processor.py

import layoutparser as lp
import pytesseract
from pdf2image import convert_from_path
from PIL import Image

def load_images(uploaded_file):
    if uploaded_file.name.endswith(".pdf"):
        return convert_from_path(uploaded_file)
    else:
        return [Image.open(uploaded_file)]
    
from processor import (
    load_images,
    analyze_layout,
    extract_text_from_blocks,
    rule_based_kv_extraction
)

def convert_pdf_to_images(pdf_path):
    return convert_from_path(pdf_path)

def analyze_layout(image):
    model = lp.EfficientDetLayoutModel(
        "lp://efficientdet/PubLayNet",
        extra_config=["MODEL.ROI_HEADS.SCORE_THRESH_TEST", 0.6],
        label_map={0: "Text", 1: "Title", 2: "List", 3: "Table", 4: "Figure"}
    )
    return model.detect(image)