o ưi9@s"dZddlZddlZddlZddlZddlZddlZddlmZddl m Z ddl m Z ddl mZmZmZmZmZddlZddlZddlmZddlmZdd lmZdd lmZmZdd lm Z dd l!m"Z"dd l#m$Z$m%Z%e Z&e%     d(de'dee'efdee'dee'deee(ej)fdee'deee'efdefddZ*e%     d(de'dee'efdee'dee'deee(ej)fdee'deee'efdeeeeeefffddZ+e,dZ-ddddddd d d!d" Z.d#e'de'fd$d%Z/dee'efdee'e'ffd&d'Z0dS))z Main OCR function for LiteLLM. N)partial)IOBase)Path)Any CoroutineDictOptionalUnion)verbose_logger)request_timeout)Logging) BaseOCRConfig OCRResponse)BaseLLMHTTPHandler)GenericLiteLLMParams)ProviderConfigManagerclientmodeldocumentapi_keyapi_basetimeoutcustom_llm_provider extra_headersreturnc st}zUt} d|d<|durtj||d\} }} } ttf|||||||d|} t} t| j | } | d| IdH}t |rJ|IdH}n|}|durWt d||WSt yo}z tj|||||dd}~ww)ar Async OCR function. Args: model: Model name (e.g., "mistral/mistral-ocr-latest") document: Document to process in Mistral format: {"type": "document_url", "document_url": "https://..."} for PDFs/docs, {"type": "image_url", "image_url": "https://..."} for images, or {"type": "file", "file": } for local files api_key: Optional API key api_base: Optional API base URL timeout: Optional timeout custom_llm_provider: Optional custom LLM provider extra_headers: Optional extra headers **kwargs: Additional parameters (e.g., include_image_base64, pages, image_limit) Returns: OCRResponse in Mistral OCR format with pages, model, usage_info, etc. Example: ```python import litellm # OCR with PDF response = await litellm.aocr( model="mistral/mistral-ocr-latest", document={ "type": "document_url", "document_url": "https://arxiv.org/pdf/2201.04234" }, include_image_base64=True ) # OCR with image response = await litellm.aocr( model="mistral/mistral-ocr-latest", document={ "type": "image_url", "image_url": "https://example.com/image.png" } ) # OCR with base64 encoded PDF response = await litellm.aocr( model="mistral/mistral-ocr-latest", document={ "type": "document_url", "document_url": f"data:application/pdf;base64,{base64_pdf}" } ) # OCR with local file response = await litellm.aocr( model="mistral/mistral-ocr-latest", document={"type": "file", "file": "/path/to/document.pdf"} ) ``` TaocrN)rr)rrrrrrrz2Got an unexpected None response from the OCR API: rrZoriginal_exceptionZcompletion_kwargs extra_kwargs)localsasyncioget_event_looplitellmget_llm_providerrocr contextvars copy_contextrunrun_in_executor iscoroutine ValueError Exceptionexception_type)rrrrrrrkwargs local_varsloop_funcctxZfunc_with_contextZ init_responseresponseer4G/home/app/Keep/.python/lib/python3.10/site-packages/litellm/ocr/main.pyrsVE    rcKst}z|d} |dd} |dddu} t|ts%tdt||d} | d kr7t|}|d} | d vrCtd | d tj ||||d \}}} }| rT| }|rX|}t j |t |d}|durmtd|t d|d|tdi|}|j|d}i}|D] }||vr||||<q|j|i|d}t d|| j||| |d|dtj||||pt| |||| ||t|d }|WSty}z tj|||||dd}~ww)a Synchronous OCR function. Args: model: Model name (e.g., "mistral/mistral-ocr-latest") document: Document to process in Mistral format: {"type": "document_url", "document_url": "https://..."} for PDFs/docs, {"type": "image_url", "image_url": "https://..."