Compare commits

...

18 Commits

Author SHA1 Message Date
dgtlmoon
cb7b12ab9c fix dupe import 2022-10-27 11:23:46 +02:00
dgtlmoon
197a3e6366 cleanup of backup creation 2022-10-27 11:21:55 +02:00
dgtlmoon
9c1b18ec22 Use a rglob that will work cross-platform more reliably 2022-10-27 11:05:39 +02:00
dgtlmoon
02b460768c Support moving the data set 2022-10-27 10:56:10 +02:00
dgtlmoon
50edc36cc4 Improve test 2022-10-27 10:29:28 +02:00
dgtlmoon
3cf2a798cf Remove dupe code 2022-10-26 11:09:49 +02:00
dgtlmoon
13daa92c92 Make snapshots path relative so the datadir is portable 2022-10-26 10:49:37 +02:00
dgtlmoon
53f4472fb8 Merge branch 'master' of github.com:dgtlmoon/changedetection.io into history-txt-snapshot-fix 2022-10-25 11:45:18 +02:00
dgtlmoon
724cb17224 Re #1052 - Dynamic URLs, use variables in the URL (such as the current date, the date in a month, and other logic see https://github.com/dgtlmoon/changedetection.io/wiki/Handling-variables-in-the-watched-URL ) (#1057) 2022-10-24 23:20:39 +02:00
dgtlmoon
a1bdf7e43f Backup zip - always use correct relative path of the snapshot text instead of the datadir path 2022-10-24 15:53:25 +02:00
dgtlmoon
4eb4b401a1 API - system info - allow 5 minutes grace before watch is considered 'overdue' 2022-10-23 23:12:28 +02:00
dgtlmoon
5d40e16c73 API - Adding basic system info/system state API (#1051) 2022-10-23 19:15:11 +02:00
dgtlmoon
492bbce6b6 Build - Fix syntax in container build test (#1050) 2022-10-23 16:02:13 +02:00
dgtlmoon
0394a56be5 Building - Test container build on PR 2022-10-23 15:54:19 +02:00
Entepotenz
7839551d6b Testing - Use same version of playwright while running tests as in production builds (#1047) 2022-10-23 11:26:32 +02:00
Entepotenz
9c5588c791 update path for validation in the CONTRIBUTING.md (#1046) 2022-10-23 11:25:29 +02:00
dgtlmoon
5a43a350de History index safety check - Be sure that only valid history index lines are read (#1042) 2022-10-19 22:41:13 +02:00
Michael McMillan
3c31f023ce Option to Hide the Referer header from monitored websites. (#996) 2022-10-18 09:16:22 +02:00
18 changed files with 217 additions and 48 deletions

View File

@@ -1,12 +1,21 @@
name: ChangeDetection.io Container Build Test
# Triggers the workflow on push or pull request events
# This line doesnt work, even tho it is the documented one
#on: [push, pull_request]
on:
push:
paths:
- requirements.txt
- Dockerfile
pull_request:
paths:
- requirements.txt
- Dockerfile
# Changes to requirements.txt packages and Dockerfile may or may not always be compatible with arm etc, so worth testing
# @todo: some kind of path filter for requirements.txt and Dockerfile
jobs:

View File

@@ -6,7 +6,7 @@ Otherwise, it's always best to PR into the `dev` branch.
Please be sure that all new functionality has a matching test!
Use `pytest` to validate/test, you can run the existing tests as `pytest tests/test_notifications.py` for example
Use `pytest` to validate/test, you can run the existing tests as `pytest tests/test_notification.py` for example
```
pip3 install -r requirements-dev

View File

@@ -64,6 +64,7 @@ EXPOSE 5000
# The actual flask app
COPY changedetectionio /app/changedetectionio
# The eventlet server wrapper
COPY changedetection.py /app/changedetection.py

View File

@@ -194,6 +194,9 @@ def changedetection_app(config=None, datastore_o=None):
watch_api.add_resource(api_v1.Watch, '/api/v1/watch/<string:uuid>',
resource_class_kwargs={'datastore': datastore, 'update_q': update_q})
watch_api.add_resource(api_v1.SystemInfo, '/api/v1/systeminfo',
resource_class_kwargs={'datastore': datastore, 'update_q': update_q})
@@ -984,9 +987,6 @@ def changedetection_app(config=None, datastore_o=None):
# create a ZipFile object
backupname = "changedetection-backup-{}.zip".format(int(time.time()))
# We only care about UUIDS from the current index file
uuids = list(datastore.data['watching'].keys())
backup_filepath = os.path.join(datastore_o.datastore_path, backupname)
with zipfile.ZipFile(backup_filepath, "w",
@@ -1002,12 +1002,12 @@ def changedetection_app(config=None, datastore_o=None):
# Add the flask app secret
zipObj.write(os.path.join(datastore_o.datastore_path, "secret.txt"), arcname="secret.txt")
