Merge branch 'master' into diff-filters

diff-proposed-for-bwees
dgtlmoon 2 years ago committed by GitHub
commit a50d6db0b2
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -0,0 +1,46 @@
name: ChangeDetection.io Container Build Test
# Triggers the workflow on push or pull request events
on:
push:
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:
test-container-build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Set up Python 3.9
uses: actions/setup-python@v2
with:
python-version: 3.9
# Just test that the build works, some libraries won't compile on ARM/rPi etc
- name: Set up QEMU
uses: docker/setup-qemu-action@v1
with:
image: tonistiigi/binfmt:latest
platforms: all
- name: Set up Docker Buildx
id: buildx
uses: docker/setup-buildx-action@v1
with:
install: true
version: latest
driver-opts: image=moby/buildkit:master
- name: Test that the docker containers can build
id: docker_build
uses: docker/build-push-action@v2
# https://github.com/docker/build-push-action#customizing
with:
context: ./
file: ./Dockerfile
platforms: linux/arm/v7,linux/arm/v6,linux/amd64,linux/arm64,
cache-from: type=local,src=/tmp/.buildx-cache
cache-to: type=local,dest=/tmp/.buildx-cache

@ -1,28 +1,25 @@
name: ChangeDetection.io Test
name: ChangeDetection.io App Test
# Triggers the workflow on push or pull request events
on: [push, pull_request]
jobs:
test-build:
test-application:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Set up Python 3.9
uses: actions/setup-python@v2
with:
python-version: 3.9
- name: Show env vars
run: set
- name: Install dependencies
run: |
python -m pip install --upgrade pip
pip install flake8 pytest
if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
if [ -f requirements-dev.txt ]; then pip install -r requirements-dev.txt; fi
- name: Lint with flake8
run: |
# stop the build if there are Python syntax errors or undefined names
@ -39,7 +36,4 @@ jobs:
# Each test is totally isolated and performs its own cleanup/reset
cd changedetectionio; ./run_all_tests.sh
# https://github.com/docker/build-push-action/blob/master/docs/advanced/test-before-push.md ?
# https://github.com/docker/buildx/issues/59 ? Needs to be one platform?
# https://github.com/docker/buildx/issues/495#issuecomment-918925854

@ -5,13 +5,14 @@ FROM python:3.8-slim as builder
ARG CRYPTOGRAPHY_DONT_BUILD_RUST=1
RUN apt-get update && apt-get install -y --no-install-recommends \
libssl-dev \
libffi-dev \
g++ \
gcc \
libc-dev \
libffi-dev \
libssl-dev \
libxslt-dev \
zlib1g-dev \
g++
make \
zlib1g-dev
RUN mkdir /install
WORKDIR /install
@ -22,7 +23,7 @@ RUN pip install --target=/dependencies -r /requirements.txt
# Playwright is an alternative to Selenium
# Excluded this package from requirements.txt to prevent arm/v6 and arm/v7 builds from failing
RUN pip install --target=/dependencies playwright~=1.24 \
RUN pip install --target=/dependencies playwright~=1.26 \
|| echo "WARN: Failed to install Playwright. The application can still run, but the Playwright option will be disabled."
# Final image stage

@ -33,7 +33,7 @@ _Need an actual Chrome runner with Javascript support? We support fetching via W
#### Key Features
- Lots of trigger filters, such as "Trigger on text", "Remove text by selector", "Ignore text", "Extract text", also using regular-expressions!
- Target elements with xPath and CSS Selectors, Easily monitor complex JSON with JsonPath rules
- Target elements with xPath and CSS Selectors, Easily monitor complex JSON with JSONPath or jq
- Switch between fast non-JS and Chrome JS based "fetchers"
- Easily specify how often a site should be checked
- Execute JS before extracting text (Good for logging in, see examples in the UI!)

