o 7i>!@sfddlZddlZddlZddlmZmZddlmZddlm Z ddl m Z dZ dZ Gdd d ZdS) N)DictOptional)Path) Anthropic)OpenAIz#https://api.deepinfra.com/v1/openaiz Qwen/Qwen3-Next-80B-A3B-Instructc@seZdZdZddedefddZdedefd d Zd edeee effd d Z d edeee effddZ ddededefddZ dS)TenderLocationExtractorz.Extract delivery location from GeM tender PDFsTapi_key use_anthropiccCs,||_|r t|d|_dSt|td|_dS)z Initialize with API key for LLM service Args: api_key: API key for OpenAI use_anthropic: If True, use Anthropic Claude; if False, use OpenAI )r)rbase_urlN)r rclientr OPENAI_URL)selfrr r>/var/www/html/minaions-tender/ai-engine/location_finder_gem.py__init__sz TenderLocationExtractor.__init__pdf_pathreturnc Csddl}|j|std||j|}|dkr"td|z.d}t|}|jD]}|}|r<||d7}q.Wdn1sGwY| WStyy}zt |} d| vsdd| vrntd |d | td | d}~ww) z Extract text from PDF file using pdfplumber Args: pdf_path: Path to the PDF file Returns: Extracted text as string rNz!PDF file does not exist at path: zPDF file is empty (0 bytes):  zNo /Root objectz not a PDFz%Invalid or corrupted PDF file (size: z bytes): z Error extracting text from PDF: ) ospathexists Exceptiongetsize pdfplumberopenpages extract_textstripstr) r rr file_sizetextpdfpage page_texte error_msgrrrextract_text_from_pdfs0      z-TenderLocationExtractor.extract_text_from_pdfr!c Cs d|d}|jr|jjjddd|dgd}|jdj}n!td t|jjj jtd d dd|dgd d id}|j dj j}td|zt dd|}t dd|}t|}|WStjy}ztd|td|ddddddWYd}~Sd}~ww)z Use LLM to extract location information from tender text Args: text: Extracted text from PDF Returns: Dictionary with city, state, and full_address uAnalyze the following GeM (Government e-Marketplace) tender document text and extract the delivery location information. Focus on finding: 1. The consignee address (परेषती/Consignee address) 2. Office location (कार्यालय का नाम/Office Name) 3. Any delivery address mentioned From this information, identify: - City name - State name (if city is mentioned but not state, infer the state from the city) - Full delivery address Tender Document Text: af Please respond in JSON format with the following structure: { "city": "city name or null", "state": "full state name or null", "full_address": "complete delivery address or null", "confidence": "high/medium/low" } Important: - For city names, map them to their respective states (e.g., Kanchipuram -> Tamil Nadu, Mumbai -> Maharashtra, Kadapa -> Andhra Pradesh) - Use full state names, not abbreviations (e.g., "Andhra Pradesh" not "AP") - Name of the state has to be one out of 28 states or 8 UTs. That is one out of: Andhra Pradesh, Arunachal Pradesh, Assam, Bihar, Chhattisgarh, Goa, Gujarat, Haryana, Himachal Pradesh, Jharkhand, Karnataka, Kerala, Madhya Pradesh, Maharashtra, Manipur, Meghalaya, Mizoram, Nagaland, Odisha, Punjab, Rajasthan, Sikkim, Tamil Nadu, Telangana, Tripura, Uttar Pradesh, Uttarakhand, and West Bengal. Andaman and Nicobar Islands, Chandigarh, Dadra and Nagar Haveli and Daman and Diu, Delhi, Jammu and Kashmir, Ladakh, Lakshadweep, and Puducherry. - If multiple addresses exist, prioritize the consignee/delivery address - Return null for fields you cannot confidently extractzclaude-sonnet-4-20250514iuser)rolecontent)model max_tokensmessagesrz Invoking LLM with OpenAI model:systemzdYou are a helpful assistant that extracts location (Indian State) information from tender documents.type json_object)r+r-response_formatz LLM response is:z ```json\s*rz```\s*$zError parsing LLM response: zRaw response: NlowzFailed to parse LLM response)citystate full_address confidenceerror)r r r-creater*r!print OPENAI_MODELchat completionschoicesmessageresubjsonloadsrJSONDecodeError)r r!promptresponse result_text location_datar%rrrextract_location_with_llmBsJ     z1TenderLocationExtractor.extract_location_with_llmcCspgd}d}t||tj}|r|dnd}d}|D]}tdt|d|tjr/|}nqd||dddS) z Fallback method using regex patterns to extract location Args: text: Extracted text from PDF Returns: Dictionary with extracted location information )zAndhra PradeshzArunachal PradeshAssamBihar ChhattisgarhGoaGujaratHaryanazHimachal Pradesh Jharkhand KarnatakaKeralazMadhya Pradesh MaharashtraManipur MeghalayaMizoramNagalandOdishaPunjab RajasthanSikkimz Tamil Nadu TelanganaTripuraz Uttar Pradesh Uttarakhandz West Bengalu=(?:पता|Address|परेषती)[:\s]*([^\n]{20,200})rNz\br2regex_fallback)r3r4r5r6method)r?findall IGNORECASEsearchescape)r r!statesaddress_patternaddress_matchesr5r4srrrextract_location_regex_fallbacks   z7TenderLocationExtractor.extract_location_regex_fallbackF use_fallbackc Cstd|||}tdt|dd}||}|dkr8|t|}|||d}td|dntd|d |rH||}n||}t|j|d |}|S) a  Main method to process tender PDF and extract location Args: pdf_path: Path to the PDF file use_fallback: If True, use regex fallback instead of LLM Returns: Dictionary with extracted information z Processing: z Extracted z characters from PDFuपपतताा/Addressiz+Trimmed text to 400 characters after last ''z' not found, using full text) file_name file_path)r9r'lenrfindrhrHrname) r rrir! search_stridx start_idx location_inforesultrrrprocess_tender_pdfs&     z*TenderLocationExtractor.process_tender_pdfN)T)F) __name__ __module__ __qualname____doc__rboolrr'rrrHrhrvrrrrrs#Q(r)rr?rAtypingrrpathlibr anthropicropenairr r:rrrrrs