AHj :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$de'dee'efdee'dee'deee(ejRfdee'deee'efdefdZ*e% d$de'dee'efdee'dee'deee(ejRfdee'deee'efdeeeeeefffdZ+ejXdZ-dddddddddd Z.d!e'de'fd"Z/dee'efdee'e'ffd#Z0y)%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 Kt} tj} d|d<|tj||\} }} } t t f|||||||d|} tj} t | j| } | jd| d{}tj|r |d{}n|}|td||S787#t$r}tj|||||d}~wwxYww)ar Async OCR function. Args: model: Model name (e.g., "mistral/mistral-ocr-latest") document: Document to process in Mistral format: {"type": "document_url", "document_url": "https://..."} for PDFs/docs, {"type": "image_url", "image_url": "https://..."} for images, or {"type": "file", "file": } for local files api_key: Optional API key api_base: Optional API base URL timeout: Optional timeout custom_llm_provider: Optional custom LLM provider extra_headers: Optional extra headers **kwargs: Additional parameters (e.g., include_image_base64, pages, image_limit) Returns: OCRResponse in Mistral OCR format with pages, model, usage_info, etc. Example: ```python import litellm # OCR with PDF response = await litellm.aocr( model="mistral/mistral-ocr-latest", document={ "type": "document_url", "document_url": "https://arxiv.org/pdf/2201.04234" }, include_image_base64=True ) # OCR with image response = await litellm.aocr( model="mistral/mistral-ocr-latest", document={ "type": "image_url", "image_url": "https://example.com/image.png" } ) # OCR with base64 encoded PDF response = await litellm.aocr( model="mistral/mistral-ocr-latest", document={ "type": "document_url", "document_url": f"data:application/pdf;base64,{base64_pdf}" } ) # OCR with local file response = await litellm.aocr( model="mistral/mistral-ocr-latest", document={"type": "file", "file": "/path/to/document.pdf"} ) ``` TaocrN)rr)rrrrrrrz2Got an unexpected None response from the OCR API: rroriginal_exceptioncompletion_kwargs extra_kwargs)localsasyncioget_event_looplitellmget_llm_providerrocr contextvars copy_contextrunrun_in_executor iscoroutine ValueError Exceptionexception_type)rrrrrrrkwargs local_varsloop_funcctxfunc_with_context init_responseresponsees H/root/tools/cai/cai_env/lib/python3.12/site-packages/litellm/ocr/main.pyrrs;JJ, %%'v  &+2+C+Ch, (A"Aq    3'    &&(#CGGT2"2249JKK   } -**H$H  DXJO L+  $$ 3 (    sM DBCCC;C<CDCC D"C<<DDc t} |jd} |jdd} |jdddu} t|tst dt ||jd} | d k(rt|}|jd} | d vrt d | d tj|||| \}}} }| r| }|r|}tj|tj|}|t d|tjd|d|tdi|}|j!|}i}|D]}||vs|j|||<|j#|i|}tjd|| j%|||| |d|t&j)||||xst*| |||| ||t | }|S#t,$r}tj.|||||d}~wwxYw)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_idNrFTz%convert_file_document_to_url_documentr%r&rget_provider_ocr_config LlmProvidersr debugrget_supported_ocr_paramsmap_ocr_paramsupdate_from_kwargsbase_llm_http_handlerr'r r.r/)rrrrrrrr0r1r<r= _is_asyncdoc_typedynamic_api_keydynamic_api_baseocr_provider_configrEsupported_paramsrCparamrDr8r9s r:r'r'sRJr 17    &56I5JK   |4G3H I .77/GGeGT% >E,2JJu,="5) > .<<1=  B?BSTU ..+#2$!4 / ),,+.+ 3!//-    $$ 3 (    sEG"BG"" H +HH z^[\w.+-]+/[\w.+-]+$zapplication/pdfz image/pngz image/jpegz image/gifz image/webpz image/tiffz image/bmp) z.pdfz.pngz.jpgz.jpegz.gifz.webpz.tiffz.tifz.bmp file_pathctjj|dj}tj |}|r|St j|\}}|xsdS)z Determine MIME type from file path extension. Falls back to mimetypes.guess_type, then to 'application/octet-stream'. application/octet-stream)ospathsplitextlower_MIME_TYPE_MAPrK mimetypes guess_type)r]extmimeguessedr3s r: get_mime_typerkfs^ ''  9 %a ( . . 0C   c "D  %%i0JGQ  000c x|jd}| tdd}d}t|ttfrt|}t j j|std|t|}t j j|}t|d5}|j}dddnt|tr|}nt|ts t|drXt|drt!|dd}|r t|}|j}t|tr*|j#d }ntd t%|d s td d |vr|d }t&j)|std|t+j,|j/d }d|d|}|j1dr-t3j4d|dt7|d|dd|dSt3j4d|dt7|d|dd|dS#1swYxYw)a Convert a file-type document dict to a document_url-type document dict with an inline base64 data URI. Accepts document dicts like: {"type": "file", "file": "/path/to/document.pdf"} # file path string {"type": "file", "file": Path("/path/to/doc.pdf")} # pathlib.Path {"type": "file", "file": } # file-like object (BinaryIO) {"type": "file", "file": b"raw bytes"} # raw bytes Returns: {"type": "document_url", "document_url": "data:;base64,"} or {"type": "image_url", "image_url": "data:;base64,"} r?Nz|document with type='file' must include a 'file' field containing a file path (str), pathlib.Path, file-like object, or bytesr`zFile not found: rbreadnamezutf-8zUnsupported file input type: zG. Expected str (file path), pathlib.Path, bytes, or a file-like object.z"File is empty or could not be read mime_typezInvalid MIME type: zdata:z;base64,zimage/z;OCR file input: Converted file to image_url data URI (mime=z, size=z bytes, name=)rA)r>rAz>OCR file input: Converted file to document_url data URI (mime=r@)r>r@)rKr-rLstrrrarbisfileFileNotFoundErrorrkbasenameopenrobytesrhasattrgetattrencoder> _MIME_PATTERNmatchbase64 b64encodedecode startswithr rQlen) r file_inputrq file_namer]f file_bytes base64_datadata_uris r:rNrNtsOf%J J  0I#I*sDk* O ww~~i(#&6yk$BC C!), GG$$Y/ )T " "aJ " " J & J '7:v+F :v & FD9I))4 __& j# &#**73J+D,<+=>T T  =>>h[)   y ).yk:;;"":.55g>Kyk+7HH%Kws:&7}YKq R $(;;Kws:&7}YKq R 'AAS " "s 'H00H9)NNNNN)1__doc__r#r~r(rfrare functoolsriorpathlibrtypingrrrr r httpxr%litellm._loggingr litellm.constantsr *litellm.litellm_core_utils.litellm_loggingr LiteLLMLoggingObj(litellm.llms.base_llm.ocr.transformationrr*litellm.llms.custom_httpx.llm_http_handlerrlitellm.types.routerr litellm.utilsrrrUrsfloatTimeoutrr'compiler|rerkrNrIrlr:rs=  88 +-SOI57+,""59)-.2q q 38nq c]q sm q eE5==01 2 q "# q DcN+q q q h""59)-.2{ { 38n{ c]{ sm { eE5==01 2 { "# { DcN+{  ; #sK"78 89{ { D 12            1S 1S 1IBDcNIBtCQTH~IBrl