLIENT_FACTORY)rs&r'set_async_client_factoryr<s $8 r*c8V^8dQhR\P/#rr)r%s"r'r(r(Os  U\\ r*c\f(\;_uu_4\4sRRR4\#\# +'gi\#;i)z Get a `httpx.Client` object, using the transport factory from the user. This client is shared between all calls made by `huggingface_hub`. Therefore you should not close it manually. Use [`set_client_factory`] to customize the `httpx.Client`. N)_GLOBAL_CLIENTrrr+r*r' get_sessionrOs5 \35N >\ s 6 A c8V^8dQhR\P/#rr)r%s"r'r(r(^s * *5,, *r*c\4#)a Return a `httpx.AsyncClient` object, using the transport factory from the user. Use [`set_async_client_factory`] to customize the `httpx.AsyncClient`. Contrary to the `httpx.Client` that is shared between all calls made by `huggingface_hub`, the `httpx.AsyncClient` is not shared. It is recommended to use an async context manager to ensure the client is properly closed when the context is exited. rr+r*r'get_async_sessionr^s  ( ))r*cV^8dQhRR/#)rr6Nr+)r%s"r'r(r(ns99t9r*c\pRsVeVP4R#R# \d$p\P RT 24Rp?R#Rp?ii;i)z Close the global `httpx.Client` used by `huggingface_hub`. If a Client is closed, it will be recreated on the next call to [`get_session`]. Can be useful if e.g. an SSL certificate has been updated. NzError closing client: )rclose Exceptionrkwarning)clientes r'rrnsTFN 9 LLN 9 NN3A37 8 8 9s" AA  Aregister_at_fork)after_in_child max_retriesbase_wait_time max_wait_timeretry_on_exceptionsretry_on_status_codesstreamFc@V^8dQhR\R\R\R\R\R\\ ,\ \\ ,R3,,R\\ \R3,,R \R \\PR R 3,/ #) rrmrKrrrr.rrr6N) rr#r$floattyperrLboolrrbry)r%s"r'r(r(s^8^8 ^8 ^8 ^8  ^8  ^8i5i#1E+FF^8sCx0^8 ^8u~~tT)*^8r*c +aaaaaa"\V\4'dV3p\S\4'dS3o^oTp RoRp RV9dE\VR,\P\ 34'dVR,P 4p \4p S^, oRoV eVR,PV 4RVVVVVV3Rllp V'dGV P!RRSRS/VB;_uu_4p V !V 4'gV xRRR4R#RRR4M+V P!RRSRS/VBp V !V 4'gV xR#Se3\!S4^,p\PR V R S R S R 24M!T p\PRV RS R S R 24\"P$!V4\'WI^,4p EK2 +'giL;i Td\p\PRT RS R S 24\T\P4'd \4SS8dThRp?LRp?ii;i5i)zfInternal implementation of HTTP backoff logic shared between `http_backoff` and `http_stream_backoff`.NdatacDV^8dQhR\PR\/#)rrwr6)rbryr)r%s"r'r((_http_backoff_base..__annotate__s   4 r*c<VPS9dR#\PRVP RS RS 24SS8d\V4R#VPR8Xd&\ VP 4pVe VP oR#)zNHandle response and return True if should retry, False if should return/yield.Fz HTTP Error z thrown while requesting  T)r|rkrhf_raise_for_statusrGr5r)rwratelimit_inforrmnb_triesratelimit_resetrrKs& r' _should_retry)_http_backoff_base.._should_retrys''/DD X-A-A,BB[\b[ccdehdijkk)'1!''3.%>KKSKFK$X.."N/  & 1A5L NN3L>AWX`Waabcnbooqr s%L NN\,y !K=XZ[ \ < Q7 ;ED#  NNQse#=fXQseL M#u1122+% &  szBI%1GG5G  GIGI#GGBI G GIGH?$AH:5I:H??IcV^8dQhR\R\R\R\R\R\\ ,\ \\ ,R3,,R\\ \R3,,R \P/#) rrmrKrrrr.rr6) rr#r$rrrrLrbry)r%s"r'r(r(sMM M M M  M  Mi5i#1E+FFMsCx0M ^^Mr*c J\\R RVRVRVRVRVRVRVRR /VB4#) ax Wrapper around httpx to retry calls on an endpoint, with exponential backoff. Endpoint call is retried on exceptions (ex: connection timeout, proxy error,...) and/or on specific status codes (ex: service unavailable). If the call failed more than `max_retries`, the exception is thrown or `raise_for_status` is called on the