Spaces:
Sleeping
Sleeping
| import os | |
| import boto3 | |
| import tempfile | |
| import fitz | |
| from io import BytesIO | |
| from fastapi import HTTPException | |
| class Loader: | |
| def __init__(self): | |
| # Create S3 and Transcribe clients with credentials | |
| self.bucket_name = "multimedika" | |
| self.s3_client = boto3.client( | |
| "s3", | |
| aws_access_key_id=os.getenv("AWS_ACCESS_KEY_ID"), | |
| aws_secret_access_key=os.getenv("AWS_SECRET_ACCESS_KEY"), | |
| region_name="us-west-2", | |
| ) | |
| # def upload_to_s3(self, file, object_name, folder_name="summarizer"): | |
| # try: | |
| # # If folder_name is provided, prepend it to the object_name | |
| # if folder_name: | |
| # object_name = f"{folder_name}/{object_name}" | |
| # # Create an in-memory file-like object | |
| # with BytesIO() as file_stream: | |
| # # Write the contents of the uploaded file to the stream | |
| # file_stream.write(file.file.read()) | |
| # file_stream.seek(0) # Move to the beginning of the stream | |
| # # Upload file to S3 | |
| # self.s3_client.upload_fileobj(file_stream, self.bucket_name, object_name) | |
| # print(f"File '{object_name}' successfully uploaded to bucket '{self.bucket_name}'.") | |
| # except Exception as e: | |
| # raise HTTPException(status_code=400, detail=f"Error uploading to AWS: {e}") | |
| def upload_to_s3(self, file, object_name, folder_name="summarizer"): | |
| try: | |
| # If folder_name is provided, prepend it to the object_name | |
| if folder_name: | |
| object_name = f"{folder_name}/{object_name}" | |
| # Open the PDF with PyMuPDF (fitz) | |
| pdf_document = fitz.open(stream=file.file.read(), filetype="pdf") | |
| # Loop through each page of the PDF | |
| for page_num in range(pdf_document.page_count): | |
| # Convert the page to bytes (as a separate PDF) | |
| page_stream = BytesIO() | |
| single_page_pdf = fitz.open() # Create a new PDF | |
| single_page_pdf.insert_pdf(pdf_document, from_page=page_num, to_page=page_num) | |
| single_page_pdf.save(page_stream) | |
| single_page_pdf.close() | |
| # Reset the stream position to the start | |
| page_stream.seek(0) | |
| # Define the object name for each page (e.g., 'summarizer/object_name/page_1.pdf') | |
| page_object_name = f"{object_name}/{page_num + 1}.pdf" | |
| # Upload each page to S3 | |
| self.s3_client.upload_fileobj(page_stream, self.bucket_name, page_object_name) | |
| print(f"Page {page_num + 1} of '{object_name}' successfully uploaded as '{page_object_name}' to bucket '{self.bucket_name}'.") | |
| except Exception as e: | |
| raise HTTPException(status_code=400, detail=f"Error uploading to AWS: {e}") | |
| def get_file_aws(self, object_name, local_file_name=None): | |
| """Downloads a PDF file from S3 and reads it using PyMuPDF.""" | |
| if local_file_name is None: | |
| local_file_name = "downloaded_pdf_file.pdf" # Default file name | |
| try: | |
| # Create a temporary directory to store the file | |
| temp_dir = tempfile.mkdtemp() | |
| file_path = os.path.join(temp_dir, local_file_name) | |
| # Download the file from S3 | |
| with open(file_path, "wb") as temp_file: | |
| self.s3_client.download_fileobj( | |
| self.bucket_name, object_name, temp_file | |
| ) | |
| # Open and read the PDF using PyMuPDF | |
| doc = fitz.open(file_path) | |
| # Example: Print the number of pages | |
| print(f"Number of pages: {doc.page_count}") | |
| # Do something with the PDF, like read text | |
| for page in doc: | |
| print(page.get_text()) | |
| # Close the document | |
| doc.close() | |
| # Clean up the downloaded file if needed | |
| os.remove(file_path) | |
| except Exception as e: | |
| raise HTTPException(status_code=400, detail=f"Error get file file in aws: {e}") | |