gist.py

import json
from reportlab.pdfgen import canvas
from reportlab.lib.pagesizes import letter

# Load the OCR results from the JSON file
with open('results.json') as f:
    ocr_results = json.load(f)

# Create a new PDF document
pdf_filename = 'output.pdf'
c = canvas.Canvas(pdf_filename, pagesize=letter)

# Iterate over each page in the OCR results
for filename, pages in ocr_results.items():
    for page_data in pages:
        # Create a new page in the PDF
        c.setFont("Helvetica", 12)
        
        # Iterate over the bounding boxes and draw the text
        for bbox_data in page_data['bboxes']:
            bbox = bbox_data['bbox']
            text = bbox_data['text']
            x, y = bbox[0], bbox[1]
            c.drawString(x, y, text)
        
        # Optionally, draw vertical and horizontal lines
        for line_data in page_data['vertical_lines']:
            line = line_data['bbox']
            c.line(line[0], line[1], line[0], line[3])
        
        for line_data in page_data['horizontal_lines']:
            line = line_data['bbox']
            c.line(line[0], line[1], line[2], line[1])
        
        # Save the page and start a new one
        c.showPage()

# Save the PDF document
c.save()