Spaces:
Running
Running
| from PyPDF2 import PdfReader | |
| import openai | |
| import fitz # PyMuPDF | |
| import gradio as gr | |
| class AggressiveContentFinder: | |
| """ | |
| This class identifies and extracts aggressive terms in a contract document using OpenAI's GPT-3. | |
| """ | |
| def __init__(self): | |
| """ | |
| Initialize the AggressiveContentFinder with your OpenAI API key. | |
| """ | |
| # openai.api_key = openai_api_key | |
| pass | |
| def _extract_aggressive_content(self, contract_text: str) -> str: | |
| """ | |
| Use OpenAI's GPT-3 to identify aggressive terms in the given contract text. | |
| Args: | |
| contract_text (str): Text extracted from the contract. | |
| Returns: | |
| str: Identified aggressive terms. | |
| """ | |
| try: | |
| response = openai.Completion.create( | |
| engine="text-davinci-003", | |
| prompt=f"""This is a contract document content. Your task is to identify aggressive terms like warning terms, penalties in the given contract: | |
| (Example: "The bank may take possession of the property.") | |
| ```contract: {contract_text}``` | |
| """, | |
| max_tokens=70, | |
| temperature=0 | |
| ) | |
| aggressive_terms = response.choices[0].text.strip() | |
| return aggressive_terms | |
| except Exception as e: | |
| print(f"An error occurred during text analysis: {str(e)}") | |
| def get_aggressive_content(self, pdf_file_path: str): | |
| """ | |
| Extract text from a PDF document and identify aggressive terms. | |
| Args: | |
| pdf_file_path (str): Path to the PDF document. | |
| Returns: | |
| str: Identified aggressive terms in the contract document. | |
| This method opens a multi-page PDF using PdfReader and iterates through each page. For each page, it extracts | |
| the text and passes it to the _extract_aggressive_content method for further processing. The identified | |
| aggressive terms are concatenated and returned. If any errors occur during PDF processing, they are logged for | |
| debugging. | |
| """ | |
| try: | |
| # Open the multi-page PDF using PdfReader | |
| pdf = PdfReader(pdf_file_path.name) | |
| aggressive_terms = "" | |
| # Extract text from each page and pass it to the process_text function | |
| for page_number in range(len(pdf.pages)): | |
| # Extract text from the page | |
| page = pdf.pages[page_number] | |
| text = page.extract_text() | |
| # Pass the text to the process_text function for further processing | |
| aggressive_terms += self._extract_aggressive_content(text) | |
| return aggressive_terms | |
| except Exception as e: | |
| print(f"An error occurred while processing the PDF document: {str(e)}") | |
| def file_output_fnn(self,file_path): | |
| file_path = file_path.name | |
| return file_path | |
| def gradio_interface(self): | |
| with gr.Blocks(css="style.css",theme='xiaobaiyuan/theme_brief') as demo: | |
| with gr.Row(elem_id = "col-container",scale=0.80): | |
| # with gr.Column(elem_id = "col-container",scale=0.80): | |
| # file1 = gr.File(label="File",elem_classes="filenameshow") | |
| # with gr.Column(elem_id = "col-container",scale=0.20): | |
| # upload_button1 = gr.UploadButton( | |
| # "Browse File",file_types=[".txt", ".pdf", ".doc", ".docx",".json",".csv"], | |
| # elem_classes="uploadbutton") | |
| aggressive_content = gr.Button("Get Aggressive Content",elem_classes="uploadbutton") | |
| with gr.Row(elem_id = "col-container",scale=0.60): | |
| headings = gr.Textbox(label = "Aggressive Content") | |
| # upload_button1.upload(self.file_output_fnn,upload_button1,file1) | |
| aggressive_content.click(self.get_aggressive_content,[],headings) | |