# Add any snapshot data we find, use the full path to access the file, but make the file 'relative' in the Zip.
for txt_file_path in Path(datastore_o.datastore_path).rglob('*.txt'):
parent_p = txt_file_path.parent
if parent_p.name in uuids:
zipObj.write(txt_file_path,
arcname=str(txt_file_path).replace(datastore_o.datastore_path, ''),
# Add any data in the watch data directory.
for uuid, w in datastore.data['watching'].items():
for f in Path(w.watch_data_dir).glob('*'):
zipObj.write(f,
# Use the full path to access the file, but make the file 'relative' in the Zip.
arcname=os.path.join(f.parts[-2], f.parts[-1]),
compress_type=zipfile.ZIP_DEFLATED,
compresslevel=8)

View File

@@ -122,3 +122,37 @@ class CreateWatch(Resource):
return {'status': "OK"}, 200
return list, 200
class SystemInfo(Resource):
def __init__(self, **kwargs):
# datastore is a black box dependency
self.datastore = kwargs['datastore']
self.update_q = kwargs['update_q']
@auth.check_token
def get(self):
import time
overdue_watches = []
# Check all watches and report which have not been checked but should have been
for uuid, watch in self.datastore.data.get('watching', {}).items():
# see if now - last_checked is greater than the time that should have been
# this is not super accurate (maybe they just edited it) but better than nothing
t = watch.threshold_seconds()
if not t:
# Use the system wide default
t = self.datastore.threshold_seconds
time_since_check = time.time() - watch.get('last_checked')
# Allow 5 minutes of grace time before we decide it's overdue
if time_since_check - (5 * 60) > t:
overdue_watches.append(uuid)
return {
'queue_size': self.update_q.qsize(),
'overdue_watches': overdue_watches,
'uptime': round(time.time() - self.datastore.start_time, 2),
'watch_count': len(self.datastore.data.get('watching', {}))
}, 200

View File

@@ -102,6 +102,14 @@ def main():
has_password=datastore.data['settings']['application']['password'] != False
)
# Monitored websites will not receive a Referer header
# when a user clicks on an outgoing link.
@app.after_request
def hide_referrer(response):
if os.getenv("HIDE_REFERER", False):
response.headers["Referrer-Policy"] = "no-referrer"
return response
# Proxy sub-directory support
# Set environment var USE_X_SETTINGS=1 on this script
# And then in your proxy_pass settings

Binary file not shown.

