2021-02-16 20:35:28 +00:00
|
|
|
#!/usr/bin/python3
|
|
|
|
|
|
|
|
|
|
import time
|
2021-02-21 12:41:00 +00:00
|
|
|
from flask import url_for
|
|
|
|
|
from urllib.request import urlopen
|
2023-01-29 21:36:55 +00:00
|
|
|
from .util import set_original_response, set_modified_response, live_server_setup, wait_for_all_checks, extract_rss_token_from_UI
|
2021-02-21 12:41:00 +00:00
|
|
|
|
2021-02-27 08:05:25 +00:00
|
|
|
sleep_time_for_fetch_thread = 3
|
|
|
|
|
|
2022-06-02 08:01:03 +00:00
|
|
|
|
2022-03-01 16:50:15 +00:00
|
|
|
# Basic test to check inscriptus is not adding return line chars, basically works etc
|
|
|
|
|
def test_inscriptus():
|
|
|
|
|
from inscriptis import get_text
|
2023-03-12 16:05:34 +00:00
|
|
|
html_content = "<html><body>test!<br>ok man</body></html>"
|
2022-03-01 16:50:15 +00:00
|
|
|
stripped_text_from_html = get_text(html_content)
|
|
|
|
|
assert stripped_text_from_html == 'test!\nok man'
|
|
|
|
|
|
2021-02-27 08:05:25 +00:00
|
|
|
|
2021-02-21 13:05:52 +00:00
|
|
|
def test_check_basic_change_detection_functionality(client, live_server):
|
2021-02-21 12:41:00 +00:00
|
|
|
set_original_response()
|
2021-05-08 01:29:41 +00:00
|
|
|
live_server_setup(live_server)
|
2021-02-21 12:41:00 +00:00
|
|
|
|
|
|
|
|
# Add our URL to the import page
|
|
|
|
|
res = client.post(
|
|
|
|
|
url_for("import_page"),
|
|
|
|
|
data={"urls": url_for('test_endpoint', _external=True)},
|
|
|
|
|
follow_redirects=True
|
|
|
|
|
)
|
2022-03-21 21:54:27 +00:00
|
|
|
|
2021-02-21 12:41:00 +00:00
|
|
|
assert b"1 Imported" in res.data
|
|
|
|
|
|
2024-02-10 23:09:12 +00:00
|
|
|
wait_for_all_checks(client)
|
2021-02-21 20:08:04 +00:00
|
|
|
|
2021-02-21 19:46:56 +00:00
|
|
|
# Do this a few times.. ensures we dont accidently set the status
|
|
|
|
|
for n in range(3):
|
2022-05-20 14:27:51 +00:00
|
|
|
client.get(url_for("form_watch_checknow"), follow_redirects=True)
|
2021-02-21 12:41:00 +00:00
|
|
|
|
2021-02-21 19:46:56 +00:00
|
|
|
# Give the thread time to pick it up
|
2022-10-28 16:36:07 +00:00
|
|
|
wait_for_all_checks(client)
|
2021-02-21 12:41:00 +00:00
|
|
|
|
2021-02-21 19:46:56 +00:00
|
|
|
# It should report nothing found (no new 'unviewed' class)
|
|
|
|
|
res = client.get(url_for("index"))
|
|
|
|
|
assert b'unviewed' not in res.data
|
|
|
|
|
assert b'test-endpoint' in res.data
|
2021-02-21 12:41:00 +00:00
|
|
|
|
2021-04-30 09:54:23 +00:00
|
|
|
# Default no password set, this stuff should be always available.
|
|
|
|
|
|
|
|
|
|
assert b"SETTINGS" in res.data
|
|
|
|
|
assert b"BACKUP" in res.data
|
|
|
|
|
assert b"IMPORT" in res.data
|
|
|
|
|
|
2021-02-21 13:26:19 +00:00
|
|
|
#####################
|
2021-02-16 20:35:28 +00:00
|
|
|
|
2022-04-12 15:36:29 +00:00
|
|
|
# Check HTML conversion detected and workd
|
|
|
|
|
res = client.get(
|
|
|
|
|
url_for("preview_page", uuid="first"),
|
|
|
|
|
follow_redirects=True
|
|
|
|
|
)
|
|
|
|
|
# Check this class does not appear (that we didnt see the actual source)
|
|
|
|
|
assert b'foobar-detection' not in res.data
|
|
|
|
|
|
2021-02-21 12:41:00 +00:00
|
|
|
# Make a change
|
|
|
|
|
set_modified_response()
|
2021-02-16 20:35:28 +00:00
|
|
|
|
2021-02-21 12:41:00 +00:00
|
|
|
res = urlopen(url_for('test_endpoint', _external=True))
|
|
|
|
|
assert b'which has this one new line' in res.read()
|
2021-02-16 20:35:28 +00:00
|
|
|
|
2021-02-21 12:41:00 +00:00
|
|
|
# Force recheck
|
2022-05-20 14:27:51 +00:00
|
|
|
res = client.get(url_for("form_watch_checknow"), follow_redirects=True)
|
2023-01-05 20:12:02 +00:00
|
|
|
assert b'1 watches queued for rechecking.' in res.data
|
2021-02-16 20:35:28 +00:00
|
|
|
|
2022-10-28 16:36:07 +00:00
|
|
|
wait_for_all_checks(client)
|
2021-02-16 20:35:28 +00:00
|
|
|
|
2021-02-21 12:41:00 +00:00
|
|
|
# Now something should be ready, indicated by having a 'unviewed' class
|
|
|
|
|
res = client.get(url_for("index"))
|
|
|
|
|
assert b'unviewed' in res.data
|
2021-02-21 19:46:56 +00:00
|
|
|
|
2021-06-03 00:50:31 +00:00
|
|
|
# #75, and it should be in the RSS feed
|
