service/reader_v3.py

import os
import nest_asyncio

from llama_parse import LlamaParse
from dotenv import load_dotenv
from fastapi import UploadFile
from fastapi.responses import JSONResponse

from script.get_metadata import Metadata

load_dotenv()
nest_asyncio.apply()


def parse_journal(content: bytes, file_name: str):
    """Parse the journal using LlamaParse."""
    try:
        # Initialize the parser
        parser = LlamaParse(
            api_key=os.getenv("LLAMA_PARSE_API_KEY"),
            result_type="markdown",
            # use_vendor_multimodal_model=True,
            # vendor_multimodal_model_name="openai-gpt-4o-mini",
        )

        # Load and process the document
        llama_parse_documents = parser.load_data(
            content, extra_info={"file_name": file_name}
        )

        return llama_parse_documents

    except Exception as e:
        return JSONResponse(status_code=400, content=f"Error processing file: {e}")

async def upload_file(reference, file: UploadFile):
    try:
        # Read the binary content of the uploaded file once
        content = await file.read()
        # Parse the journal
        parsed_documents = parse_journal(content, file.filename)
        # Extract metadata
        # metadata_dict = await extract_metadata(content)
        # print("Metadata Dictionary : \n\n", metadata_dict)

        metadata_gen = Metadata(reference)
        documents_with_metadata = metadata_gen.apply_metadata(parsed_documents)
        
        # document_with_metadata = 

        print("Document with Metadata : \n\n", documents_with_metadata)
        print("Banyak documents : \n", len(documents_with_metadata))

        # Return both parsed documents and metadata
        return documents_with_metadata

    except Exception as e:
        return JSONResponse(status_code=500, content=f"Error processing file: {e}")