pdf_parser.py

import os
import shutil
import fitz

TEMP_DIR = os.path.join(os.path.dirname(__file__), "temp")


def extract_text_from_pdf(pdf_path):
    doc = fitz.open(pdf_path)
    text = ""

    for page in doc:
        text += page.get_text()

    doc.close()
    return text


def process_uploaded_pdfs(file_inputs):
    """
    Saves uploaded PDF files to temp/, extracts text,
    combines content, and cleans up temp files.
    """
    os.makedirs(TEMP_DIR, exist_ok=True)

    if file_inputs is None:
        return ""

    if isinstance(file_inputs, (str, bytes)):
        file_inputs = [file_inputs]

    combined_text = ""

    for file_item in file_inputs:
        file_path = getattr(file_item, "name", file_item)
        if not file_path or not os.path.exists(file_path):
            continue

        filename = os.path.basename(file_path)
        temp_file_path = os.path.join(TEMP_DIR, filename)

        # 1. Save / Copy to temp/ directory
        shutil.copy(file_path, temp_file_path)

        try:
            # 2. Extract text from temp PDF
            extracted = extract_text_from_pdf(temp_file_path)
            combined_text += f"\n--- Document: {filename} ---\n{extracted}\n"
        finally:
            # 3. Delete temporary file when session / extraction ends
            if os.path.exists(temp_file_path):
                os.remove(temp_file_path)

    return combined_text