View File

@@ -35,6 +35,8 @@ class perform_site_check():
def run(self, uuid):
from jinja2 import Environment
changed_detected = False
screenshot = False # as bytes
stripped_text_from_html = ""
@@ -65,7 +67,11 @@ class perform_site_check():
request_headers['Accept-Encoding'] = request_headers['Accept-Encoding'].replace(', br', '')
timeout = self.datastore.data['settings']['requests'].get('timeout')
url = watch.get('url')
# Jinja2 available in URLs along with https://pypi.org/project/jinja2-time/
jinja2_env = Environment(extensions=['jinja2_time.TimeExtension'])
url = str(jinja2_env.from_string(watch.get('url')).render())
request_body = self.datastore.data['watching'][uuid].get('body')
request_method = self.datastore.data['watching'][uuid].get('method')
ignore_status_codes = self.datastore.data['watching'][uuid].get('ignore_status_codes', False)

View File

@@ -1,6 +1,8 @@
import os
import uuid as uuid_builder
from distutils.util import strtobool
import logging
import os
import time
import uuid
minimum_seconds_recheck_time = int(os.getenv('MINIMUM_SECONDS_RECHECK_TIME', 60))
mtable = {'seconds': 1, 'minutes': 60, 'hours': 3600, 'days': 86400, 'weeks': 86400 * 7}
@@ -22,7 +24,7 @@ class model(dict):
#'newest_history_key': 0,
'title': None,
'previous_md5': False,
'uuid': str(uuid_builder.uuid4()),
'uuid': str(uuid.uuid4()),
'headers': {}, # Extra headers to send
'body': None,
'method': 'GET',
@@ -60,7 +62,7 @@ class model(dict):
self.update(self.__base_config)
self.__datastore_path = kw['datastore_path']
self['uuid'] = str(uuid_builder.uuid4())
self['uuid'] = str(uuid.uuid4())
del kw['datastore_path']
@@ -82,10 +84,9 @@ class model(dict):
return False
def ensure_data_dir_exists(self):
target_path = os.path.join(self.__datastore_path, self['uuid'])
if not os.path.isdir(target_path):
print ("> Creating data dir {}".format(target_path))
os.mkdir(target_path)
if not os.path.isdir(self.watch_data_dir):
print ("> Creating data dir {}".format(self.watch_data_dir))
os.mkdir(self.watch_data_dir)
@property
def label(self):
@@ -109,16 +110,40 @@ class model(dict):
@property
def history(self):
"""History index is just a text file as a list
{watch-uuid}/history.txt
contains a list like
{epoch-time},{filename}\n
We read in this list as the history information
"""
tmp_history = {}
import logging
import time
# Read the history file as a dict
fname = os.path.join(self.__datastore_path, self.get('uuid'), "history.txt")
fname = os.path.join(self.watch_data_dir, "history.txt")
if os.path.isfile(fname):
logging.debug("Reading history index " + str(time.time()))
with open(fname, "r") as f:
tmp_history = dict(i.strip().split(',', 2) for i in f.readlines())
for i in f.readlines():
if ',' in i:
k, v = i.strip().split(',', 2)
# The index history could contain a relative path, so we need to make the fullpath
# so that python can read it
if not '/' in v and not '\'' in v:
v = os.path.join(self.watch_data_dir, v)
else:
# It's possible that they moved the datadir on older versions
# So the snapshot exists but is in a different path
snapshot_fname = v.split('/')[-1]
proposed_new_path = os.path.join(self.watch_data_dir, snapshot_fname)
if not os.path.exists(v) and os.path.exists(proposed_new_path):
v = proposed_new_path
tmp_history[k] = v
if len(tmp_history):
self.__newest_history_key = list(tmp_history.keys())[-1]
@@ -129,7 +154,7 @@ class model(dict):
@property
def has_history(self):
fname = os.path.join(self.__datastore_path, self.get('uuid'), "history.txt")
fname = os.path.join(self.watch_data_dir, "history.txt")
return os.path.isfile(fname)
# Returns the newest key, but if theres only 1 record, then it's counted as not being new, so return 0.
@@ -148,25 +173,19 @@ class model(dict):
# Save some text file to the appropriate path and bump the history
# result_obj from fetch_site_status.run()
def save_history_text(self, contents, timestamp):
import uuid
import logging
output_path = os.path.join(self.__datastore_path, self['uuid'])
self.ensure_data_dir_exists()
snapshot_fname = os.path.join(output_path, str(uuid.uuid4()))
logging.debug("Saving history text {}".format(snapshot_fname))
snapshot_fname = "{}.txt".format(str(uuid.uuid4()))
# in /diff/ and /preview/ we are going to assume for now that it's UTF-8 when reading
# most sites are utf-8 and some are even broken utf-8
with open(snapshot_fname, 'wb') as f:
with open(os.path.join(self.watch_data_dir, snapshot_fname), 'wb') as f:
