Phân tích PDF để trích xuất Bảng trong Python SDK

import shutil
import json
import logging
from pathlib import Path
from asposepdfcloud import ApiClient, PdfApi
import logging

# Configure logging
logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")


class ParseExtractTables:
    """Class for extracting tables from PDF document page using Aspose PDF Cloud API."""
    def __init__(self):
        self.pdf_api = PdfApi(ApiClient(APP_KEY, APP_SID)

    def upload_document(self, documentName: str, localFolder: str, remoteFolder: str):
        """Upload a PDF document to the Aspose Cloud server."""
        if self.pdf_api:
            file_path = localFolder / documentName
            try:
                if remoteFolder == None:
                    self.pdf_api.upload_file(documentName, str(file_path))
                else:
                    opts = { "folder": remoteFolder }
                    self.pdf_api.upload_file(remoteFolder + '/' + documentName, file_path)
                logging.info(f"File {documentName} uploaded successfully.")
            except Exception as e:
                logging.error(f"Failed to upload file: {e}")

   def extract(self, documentName: str, localFolder: Path, remoteFolder: Path):
        self.upload_document(documentName, remoteFolder)

        opts = {
            "folder": remoteFolder
        }
        respTables = self.pdf_api.get_document_tables(documentName, **opts)
        if respTables.code != 200:
            logging.error("GetTables(): Unexpected error!")
        else:
            localJsonFile = str(Path.joinpath(localFolder, "tables_objects.json"))
            with open(localJsonFile, "w", encoding="utf-8") as localFile:
                localFile.write("[\n")

                for tab in respTables.tables.list:
                    logging.info(f"GetTabels(): Table '{tab.id}' successfully extracted from the document '{documentName}'.")
                    jsText = json.dumps(tab.__dict__, ensure_ascii=False, default=str, indent=4) + ",\n\n"
                    localFile.write(jsText)
                
                localFile.write("]")