2023-01-29 21:36:55 +00:00
|
|
|
rss_token = extract_rss_token_from_UI(client)
|
|
|
|
|
res = client.get(url_for("rss", token=rss_token, _external=True))
|
2021-06-03 00:50:31 +00:00
|
|
|
expected_url = url_for('test_endpoint', _external=True)
|
|
|
|
|
assert b'<rss' in res.data
|
2022-03-30 08:51:10 +00:00
|
|
|
|
|
|
|
|
# re #16 should have the diff in here too
|
2023-03-12 15:21:47 +00:00
|
|
|
assert b'(into) which has this one new line' in res.data
|
2022-03-30 08:51:10 +00:00
|
|
|
assert b'CDATA' in res.data
|
2022-06-02 08:01:03 +00:00
|
|
|
|
2021-06-03 00:50:31 +00:00
|
|
|
assert expected_url.encode('utf-8') in res.data
|
|
|
|
|
|
2021-02-21 19:46:56 +00:00
|
|
|
# Following the 'diff' link, it should no longer display as 'unviewed' even after we recheck it a few times
|
2021-02-27 08:05:25 +00:00
|
|
|
res = client.get(url_for("diff_history_page", uuid="first"))
|
2023-10-11 14:25:36 +00:00
|
|
|
assert b'selected=""' in res.data, "Confirm diff history page loaded"
|
2021-02-21 19:46:56 +00:00
|
|
|
|
2022-07-30 18:11:56 +00:00
|
|
|
# Check the [preview] pulls the right one
|
|
|
|
|
res = client.get(
|
|
|
|
|
url_for("preview_page", uuid="first"),
|
|
|
|
|
follow_redirects=True
|
|
|
|
|
)
|
|
|
|
|
assert b'which has this one new line' in res.data
|
|
|
|
|
assert b'Which is across multiple lines' not in res.data
|
|
|
|
|
|
2022-10-28 16:36:07 +00:00
|
|
|
wait_for_all_checks(client)
|
2021-02-21 19:54:15 +00:00
|
|
|
|
2021-02-21 19:46:56 +00:00
|
|
|
# Do this a few times.. ensures we dont accidently set the status
|
2021-02-27 08:05:25 +00:00
|
|
|
for n in range(2):
|
2022-05-20 14:27:51 +00:00
|
|
|
client.get(url_for("form_watch_checknow"), follow_redirects=True)
|
2021-02-21 19:46:56 +00:00
|
|
|
|
|
|
|
|
# Give the thread time to pick it up
|
2022-10-28 16:36:07 +00:00
|
|
|
wait_for_all_checks(client)
|
2021-02-21 19:46:56 +00:00
|
|
|
|
|
|
|
|
# It should report nothing found (no new 'unviewed' class)
|
|
|
|
|
res = client.get(url_for("index"))
|
|
|
|
|
assert b'unviewed' not in res.data
|
2022-06-02 08:01:03 +00:00
|
|
|
assert b'Mark all viewed' not in res.data
|
|
|
|
|
assert b'head title' not in res.data # Should not be present because this is off by default
|
2021-02-21 19:46:56 +00:00
|
|
|
assert b'test-endpoint' in res.data
|
2021-02-21 19:54:15 +00:00
|
|
|
|
|
|
|
|
set_original_response()
|
|
|
|
|
|
2021-06-24 09:10:19 +00:00
|
|
|
# Enable auto pickup of <title> in settings
|
|
|
|
|
res = client.post(
|
|
|
|
|
url_for("settings_page"),
|
2022-06-02 08:01:03 +00:00
|
|
|
data={"application-extract_title_as_title": "1", "requests-time_between_check-minutes": 180,
|
|
|
|
|
'application-fetch_backend': "html_requests"},
|
2021-06-24 09:10:19 +00:00
|
|
|
follow_redirects=True
|
|
|
|
|
)
|
|
|
|
|
|
2022-05-20 14:27:51 +00:00
|
|
|
client.get(url_for("form_watch_checknow"), follow_redirects=True)
|
2022-10-28 16:36:07 +00:00
|
|
|
wait_for_all_checks(client)
|
2021-06-24 09:10:19 +00:00
|
|
|
|
2021-02-21 19:54:15 +00:00
|
|
|
res = client.get(url_for("index"))
|
2021-02-27 08:05:25 +00:00
|
|
|
assert b'unviewed' in res.data
|
2022-06-02 08:01:03 +00:00
|
|
|
assert b'Mark all viewed' in res.data
|
|
|
|
|
|
2021-06-24 09:10:19 +00:00
|
|
|
# It should have picked up the <title>
|
|
|
|
|
assert b'head title' in res.data
|
2021-02-27 08:05:25 +00:00
|
|
|
|
2024-06-05 09:46:04 +00:00
|
|
|
# Be sure the last_viewed is going to be greater than the last snapshot
|
|
|
|
|
time.sleep(1)
|
|
|
|
|
|
2022-06-02 08:01:03 +00:00
|
|
|
# hit the mark all viewed link
|
|
|
|
|
res = client.get(url_for("mark_all_viewed"), follow_redirects=True)
|
|
|
|
|
|
|
|
|
|
assert b'Mark all viewed' not in res.data
|
|
|
|
|
assert b'unviewed' not in res.data
|
|
|
|
|
|
2021-06-24 09:10:19 +00:00
|
|
|
#
|
2021-02-27 21:01:42 +00:00
|
|
|
# Cleanup everything
|
2022-05-20 14:27:51 +00:00
|
|
|
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
|
2021-02-27 08:37:40 +00:00
|
|
|
assert b'Deleted' in res.data
|
