lee/changedetection.io
1
0
Fork 0
mirror of https://github.com/dgtlmoon/changedetection.io.git synced 2025-12-12 02:55:43 +00:00
Code Issues Actions 7 Packages Projects Releases Wiki Activity

feature: Support XPath2.0 to 3.1 (#1774)

Browse Source
This commit is contained in:
Constantin Hong
2023-11-14 00:42:21 +09:00
committed by GitHub
parent 5229094e44
commit 26931e0167
10 changed files with 541 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

225
changedetectionio/tests/test_xpath_selector.py
View File

@@ -6,9 +6,11 @@ from .util import live_server_setup, wait_for_all_checks
from ..html_tools import *
def test_setup(live_server):
live_server_setup(live_server)
def set_original_response():
test_return_data = """<html>
<body>
@@ -26,6 +28,7 @@ def set_original_response():
f.write(test_return_data)
return None
def set_modified_response():
test_return_data = """<html>
<body>
@@ -44,11 +47,12 @@ def set_modified_response():
return None
# Handle utf-8 charset replies https://github.com/dgtlmoon/changedetection.io/pull/613
def test_check_xpath_filter_utf8(client, live_server):
filter='//item/*[self::description]'
filter = '//item/*[self::description]'
d='''<?xml version="1.0" encoding="UTF-8"?>
d = '''<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:dc="http://purl.org/dc/elements/1.1/" version="2.0">
<channel>
<title>rpilocator.com</title>
@@ -102,9 +106,9 @@ def test_check_xpath_filter_utf8(client, live_server):
# Handle utf-8 charset replies https://github.com/dgtlmoon/changedetection.io/pull/613
def test_check_xpath_text_function_utf8(client, live_server):
filter='//item/title/text()'
filter = '//item/title/text()'
d='''<?xml version="1.0" encoding="UTF-8"?>
d = '''<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:dc="http://purl.org/dc/elements/1.1/" version="2.0">
<channel>
<title>rpilocator.com</title>
@@ -163,15 +167,12 @@ def test_check_xpath_text_function_utf8(client, live_server):
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
assert b'Deleted' in res.data
def test_check_markup_xpath_filter_restriction(client, live_server):
def test_check_markup_xpath_filter_restriction(client, live_server):
xpath_filter = "//*[contains(@class, 'sametext')]"
set_original_response()
# Give the endpoint time to spin up
time.sleep(1)
# Add our URL to the import page
test_url = url_for('test_endpoint', _external=True)
res = client.post(
@@ -214,7 +215,6 @@ def test_check_markup_xpath_filter_restriction(client, live_server):
def test_xpath_validation(client, live_server):
# Add our URL to the import page
test_url = url_for('test_endpoint', _external=True)
res = client.post(
@@ -235,6 +235,48 @@ def test_xpath_validation(client, live_server):
assert b'Deleted' in res.data
def test_xpath23_prefix_validation(client, live_server):
# 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
wait_for_all_checks(client)
res = client.post(
url_for("edit_page", uuid="first"),
data={"include_filters": "xpath:/something horrible", "url": test_url, "tags": "", "headers": "", 'fetch_backend': "html_requests"},
follow_redirects=True
)
assert b"is not a valid XPath expression" in res.data
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
assert b'Deleted' in res.data
def test_xpath1_validation(client, live_server):
# 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
wait_for_all_checks(client)
res = client.post(
url_for("edit_page", uuid="first"),
data={"include_filters": "xpath1:/something horrible", "url": test_url, "tags": "", "headers": "", 'fetch_backend': "html_requests"},
follow_redirects=True
)
assert b"is not a valid XPath expression" in res.data
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
assert b'Deleted' in res.data
# actually only really used by the distll.io importer, but could be handy too
def test_check_with_prefix_include_filters(client, live_server):
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
@@ -254,7 +296,8 @@ def test_check_with_prefix_include_filters(client, live_server):
res = client.post(
url_for("edit_page", uuid="first"),
data={"include_filters": "xpath://*[contains(@class, 'sametext')]", "url": test_url, "tags": "", "headers": "", 'fetch_backend': "html_requests"},
data={"include_filters": "xpath://*[contains(@class, 'sametext')]", "url": test_url, "tags": "", "headers": "",
'fetch_backend': "html_requests"},
follow_redirects=True
)
@@ -266,13 +309,15 @@ def test_check_with_prefix_include_filters(client, live_server):