f.write(contents)
f.close()
# Append to index
# @todo check last char was \n
index_fname = os.path.join(output_path, "history.txt")
index_fname = os.path.join(self.watch_data_dir, "history.txt")
with open(index_fname, 'a') as f:
f.write("{},{}\n".format(timestamp, snapshot_fname))
f.close()
@@ -207,14 +226,14 @@ class model(dict):
return not local_lines.issubset(existing_history)
def get_screenshot(self):
fname = os.path.join(self.__datastore_path, self['uuid'], "last-screenshot.png")
fname = os.path.join(self.watch_data_dir, "last-screenshot.png")
if os.path.isfile(fname):
return fname
return False
def __get_file_ctime(self, filename):
fname = os.path.join(self.__datastore_path, self['uuid'], filename)
fname = os.path.join(self.watch_data_dir, filename)
if os.path.isfile(fname):
return int(os.path.getmtime(fname))
return False
@@ -239,9 +258,14 @@ class model(dict):
def snapshot_error_screenshot_ctime(self):
return self.__get_file_ctime('last-error-screenshot.png')
@property
def watch_data_dir(self):
# The base dir of the watch data
return os.path.join(self.__datastore_path, self['uuid'])
def get_error_text(self):
"""Return the text saved from a previous request that resulted in a non-200 error"""
fname = os.path.join(self.__datastore_path, self['uuid'], "last-error.txt")
fname = os.path.join(self.watch_data_dir, "last-error.txt")
if os.path.isfile(fname):
with open(fname, 'r') as f:
return f.read()
@@ -249,7 +273,7 @@ class model(dict):
def get_error_snapshot(self):
"""Return path to the screenshot that resulted in a non-200 error"""
fname = os.path.join(self.__datastore_path, self['uuid'], "last-error-screenshot.png")
fname = os.path.join(self.watch_data_dir, "last-error-screenshot.png")
if os.path.isfile(fname):
return fname
return False

View File

@@ -9,6 +9,8 @@
# exit when any command fails
set -e
SCRIPT_DIR=$( cd -- "$( dirname -- "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )
find tests/test_*py -type f|while read test_name
do
echo "TEST RUNNING $test_name"
@@ -45,7 +47,9 @@ docker kill $$-test_selenium
echo "TESTING WEBDRIVER FETCH > PLAYWRIGHT/BROWSERLESS..."
# Not all platforms support playwright (not ARM/rPI), so it's not packaged in requirements.txt
pip3 install playwright~=1.24
PLAYWRIGHT_VERSION=$(grep -i -E "RUN pip install.+" "$SCRIPT_DIR/../Dockerfile" | grep --only-matching -i -E "playwright[=><~+]+[0-9\.]+")
echo "using $PLAYWRIGHT_VERSION"
pip3 install "$PLAYWRIGHT_VERSION"
docker run -d --name $$-test_browserless -e "DEFAULT_LAUNCH_ARGS=[\"--window-size=1920,1080\"]" --rm -p 3000:3000 --shm-size="2g" browserless/chrome:1.53-chrome-stable
# takes a while to spin up
sleep 5

View File

@@ -30,14 +30,14 @@ class ChangeDetectionStore:
def __init__(self, datastore_path="/datastore", include_default_watches=True, version_tag="0.0.0"):
# Should only be active for docker
# logging.basicConfig(filename='/dev/stdout', level=logging.INFO)
self.needs_write = False
self.__data = App.model()
self.datastore_path = datastore_path
self.json_store_path = "{}/url-watches.json".format(self.datastore_path)
self.needs_write = False
self.proxy_list = None
self.start_time = time.time()
self.stop_thread = False
self.__data = App.model()
# Base definition for all watchers
# deepcopy part of #569 - not sure why its needed exactly
self.generic_definition = deepcopy(Watch.model(datastore_path = datastore_path, default={}))

View File

@@ -40,7 +40,8 @@
<fieldset>
<div class="pure-control-group">
{{ render_field(form.url, placeholder="https://...", required=true, class="m-d") }}
<span class="pure-form-message-inline">Some sites use JavaScript to create the content, for this you should <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Fetching-pages-with-WebDriver">use the Chrome/WebDriver Fetcher</a></span>
<span class="pure-form-message-inline">Some sites use JavaScript to create the content, for this you should <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Fetching-pages-with-WebDriver">use the Chrome/WebDriver Fetcher</a></span><br/>
<span class="pure-form-message-inline">You can use variables in the URL, perfect for inserting the current date and other logic, <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Handling-variables-in-the-watched-URL">help and examples here</a></span><br/>
</div>
<div class="pure-control-group">
{{ render_field(form.title, class="m-d") }}

View File

@@ -147,6 +147,16 @@ def test_api_simple(client, live_server):