response object. Re-implement mechanisms from the `backoff` library to avoid adding an external dependencies to `hugging_face_hub`. See https://github.com/litl/backoff. Args: method (`Literal["GET", "OPTIONS", "HEAD", "POST", "PUT", "PATCH", "DELETE"]`): HTTP method to perform. url (`str`): The URL of the resource to fetch. max_retries (`int`, *optional*, defaults to `5`): Maximum number of retries, defaults to 5 (no retries). base_wait_time (`float`, *optional*, defaults to `1`): Duration (in seconds) to wait before retrying the first time. Wait time between retries then grows exponentially, capped by `max_wait_time`. max_wait_time (`float`, *optional*, defaults to `8`): Maximum duration (in seconds) to wait before retrying. retry_on_exceptions (`type[Exception]` or `tuple[type[Exception]]`, *optional*): Define which exceptions must be caught to retry the request. Can be a single type or a tuple of types. By default, retry on `httpx.TimeoutException` and `httpx.NetworkError`. retry_on_status_codes (`int` or `tuple[int]`, *optional*, defaults to `(429, 500, 502, 503, 504)`): Define on which status codes the request must be retried. By default, retries on rate limit (429) and server errors (5xx). **kwargs (`dict`, *optional*): kwargs to pass to `httpx.request`. Example: ``` >>> from huggingface_hub.utils import http_backoff # Same usage as "httpx.request". >>> response = http_backoff("GET", "https://www.google.com") >>> response.raise_for_status() # If you expect a Gateway Timeout from time to time >>> http_backoff("PUT", upload_url, data=data, retry_on_status_codes=504) >>> response.raise_for_status() ``` > [!WARNING] > When using `requests` it is possible to stream data by passing an iterator to the > `data` argument. On http backoff this is a problem as the iterator is not reset > after a failed call. This issue is mitigated for file objects or any IO streams > by saving the initial position of the cursor (with `data.tell()`) and resetting the > cursor between each call (with `data.seek()`). For arbitrary iterators, http backoff > will fail. If this is a hard constraint for you, please let us know by opening an > issue on [Github](https://github.com/huggingface/huggingface_hub). rmrKrrrrrrFr+)nextrrmrKrrrrrrs&&$$$$$,r' http_backoffrspB    $ * ( !4 #8     r*c4V^8dQhR\R\R\R\R\R\\ ,\ \\ ,R3,,R\\ \R3,,R \\PR R 3,/#) rrmrKrrrr.rr6N) rr#r$rrrrLrrbry)r%s"r'r(r(?sOO O O O  O  Oi5i#1E+FFOsCx0Ou~~tT)*Or*c+X"\R RVRVRVRVRVRVRVRR /VBR jxL R #L5i) a Wrapper around httpx to retry calls on an endpoint, with exponential backoff. Endpoint call is retried on exceptions (ex: connection timeout, proxy error,...) and/or on specific status codes (ex: service unavailable). If the call failed more than `max_retries`, the exception is thrown or `raise_for_status` is called on the response object. Re-implement mechanisms from the `backoff` library to avoid adding an external dependencies to `hugging_face_hub`. See https://github.com/litl/backoff. Args: method (`Literal["GET", "OPTIONS", "HEAD", "POST", "PUT", "PATCH", "DELETE"]`): HTTP method to perform. url (`str`): The URL of the resource to fetch. max_retries (`int`, *optional*, defaults to `5`): Maximum number of retries, defaults to 5 (no retries). base_wait_time (`float`, *optional*, defaults to `1`): Duration (in seconds) to wait before retrying the first time. Wait time between retries then grows exponentially, capped by `max_wait_time`. max_wait_time (`float`, *optional*, defaults to `8`): Maximum duration (in seconds) to wait before retrying. retry_on_exceptions (`type[Exception]` or `tuple[type[Exception]]`, *optional*): Define which exceptions must be caught to retry the request. Can be a single type or a tuple of types. By default, retry on `httpx.TimeoutException` and `httpx.NetworkError`. retry_on_status_codes (`int` or `tuple[int]`, *optional*, defaults to `(429, 500, 502, 503, 504)`): Define on which status codes the request must be retried. By default, retries on rate limit (429) and server errors (5xx). **kwargs (`dict`, *optional*): kwargs to pass to `httpx.request`. Example: ``` >>> from huggingface_hub.utils import http_stream_backoff # Same usage as "httpx.stream". >>> with http_stream_backoff("GET", "https://www.google.com") as response: ... for chunk in response.iter_bytes(): ... print(chunk) # If you expect a Gateway Timeout from time to time >>> with http_stream_backoff("PUT", upload_url, data=data, retry_on_status_codes=504) as response: ... response.raise_for_status() ``` When using `httpx` it is possible to stream data by passing an iterator to the `data` argument. On http backoff this is a problem as the iterator is not reset after a failed call. This issue is mitigated for file objects or any IO streams by saving the initial position of the cursor (with `data.tell()`) and resetting the cursor between each call (with `data.seek()`). For arbitrary iterators, http backoff will fail. If this is a hard constraint for you, please let us know by opening an issue on [Github](https://github.com/huggingface/huggingface_hub). rmrKrrrrrrTNr+)rrs&&$$$$$,r'http_stream_backoffr>spL"      &  $  0 4      s *(*retry_on_errorsc\V^8dQhR\R\R\R\P/#)rrmrKrr6)rr#rrbry)r%s"r'r(r(s000 0 #09=0 ^^0r*c jV'd/MRR RR /p\R RVRV/VBRR/BVBp\V4RVPu;8:dR8:dhMV#\VPR ,4pVP R 8Xd6\V4P VPR 7P4pKV#) aPerform an HTTP request with backoff and follow relative redirects only. Used to fetch HEAD /resolve on repo or bucket files. This is useful to follow a redirection to a renamed repository without following redirection to a CDN. A backoff mechanism retries the HTTP call on errors (429, 5xx, timeout, network errors). Args: method (`str`): HTTP method, such as 'GET' or 'HEAD'. url (`str`): The URL of the resource to fetch. retry_on_errors (`bool`, *optional*, defaults to `False`): Whether to retry on errors. If False, no retry is performed (fast fallback to local cache). If True, uses default retry behavior (429, 5xx, timeout, network errors). **httpx_kwargs (`dict`, *optional*): Params to pass to `httpx.request`. rrrmrKrFi,iLocation)pathr+) rrr|r r5netloc_replacergeturl)rmrKr httpx_kwargsno_retry_kwargsrw parsed_targets&&$, r'-_httpx_follow_relative_redirects_with_backoffrs0$92?VXZ#[     #     H% (&& -# -  O%X%5%5j%ABM##r)sm,,-2D2D,ELLN  Or*cJV^8dQhR\R\R,R\/#)rrKendpointNr6rZ)r%s"r'r(r(s%   sTz c r*c"V'dVPR4M\PpV\P\P39dAVP \PV4pVP \PV4pV#)zReplace the default endpoint in a URL by a custom one. This is useful when using a proxy and the Hugging Face Hub returns a URL with the default endpoint. rN)rstripr ENDPOINT_HF_DEFAULT_ENDPOINT_HF_DEFAULT_STAGING_ENDPOINTreplace)rKrs&&r'fix_hf_endpoint_in_urlrsf (0xs#Y5G5GH 66 8^8^__kk)88(Ckk)@@(K Jr*cVV^8dQhR\PR\R,RR/#)rrw endpoint_nameNr6)rbryr#)r%s"r'r(r(s/\?\?%..\?t\?W[\?r*c 6 \V4VP 4R# \d\PRRR7L7i;i \ P EdpTP^d,^8XdRp?R#TPPR4pTPPR4pTPe8TPPe \TPP4MRpT'd \T4MR6wrgTR8Xd?TP R2R ,R TP R 2,p\\YYgR 7ThTR 8Xd?TP R2R ,RTP R 2,p\\ YYgR 7ThTR8Xd?TP R2R ,RTP R 2,p\\"YYgR 7ThTR8XdKTP R2R ,RTP R 2,R,R,p\\$Y4ThTR8XdqTem\&P)T4eVTP R2R ,RTP R 2,R,R,p\\*Y\-T4R7ThTR8Xg4TPR8XdoTR8wdhTed\.P)T4eMTP R2R ,RTP R 2,R,R,p\\0YYgR 7ThTPR8XdTeRT R 2MR!p\\2Y4ThTPR"8XdAR TP R#T R 2R$TP R 2,R%,p\\4Y4ThTPR&8Xd\7TP4p T eR'T P8 R(2pTR)T P: R*2, pT P<eAT P>e3TR+T P@ R,T P< R-T P> R.2, pM TR , pTR/TP R 2, pMR0TP R 2p\\4Y4ThTPR18Xd[TPPPR24p T R3T R4TPPR54 R 2p\\4Y4Th\\4\T4T4ThRp?ii;i)7ad Internal version of `response.raise_for_status()` that will refine a potential HTTPError. Raised exception will be an instance of [`~errors.HfHubHTTPError`]. This helper is meant to be the unique method to raise_for_status when making a call to the Hugging Face Hub. Args: response (`Response`): Response from the server. endpoint_name (`str`, *optional*): Name of the endpoint that has been called. If provided, the error message will be more complete. > [!WARNING] > Raises when the request has failed: > > - [`~utils.RepositoryNotFoundError`] > If the repository to download from cannot be found. This may be because it > doesn't exist, because `repo_type` is not set correctly, or because the repo > is `private` and you do not have access. > - [`~utils.GatedRepoError`] > If the repository exists but is gated and the user is not on the authorized > list. > - [`~utils.RevisionNotFoundError`] > If the repository exists but the revision couldn't be found. > - [`~utils.EntryNotFoundError`] > If the repository exists but the entry (e.g. the requested file) couldn't be > find. > - [`~utils.BadRequestError`] > If request failed with a HTTP 400 BadRequest error. > - [`~utils.HfHubHTTPError`] > If request failed for a reason not listed above. zFailed to parse warning headersT)exc_infoNz X-Error-CodeX-Error-MessageRevisionNotFoundz Client Error. zRevision Not Found for url: .)rTrW EntryNotFoundzEntry Not Found for url: GatedRepoz!Cannot access gated repo for url z$Access to this resource is disabled.z!Cannot access repository for url  RepoNotFoundzBucket Not Found for url: zG Please make sure you specified the correct bucket id (namespace/name).zg If the bucket is private, make sure you are authenticated