blob: 3bf94e2dcdd93e53fba71434a0c57740516ab1b5 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
|
import datetime
import pytest
from sandcrawler.ingest_html import *
def test_html_extract_ojs3() -> None:
with open('tests/files/first_monday_ojs3_fulltext.html', 'rb') as f:
ojs3_html = f.read()
fulltext = html_extract_body_teixml(ojs3_html)
assert fulltext['status'] == 'success'
|
