uPj"xUdZddlmZddlmZmZmZmZddlZddl m Z ddl m Z ddl mZerddlmZneZeeefZed Zd Zeeed fed <d Zeed <dZeed<dZeed<dedefdZGddeZGddeZGddeZ GddeZ!GddeZ"Gdd eZ#Gd!d"Z$y)#z( Base OCR transformation configuration. )Mapping) TYPE_CHECKINGAnyFinalLiteralN) PrivateAttrBaseLLMException)LiteLLMPydanticObjectBase)Logging)litellmnative.OCR_REQUEST_FORMATS req_formatOCR_REQUEST_FORMAT_PARAMz x-req-formatOCR_REQUEST_FORMAT_HEADERprovider_native_responsePROVIDER_NATIVE_RESPONSE_KEYvaluereturnc r|dk(ry|dk(rytdtd|ddjtd)Nr rz Invalid `z`: z. Expected one of z, .) ValueErrorrjoinr)rs m/home/oem/CyberBrain26-unified/.venv/lib/python3.12/site-packages/litellm/llms/base_llm/ocr/transformation.pyparse_ocr_request_formatr$sL    ,-S 9KDIIViLjKkklm cNeZdZUdZdZedzed<dZedzed<dZedzed<y)OCRPageDimensionsz"Page dimensions from OCR response.Ndpiheightwidth) __name__ __module__ __qualname____doc__r int__annotations__r!r"rrrr.s/,CtFC$JE3:rrcLeZdZUdZdZedzed<dZeee fdzed<ddiZ y) OCRPageImagezImage extracted from OCR page.N image_base64bboxextraallow) r#r$r%r&r,strr(r-dictr model_configr)rrr+r+6s4(#L#*#"&D$sCx.4 &W%Lrr+c\eZdZUdZeed<eed<dZee dzed<dZ e dzed<ddiZ y) OCRPagezSingle page from OCR response.indexmarkdownNimages dimensionsr.r/) r#r$r%r&r'r(r0r7listr+r8rr2r)rrr4r4?s=( JM(,FD  %,+/J!D(/W%Lrr4cVeZdZUdZdZedzed<dZedzed<dZ edzed<ddiZ y) OCRUsageInfoz$Usage information from OCR response.Npages_processedcreditsdoc_size_bytesr.r/) r#r$r%r&r<r'r(r=floatr>r2r)rrr;r;Js9."&OS4Z& GUT\ !%NC$J%W%Lrr;ceZdZUdZeeed<eed<dZe dzed<dZ e dzed<dZ edzed<dZ eeeefdzed<dZeeeefdzed <d Zeed <d d iZeeZeed<deeefddfdZdeeefdzfdZy) OCRResponsez Standard OCR response format. Standardized to Mistral OCR format - other providers should transform to this format. pagesmodelNdocument_annotation usage_infocontenttables keyValuePairsocrobjectr.r/)default_factory_hidden_paramsnative_responserc*||jt<y)zFKeep the provider's own response payload alongside the normalized one.N)rLrselfrMs rset_provider_native_responsez(OCRResponse.set_provider_native_responsehsD 'rrWc `eZdZdZddZdedefdZdedzfdZde d e dede fd Z dd e ded edzd edzde dzde f dZ dd edzded e de dzdef dZ dede d e d e def dZdede d e d e def dZdedej$dedefdZdedej$dedefdZdeded e defdZy) BaseOCRConfigzc Base configuration for OCR transformations. Handles provider-agnostic OCR operations. rNcyNr)rPs r__init__zBaseOCRConfig.__init__s rrCcgS)z Get supported OCR parameters for this provider. Override this method in provider-specific implementations. r))rPrCs rget_supported_ocr_paramsz&BaseOCRConfig.get_supported_ocr_paramss  rcy)zY Return the provider-specific API key environment variable name, if any. Nr)r_s rget_api_key_env_varz!BaseOCRConfig.get_api_key_env_varsrnon_default_paramsoptional_paramsc|S)z3Map OCR parameters to provider-specific parameters.r))rPrerfrCs rmap_ocr_paramszBaseOCRConfig.map_ocr_paramss rheadersapi_keyapi_baselitellm_paramsc |S)zq Validate environment and return headers. Override in provider-specific implementations. r))rPrirCrjrkrlkwargss rvalidate_environmentz"BaseOCRConfig.validate_environments rc td)zk Get complete URL for OCR endpoint. Override in provider-specific implementations. z0get_complete_url must be implemented by providerNotImplementedError)rPrkrCrfrlrns rget_complete_urlzBaseOCRConfig.get_complete_urls""TUUrdocumentc td)a Transform OCR request to provider-specific format. Override in provider-specific implementations. Note: By the time this method is called, any file-type documents have already been converted to document_url/image_url format with base64 data URIs by the preprocessing in litellm/ocr/main.py. Args: model: Model name document: Document to process - always a dict with type="document_url" or type="image_url" optional_params: Optional parameters for the request headers: Request headers Returns: OCRRequestData with data and files fields z5transform_ocr_request must be implemented by providerrqrPrCrtrfrirns rtransform_ocr_requestz#BaseOCRConfig.transform_ocr_requests2""YZZrc8K|jd||||d|Sw)aU Async transform OCR request to provider-specific format. Optional method - providers can override if they need async transformations (e.g., Azure AI for URL-to-base64 conversion). Default implementation falls back to sync transform_ocr_request. Args: model: Model name document: Document to process (Mistral format dict, or file path, bytes, etc.) optional_params: Optional parameters for the request headers: Request headers Returns: OCRRequestData with data and files fields )rCrtrfrir))rwrvs rasync_transform_ocr_requestz)BaseOCRConfig.async_transform_ocr_requests82*t)) +     s raw_response logging_objc td)z Transform provider-specific OCR response to standard format. Override in provider-specific implementations. z6transform_ocr_response must be implemented by providerrqrPrCrzr{rns rtransform_ocr_responsez$BaseOCRConfig.transform_ocr_responses""Z[[rc6K|jd|||d|Sw)a Async transform provider-specific OCR response to standard format. Optional method - providers can override if they need async transformations (e.g., Azure Document Intelligence for async operation polling). Default implementation falls back to sync transform_ocr_response. Args: model: Model name raw_response: Raw HTTP response logging_obj: Logging object Returns: OCRResponse in standard format )rCrzr{r))r~r}s rasync_transform_ocr_responsez*BaseOCRConfig.async_transform_ocr_responses5.+t** %#    s error_message status_codect|||S)z-Get appropriate error class for the provider.)rmessagerir )rPrrris rget_error_classzBaseOCRConfig.get_error_classs #!  r)rN)NNNr^)r#r$r%r&r`r0r9rbrdr1rhrors DocumentTyperWrwryhttpxResponseLiteLLMLoggingObjrAr~rr' Exceptionrr)rrr\r\ys  cdS4Z     ##&*   t  *  t   ('+ V* V V V t V  V[[[ [  [ [6         B \ \nn \' \  \  nn '    <          rr\)%r&collections.abcrtypingrrrrrpydanticr)litellm.llms.base_llm.chat.transformationr litellm.types.llms.baser *litellm.litellm_core_utils.litellm_loggingr rr1r0rOCRRequestFormatrtupler(rrrrJrrr+r4r;rArWr\r)rrrs$55 F=W CH~ ./;PU5!13!678P".%.#151&@e@F/?1&,&&'&&,&N+N<(.(n n r