More pdf stats
No One
Actions

Authored By

	akosiaris
	Nov 10 2020, 1:19 PM

Size

1 KB

Referenced Files

None

Subscribers

None

More pdf stats
View Options

	"""
	Just get some status for PDF rendering
	"""
	import io
	import pprint
	import random
	import requests
	from pdfminer.high_level import extract_text

	stats = {
	'200s': 0,
	'4xx': 0,
	'5xx': 0,
	'valid': 0,
	'invalid': 0,
	'cl-matches-bytes': 0,
	'cl-fails-bytes': 0,
	}

	WIKIS = ['en', 'es', 'el', 'it', 'fr', 'simple', 'de', 'ar', 'bg', 'no', 'tr']

	for i in range(100):
	wiki = random.choice(WIKIS)
	r = requests.get('https://{}.wikipedia.org/api/rest_v1/page/random/title'.format(wiki))
	title = r.json()['items'][0]['title']
	print('Random page title: {}'.format(title))
	pdf = requests.get('https://en.wikipedia.org/api/rest_v1/page/pdf/{}'.format(title))
	code = pdf.status_code
	if code == 200:
	stats['200s'] += 1
	elif 400 <= code < 500:
	stats['4xx'] += 1
	if 500 <= code < 600:
	stats['5xx'] += 1
	if int(pdf.headers['content-length']) == len(pdf.content):
	stats['cl-matches-bytes'] += 1
	else:
	stats['cl-fails-bytes'] += 1
	f = io.BytesIO(pdf.content)
	try:
	extract_text(f)
	stats['valid'] += 1
	except :
	stats['invalid'] += 1

	pprint.pprint(stats)