follow_redirects=True
)
assert b"Some text thats the same" in res.data #in selector
assert b"Some text that will change" not in res.data #not in selector
assert b"Some text thats the same" in res.data # in selector
assert b"Some text that will change" not in res.data # not in selector
client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
def test_various_rules(client, live_server):
# Just check these don't error
#live_server_setup(live_server)
# live_server_setup(live_server)
with open("test-datastore/endpoint-content.txt", "w") as f:
f.write("""<html>
<body>
@@ -285,10 +330,11 @@ def test_various_rules(client, live_server):
<a href=''>some linky </a>
<a href=''>another some linky </a>
<!-- related to https://github.com/dgtlmoon/changedetection.io/pull/1774 -->
<input type="email" id="email" />
<input type="email" id="email" />
</body>
</html>
""")
test_url = url_for('test_endpoint', _external=True)
res = client.post(
url_for("import_page"),
@@ -298,7 +344,6 @@ def test_various_rules(client, live_server):
assert b"1 Imported" in res.data
wait_for_all_checks(client)
for r in ['//div', '//a', 'xpath://div', 'xpath://a']:
res = client.post(
url_for("edit_page", uuid="first"),
@@ -313,3 +358,153 @@ def test_various_rules(client, live_server):
assert b"Updated watch." in res.data
res = client.get(url_for("index"))
assert b'fetch-error' not in res.data, f"Should not see errors after '{r} filter"
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
assert b'Deleted' in res.data
def test_xpath_20(client, live_server):
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
wait_for_all_checks(client)
set_original_response()
test_url = url_for('test_endpoint', _external=True)
res = client.post(
url_for("edit_page", uuid="first"),
data={"include_filters": "//*[contains(@class, 'sametext')]|//*[contains(@class, 'changetext')]",
"url": test_url,
"tags": "",
"headers": "",
'fetch_backend': "html_requests"},
follow_redirects=True
)
assert b"Updated watch." in res.data
wait_for_all_checks(client)
res = client.get(
url_for("preview_page", uuid="first"),
follow_redirects=True
)
assert b"Some text thats the same" in res.data # in selector
assert b"Some text that will change" in res.data # in selector
client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
def test_xpath_20_function_count(client, live_server):
set_original_response()
# 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
wait_for_all_checks(client)
res = client.post(
url_for("edit_page", uuid="first"),
data={"include_filters": "xpath:count(//div) * 123456789987654321",
"url": test_url,
"tags": "",
"headers": "",
'fetch_backend': "html_requests"},
follow_redirects=True
)
assert b"Updated watch." in res.data
wait_for_all_checks(client)
res = client.get(
url_for("preview_page", uuid="first"),
follow_redirects=True
)
assert b"246913579975308642" in res.data # in selector
client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
def test_xpath_20_function_count2(client, live_server):
set_original_response()
# 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
wait_for_all_checks(client)
res = client.post(
url_for("edit_page", uuid="first"),
data={"include_filters": "/html/body/count(div) * 123456789987654321",
"url": test_url,
"tags": "",
"headers": "",
'fetch_backend': "html_requests"},
follow_redirects=True
)
assert b"Updated watch." in res.data
wait_for_all_checks(client)
res = client.get(
url_for("preview_page", uuid="first"),
follow_redirects=True
)
assert b"246913579975308642" in res.data # in selector
client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
def test_xpath_20_function_string_join_matches(client, live_server):
set_original_response()
# 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
wait_for_all_checks(client)
res = client.post(
url_for("edit_page", uuid="first"),
data={
"include_filters": "xpath:string-join(//*[contains(@class, 'sametext')]|//*[matches(@class, 'changetext')], 'specialconjunction')",
"url": test_url,
"tags": "",
"headers": "",
'fetch_backend': "html_requests"},
follow_redirects=True
)
assert b"Updated watch." in res.data
wait_for_all_checks(client)
res = client.get(
url_for("preview_page", uuid="first"),
follow_redirects=True
)
assert b"Some text thats the samespecialconjunctionSome text that will change" in res.data # in selector
client.get(url_for("form_delete", uuid="all"), follow_redirects=True)