a bg@sUddlZddlZddlZddlmZddlmZmZmZddl Z ddl Z ddl m Z ddl mZddlmZddlmZddlmZerdd lmZeeZGd d d eZGd d d eZGdddeZdZeed<GdddeZ dS)N)abstractmethod) TYPE_CHECKINGIterableIterator)Document) BaseChatModel) HumanMessage)BaseBlobParser)BlobImagec@s8eZdZdZededddZeee dddZ d S) BaseImageBlobParserz6Abstract base class for parsing image blobs into text.r imgreturncCsdS)zAbstract method to analyze an image and extract textual content. Args: img: The image to be analyzed. Returns: The extracted text content. N)selfrrr/var/www/html/cobodadashboardai.evdpl.com/venv/lib/python3.9/site-packages/langchain_community/document_loaders/parsers/images.py_analyze_imagesz"BaseImageBlobParser._analyze_image)blobrccszddlm}|r}|jdkr4|t|}n ||}||}t d| ddt |i|j d|jidVWd n1s0YWntytd Yn0d S) zLazily parse a blob and yields Documents containing the parsed content. Args: blob (Blob): The blob to be parsed. Yields: Document: A document containing the parsed content and metadata. rr zapplication/x-npyzImage text: %s z\nsource)Z page_contentmetadataNzG`Pillow` package not found, please install it with `pip install Pillow`)ZPILr Z as_bytes_iomimetypeZ fromarraynumpyloadopenrloggerdebugreplacerrr ImportError)rrZImgbufrcontentrrr lazy_parse$s      * zBaseImageBlobParser.lazy_parseN) __name__ __module__ __qualname____doc__rstrrr rrr#rrrrr s r cs6eZdZdZddfdd Zdeddd ZZS) RapidOCRBlobParserzParser for extracting text from images using the RapidOCR library. Attributes: ocr: The RapidOCR instance for performing OCR. N)rcstd|_dS)z5 Initializes the RapidOCRBlobParser. N)super__init__ocr)r __class__rrr+Ks zRapidOCRBlobParser.__init__r rcCsr|js:zddlm}||_Wnty8tdYn0|t|\}}d}|rnddd|D}|S)z Analyzes an image and extracts text using RapidOCR. Args: img (Image): The image to be analyzed. Returns: str: The extracted text content. r)RapidOCRzc`rapidocr-onnxruntime` package not found, please install it with `pip install rapidocr-onnxruntime`rcSsg|] }|dqS)r).0textrrr mz5RapidOCRBlobParser._analyze_image..)r,Zrapidocr_onnxruntimer/r nparrayjoinstrip)rrr/Z ocr_result_r"rrrrTs     z!RapidOCRBlobParser._analyze_image)r$r%r&r'r+r(r __classcell__rrr-rr)Cs  r)cs@eZdZdZddeedfddZdeddd ZZS) TesseractBlobParserzFParse for extracting text from images using the Tesseract OCR library.)Zeng)langscstt||_dS)zInitialize the TesseractBlobParser. Args: langs (list[str]): The languages to use for OCR. N)r*r+listr=)rr=r-rrr+ts zTesseractBlobParser.__init__r rcCsBz ddl}Wnty&tdYn0|j|d|jdS)zAnalyze an image and extracts text using Tesseract OCR. Args: img: The image to be analyzed. Returns: str: The extracted text content. rNzQ`pytesseract` package not found, please install it with `pip install pytesseract`+)lang) pytesseractr Zimage_to_stringr8r=r9)rrrArrrrs    z"TesseractBlobParser._analyze_image) r$r%r&r'rr(r+rr;rrr-rr<qs r<aYou are an assistant tasked with summarizing images for retrieval. 1. These summaries will be embedded and used to retrieve the raw image. Give a concise summary of the image that is well optimized for retrieval 2. extract all the text from the image. Do not exclude any content from the page. Format answer in markdown without explanatory text and without markdown delimiter ``` at the beginning. _PROMPT_IMAGES_TO_DESCRIPTIONcs>eZdZdZedeedfddZdeddd ZZ S) LLMImageBlobParserzParser for analyzing images using a language model (LLM). Attributes: model (BaseChatModel): The language model to use for analysis. prompt (str): The prompt to provide to the language model. )prompt)modelrDcst||_||_dS)zInitializes the LLMImageBlobParser. Args: model (BaseChatModel): The language model to use for analysis. prompt (str): The prompt to provide to the language model. N)r*r+rErD)rrErDr-rrr+s zLLMImageBlobParser.__init__r rcCszt}|j|ddt|d}|jt d|j j t ddddd|id gd g}|j }t |tsvJ|S) zAnalyze an image using the provided language model. Args: img: The image to be analyzed. Returns: The extracted textual content. ZPNG)formatzutf-8r3)typer3 image_urlurlzdata:image/jpeg;base64,)rGrH)r")ioBytesIOsavebase64 b64encodegetvaluedecoderEZinvokerrDrFr" isinstancer()rrZ image_bytesZ img_base64msgresultrrrrs&   z!LLMImageBlobParser._analyze_image) r$r%r&r'rBrr(r+rr;rrr-rrCs rC)!rMrJloggingabcrtypingrrrrr6Zlangchain_core.documentsrZlangchain_core.language_modelsrZlangchain_core.messagesrZ)langchain_community.document_loaders.baser Z1langchain_community.document_loaders.blob_loadersr Z PIL.Imager getLoggerr$rr r)r<rBr(__annotations__rCrrrrs(         -.%