} for images, or {"type": "file", "file": } for local files api_key: Optional API key api_base: Optional API base URL timeout: Optional timeout custom_llm_provider: Optional custom LLM provider extra_headers: Optional extra headers **kwargs: Additional parameters (e.g., include_image_base64, pages, image_limit) Returns: OCRResponse in Mistral OCR format with pages, model, usage_info, etc. Example: ```python import litellm # OCR with PDF response = litellm.ocr( model="mistral/mistral-ocr-latest", document={ "type": "document_url", "document_url": "https://arxiv.org/pdf/2201.04234" }, include_image_base64=True ) # OCR with image response = litellm.ocr( model="mistral/mistral-ocr-latest", document={ "type": "image_url", "image_url": "https://example.com/image.png" } ) # OCR with base64 encoded PDF response = litellm.ocr( model="mistral/mistral-ocr-latest", document={ "type": "document_url", "document_url": f"data:application/pdf;base64,{base64_pdf}" } ) # OCR with local file response = litellm.ocr( model="mistral/mistral-ocr-latest", document={"type": "file", "file": "/path/to/document.pdf"} ) # Access pages for page in response.pages: print(f"Page {page.index}: {page.markdown}") ``` litellm_logging_objlitellm_call_idNrFTzlitellm_paramsr) rrr>rZ logging_objrrrrheadersZprovider_configr?rr4)rpopget isinstancedictr)r8%convert_file_document_to_url_documentr!r"rZget_provider_ocr_configZ LlmProvidersr debugrZget_supported_ocr_paramsZmap_ocr_paramsZupdate_environment_variablesbase_llm_http_handlerr#r r*r+)rrrrrrrr,r-r6r7Z _is_asyncZdoc_typeZdynamic_api_keyZdynamic_api_baseZocr_provider_configr?Zsupported_paramsr=paramr>r2r3r4r4r5r#sI           r#z^[\w.+-]+/[\w.+-]+$zapplication/pdfz image/pngz image/jpegz image/gifz image/webpz image/tiffz image/bmp) z.pdfz.pngz.jpgz.jpegz.gifz.webpz.tiffz.tifz.bmp file_pathcCs<tj|d}t|}|r|St|\}}|pdS)z Determine MIME type from file path extension. Falls back to mimetypes.guess_type, then to 'application/octet-stream'. application/octet-stream)ospathsplitextlower_MIME_TYPE_MAPrB mimetypes guess_type)rIextmimeZguessedr/r4r4r5 get_mime_typees  rUc Cs|d}|dur tdd}d}t|ttfrMt|}tj|s)td|t |}tj |}t |d }| }Wdn1sGwYn} # file-like object (BinaryIO) {"type": "file", "file": b"raw bytes"} # raw bytes Returns: {"type": "document_url", "document_url": "data:;base64,"} or {"type": "image_url", "image_url": "data:;base64,"} r9Nz|document with type='file' must include a 'file' field containing a file path (str), pathlib.Path, file-like object, or bytesrKzFile not found: rbreadnamezutf-8zUnsupported file input type: zG. Expected str (file path), pathlib.Path, bytes, or a file-like object.z"File is empty or could not be read mime_typezInvalid MIME type: zdata:z;base64,zimage/z;OCR file input: Converted file to image_url data URI (mime=z, size=z bytes, name=)r;)r8r;z>OCR file input: Converted file to document_url data URI (mime=r:)r8r:)rBr)rCstrrrLrMisfileFileNotFoundErrorrUbasenameopenrWbytesrhasattrgetattrencoder8 _MIME_PATTERNmatchbase64 b64encodedecode startswithr rFlen) rZ file_inputrY file_namerIfZ file_bytesZ base64_dataZdata_urir4r4r5rEssx              rE)NNNNN)1__doc__rrfr$rQrLre functoolsriorpathlibrtypingrrrrr Zhttpxr!Zlitellm._loggingr Zlitellm.constantsr Z*litellm.litellm_core_utils.litellm_loggingr ZLiteLLMLoggingObjZ(litellm.llms.base_llm.ocr.transformationr rZ*litellm.llms.custom_httpx.llm_http_handlerrZlitellm.types.routerrZ litellm.utilsrrrGr[floatTimeoutrr#compilerdrPrUrEr4r4r4r5s          t   B &