and your token has the required permissions.) bucket_idiz+Invalid credentials in Authorization headerzRepository Not Found for url: z Please make sure you specified the correct `repo_id` and `repo_type`. If you are trying to access a private or gated repo, make sure you are authenticated and your token has the required permissions.zQ For more details, see https://huggingface.co/docs/huggingface_hub/authenticationr{z Bad request for z endpoint:z Bad request:iz Forbidden: z Cannot access content at: z2 Make sure your token has the correct permissions.rz0 429 Too Many Requests: you have reached your 'z ' rate limit.z Retry after z seconds (rNz requests remaining in current z s window).z Url: z! 429 Too Many Requests for url: iRangez. Requested range: z. Content-Range: z Content-RangerO)!_warn_on_warning_headersrrkrlraise_for_statusrbHTTPStatusErrorr|r5rRr`rKr#rX_formatrrrrBUCKET_API_REGEXr?rr]REPO_API_REGEXrr rrGrrr r!r) rwrr error_code error_message request_urlrTrWmessager range_headers && r'rrsnBG * u?!!# G 6 FG  s?   3 &! + %%)).9  ((,,->? *2)9)9)E(JZJZJ^J^JjC  $$ %pt HS6{CXd + +!--.n=FKghphthtguuvIwwG/iiop p ? *!--.n=FKdemeqeqdrrsIttG2GQZlrs s ; &''(7&@EfgogsgsfttuCvv .'ybhi i D D''(75hll^1EF9 9 +W?Q F . (' '' 4@''(7.x||nA>?]]} } #WB[\gBh > )  C '!NN'%%k2>''(728<<.BC`` g g 17PYkqr r  ! !S (DQD]&}oZ@cu /7=1 D  ! !S (x++,LqI0a@AGH  .'gs?s4 A"AATTz-Hr#zutf-8ignore)r!i:NiNz ... [truncated]zz-drrr)curlN)rmsortedr5r4r=contentdecodelenrbRequestNotReadrrKr&rr#r1)r`partsr@rAbody flat_partss& r'rorosb  w~~$E w,,./ 779 'A 4A3b%&&0 D" ?? &??))'()CD4y4u+&67  4dB/011 tW[[! ""EJ   eCFm , 1   eCFm ,  88J    "!"s(AEE54E5z%^\s*bytes\s*=\s*(\d*)\s*-\s*(\d*)\s*$cXV^8dQhR\R,R\R\R,/#)roriginal_rangeN resume_sizer6r")r%s"r'r(r(s*"!"!t"!#"!#PT*"!r*cV'gRV R2#RV9d\RV: R24h\PV4pV'g\RV: R24hVP 4wr4V'gIV'g\RV: R24h\ V4V, pRV 2pV^8:d\R V: R24hV#\ V4pW1,pV'd,\ V4pRV RV 2pWt8d\R V: R24hV#RV R2#) z: Adjust HTTP Range header to account for resume position. zbytes=-,zMultiple ranges detected - z, not supported yet.zInvalid range format - rzbytes=-zEmpty new range - )r} RANGE_REGEXrEr*groupsr$)rOrPrEstartend new_suffix new_range new_starts&& r'_adjust_range_headerr[s!  }A&& n6~6HH\]^^   n -E 4^4FaHIIJE !88J!LM MX + j\* ?!3I=BC C JE#I #hYKq. ?!3I=BC C I;a  r*cLV^8dQh/^\9d \;R&^\9d \;R&^\9d\PR,;R&^\9d"\ \ \,R3,;R&^\9d\ \R3,;R&#)rrrNr._DEFAULT_RETRY_ON_EXCEPTIONS_DEFAULT_RETRY_ON_STATUS_CODES) __conditional_annotations__rrrbrrLrrr$)r%s"r'r(r(sDDpBA(AqErTS4SsEt+* t#*uEv iheDOS$89hw Ex LKc3hKy Er*> rawblobrefstreecommitresolverevisionsettings discussions)riiii)N)ir_r0atexitrr'osre threadingrricollections.abcrrr contextlibr dataclassesrshlexrtypingrr urllib.parser rbhuggingface_hub.errorsr rr r!r rrrrrrrr_lfsr_typingr get_loggerr,rkrcompiler>rArGrhrgr2VERBOSErrrPr\rSrXr]rqrtrrrrrrrLockrrrrrrrrrregisterhasattrrTimeoutException NetworkErrorr]r^rrrrrrsetrrrrro IGNORECASErTr[r()r_s@r'rsLE  88%!! 7   "   H % $&&&,::mn**%OP8z #   ** :: ****%ijJJ'STq2- B* 'B ,-!"e&7&7"78~~ +AA7SS&**0"8&  * 9,  2!""}5>C=S=SUZUgUg