import asyncio
import os
import tempfile
import unittest
from unittest.mock import MagicMock, patch
from bs4 import BeautifulSoup
from olmocr.bench.tests import TestType
from olmocr.synth.mine_html_templates import render_pdf_with_playwright
class TestIncompleteContent(unittest.TestCase):
def test_pg39(self):
html = """
History of Computing
9/2/2010
HISTORY OF COMPUTING
Fall 2010 - CSE 111
Friday, September 3 & Wednesday, September 8
PRE 1642
3000 BC – Abacus is invented in Babylonia
800 AD - Chinese begin to use zero.
← Exo
← Span
← Rods
← Frame
From 800 to 1641 some kinda important stuff happens in math – algebra, calculus, geometry, etc.
"""
with tempfile.TemporaryDirectory() as td:
result = asyncio.run(render_pdf_with_playwright(html, os.path.join(td, "temp.pdf"), 768, 1024))
# Somehow we need to error out and show that this page didn't render fully
self.assertFalse(result.success)
if __name__ == "__main__":
unittest.main()