Handle simple obfuscations - HomeDepot.com style price obfuscation (#764)

pull/767/head
dgtlmoon 2 years ago
parent a95468be08
commit dd0eeca056

@ -150,7 +150,9 @@ class perform_site_check():
is_html = False is_html = False
if is_html or is_source: if is_html or is_source:
# CSS Filter, extract the HTML that matches and feed that into the existing inscriptis::get_text # CSS Filter, extract the HTML that matches and feed that into the existing inscriptis::get_text
fetcher.content = html_tools.workarounds_for_obfuscations(fetcher.content)
html_content = fetcher.content html_content = fetcher.content
# If not JSON, and if it's not text/plain.. # If not JSON, and if it's not text/plain..

@ -202,3 +202,17 @@ def html_to_text(html_content: str, render_anchor_tag_content=False) -> str:
return text_content return text_content
def workarounds_for_obfuscations(content):
"""
Some sites are using sneaky tactics to make prices and other information un-renderable by Inscriptis
This could go into its own Pip package in the future, for faster updates
"""
# HomeDepot.com style <span>$<!-- -->90<!-- -->.<!-- -->74</span>
# https://github.com/weblyzard/inscriptis/issues/45
if not content:
return content
content = re.sub('<!--\s+-->', '', content)
return content

File diff suppressed because it is too large Load Diff

@ -0,0 +1,43 @@
#!/usr/bin/python3
import time
from flask import url_for
from .util import live_server_setup
def set_original_ignore_response():
test_return_data = """<html>
<body>
<span>The price is</span><span>$<!-- -->90<!-- -->.<!-- -->74</span>
</body>
</html>
"""
with open("test-datastore/endpoint-content.txt", "w") as f:
f.write(test_return_data)
def test_obfuscations(client, live_server):
set_original_ignore_response()
live_server_setup(live_server)
time.sleep(1)
# Add our URL to the import page
test_url = url_for('test_endpoint', _external=True)
res = client.post(
url_for("import_page"),
data={"urls": test_url},
follow_redirects=True
)
assert b"1 Imported" in res.data
# Give the thread time to pick it up
time.sleep(3)
# Check HTML conversion detected and workd
res = client.get(
url_for("preview_page", uuid="first"),
follow_redirects=True
)
assert b'$90.74' in res.data
Loading…
Cancel
Save