# @todo how to handle None/default global values?
assert watch['history_n'] == 2, "Found replacement history section, which is in its own API"
# basic systeminfo check
res = client.get(
url_for("systeminfo"),
headers={'x-api-key': api_key},
)
info = json.loads(res.data)
assert info.get('watch_count') == 1
assert info.get('uptime') > 0.5
# Finally delete the watch
res = client.delete(
url_for("watch", uuid=watch_uuid),

View File

@@ -1,18 +1,31 @@
#!/usr/bin/python3
import time
from .util import set_original_response, set_modified_response, live_server_setup
from flask import url_for
from urllib.request import urlopen
from . util import set_original_response, set_modified_response, live_server_setup
from zipfile import ZipFile
import re
import time
def test_backup(client, live_server):
live_server_setup(live_server)
set_original_response()
# Give the endpoint time to spin up
time.sleep(1)
# 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
)
assert b"1 Imported" in res.data
time.sleep(3)
res = client.get(
url_for("get_backup"),
follow_redirects=True
@@ -20,6 +33,19 @@ def test_backup(client, live_server):
# Should get the right zip content type
assert res.content_type == "application/zip"
# Should be PK/ZIP stream
assert res.data.count(b'PK') >= 2
# ZipFile from buffer seems non-obvious, just save it instead
with open("download.zip", 'wb') as f:
f.write(res.data)
zip = ZipFile('download.zip')
l = zip.namelist()
uuid4hex = re.compile('^[a-f0-9]{8}-?[a-f0-9]{4}-?4[a-f0-9]{3}-?[89ab][a-f0-9]{3}-?[a-f0-9]{12}.*txt', re.I)
newlist = list(filter(uuid4hex.match, l)) # Read Note below
# Should be two txt files in the archive (history and the snapshot)
assert len(newlist) == 2

View File

@@ -0,0 +1,33 @@
#!/usr/bin/python3
import time
from flask import url_for
from .util import live_server_setup
# If there was only a change in the whitespacing, then we shouldnt have a change detected
def test_jinja2_in_url_query(client, live_server):
live_server_setup(live_server)
# Give the endpoint time to spin up
time.sleep(1)
# Add our URL to the import page
test_url = url_for('test_return_query', _external=True)
# because url_for() will URL-encode the var, but we dont here
full_url = "{}?{}".format(test_url,
"date={% now 'Europe/Berlin', '%Y' %}.{% now 'Europe/Berlin', '%m' %}.{% now 'Europe/Berlin', '%d' %}", )
res = client.post(
url_for("form_quick_watch_add"),
data={"url": full_url, "tag": "test"},
follow_redirects=True
)
assert b"Watch added" in res.data
time.sleep(3)
# It should report nothing found (no new 'unviewed' class)
res = client.get(
url_for("preview_page", uuid="first"),
follow_redirects=True
)
assert b'date=2' in res.data

View File

@@ -159,5 +159,10 @@ def live_server_setup(live_server):
ret = " ".join([auth.username, auth.password, auth.type])
return ret
# Just return some GET var
@live_server.app.route('/test-return-query', methods=['GET'])
def test_return_query():
return request.query_string
live_server.start()

View File

@@ -45,6 +45,9 @@ services:
# Respect proxy_pass type settings, `proxy_set_header Host "localhost";` and `proxy_set_header X-Forwarded-Prefix /app;`
# More here https://github.com/dgtlmoon/changedetection.io/wiki/Running-changedetection.io-behind-a-reverse-proxy-sub-directory
# - USE_X_SETTINGS=1
#
# Hides the `Referer` header so that monitored websites can't see the changedetection.io hostname.
# - HIDE_REFERER=true
# Comment out ports: when using behind a reverse proxy , enable networks: etc.
ports:

View File

@@ -1,8 +1,8 @@
flask~= 2.0
flask ~= 2.0
flask_wtf
eventlet>=0.31.0
eventlet >= 0.31.0
validators
timeago ~=1.0
timeago ~= 1.0
inscriptis ~= 2.2
feedgen ~= 0.9
flask-login ~= 0.5
@@ -46,4 +46,9 @@ selenium ~= 4.1.0
# need to revisit flask login versions
werkzeug ~= 2.0.0
# Templating, so far just in the URLs but in the future can be for the notifications also
jinja2
jinja2-time
# playwright is installed at Dockerfile build time because it's not available on all platforms