@ -12,11 +12,14 @@ Know when important content changes, we support notifications via Discord, Teleg
[**Don't have time? Let us host it for you! try our $6.99/month subscription - use our proxies and support!**](https://lemonade.changedetection.io/start) , _half the price of other website change monitoring services and comes with unlimited watches & checks!_
- Chrome browser included.
- Super fast, no registration needed setup.
- Start watching and receiving change notifications instantly.
- Automatic Updates, Automatic Backups, No Heroku "paused application", don't miss a change!
- Javascript browser included
- Unlimited checks and watches!
Easily see what changed, examine by word, line, or individual character.
<img src="https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/docs/screenshot-diff.png" style="max-width:100%;" alt="Self-hosted web page change monitoring context difference " title="Self-hosted web page change monitoring context difference " />
#### Example use cases
@ -44,22 +47,18 @@ _Need an actual Chrome runner with Javascript support? We support fetching via W
#### Key Features
- Lots of trigger filters, such as "Trigger on text", "Remove text by selector", "Ignore text", "Extract text", also using regular-expressions!
- Target elements with xPath and CSS Selectors, Easily monitor complex JSON with JsonPath rules
- Target elements with xPath and CSS Selectors, Easily monitor complex JSON with JSONPath or jq
- Switch between fast non-JS and Chrome JS based "fetchers"
- Easily specify how often a site should be checked
- Execute JS before extracting text (Good for logging in, see examples in the UI!)
- Override Request Headers, Specify `POST` or `GET` and other methods
- Use the "Visual Selector" to help target specific elements
- Configurable [proxy per watch](https://github.com/dgtlmoon/changedetection.io/wiki/Proxy-configuration)
We [recommend and use Bright Data](https://brightdata.grsm.io/n0r16zf7eivq) global proxy services, Bright Data will match any first deposit up to $100 using our signup link.
## Screenshots
### Examine differences in content.
Easily see what changed, examine by word, line, or individual character.
<img src="https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/docs/screenshot-diff.png" style="max-width:100%;" alt="Self-hosted web page change monitoring context difference " title="Self-hosted web page change monitoring context difference " />
Please :star: star :star: this project and help it grow! https://github.com/dgtlmoon/changedetection.io/
### Filter by elements using the Visual Selector tool.
@ -122,7 +121,7 @@ See the wiki for more information https://github.com/dgtlmoon/changedetection.io
## Filters
XPath, JSONPath and CSS support comes baked in! You can be as specific as you need, use XPath exported from various XPath element query creation tools.
XPath, JSONPath, jq, and CSS support comes baked in! You can be as specific as you need, use XPath exported from various XPath element query creation tools.
(We support LXML `re:test`, `re:math` and `re:replace`.)
@ -152,7 +151,7 @@ Now you can also customise your notification content!
## JSON API Monitoring
Detect changes and monitor data in JSON API's by using the built-in JSONPath selectors as a filter / selector.
Detect changes and monitor data in JSON API's by using either JSONPath or jq to filter, parse, and restructure JSON as needed.
![image](https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/docs/json-filter-field-example.png)
@ -160,9 +159,52 @@ This will re-parse the JSON and apply formatting to the text, making it super ea
![image](https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/docs/json-diff-example.png)
### JSONPath or jq?
For more complex parsing, filtering, and modifying of JSON data, jq is recommended due to the built-in operators and functions. Refer to the [documentation](https://stedolan.github.io/jq/manual/) for more information on jq.
The example below adds the price in dollars to each item in the JSON data, and then filters to only show items that are greater than 10.
#### Sample input data from API
```
{
"items": [
{
"name": "Product A",
"priceInCents": 2500
},
{
"name": "Product B",
"priceInCents": 500
},
{
"name": "Product C",
"priceInCents": 2000
}
]
}
```
#### Sample jq
`jq:.items[] | . + { "priceInDollars": (.priceInCents / 100) } | select(.priceInDollars > 10)`
#### Sample output data
```
{
"name": "Product A",
"priceInCents": 2500,
"priceInDollars": 25
}
{
"name": "Product C",
"priceInCents": 2000,
"priceInDollars": 20
}
```
### Parse JSON embedded in HTML!
When you enable a `json:` filter, you can even automatically extract and parse embedded JSON inside a HTML page! Amazingly handy for sites that build content based on JSON, such as many e-commerce websites.
When you enable a `json:` or `jq:` filter, you can even automatically extract and parse embedded JSON inside a HTML page! Amazingly handy for sites that build content based on JSON, such as many e-commerce websites.
```
<html>
@ -172,7 +214,7 @@ When you enable a `json:` filter, you can even automatically extract and parse e
</script>
```
`json:$.price` would give `23.50`, or you can extract the whole structure
`json:$.price` or `jq:.price` would give `23.50`, or you can extract the whole structure
## Proxy configuration

@ -33,7 +33,7 @@ from flask_wtf import CSRFProtect
from changedetectionio import html_tools
from changedetectionio.api import api_v1
__version__ = '0.39.19'
__version__ = '0.39.20.1'
datastore = None
@ -547,6 +547,7 @@ def changedetection_app(config=None, datastore_o=None):
# Defaults for proxy choice
if datastore.proxy_list is not None: # When enabled
# @todo
# Radio needs '' not None, or incase that the chosen one no longer exists
if default['proxy'] is None or not any(default['proxy'] in tup for tup in datastore.proxy_list):
default['proxy'] = ''
@ -560,7 +561,10 @@ def changedetection_app(config=None, datastore_o=None):
# @todo - Couldn't get setattr() etc dynamic addition working, so remove it instead
del form.proxy
else:
form.proxy.choices = [('', 'Default')] + datastore.proxy_list
form.proxy.choices = [('', 'Default')]
for p in datastore.proxy_list:
form.proxy.choices.append(tuple((p, datastore.proxy_list[p]['label'])))
if request.method == 'POST' and form.validate():
extra_update_obj = {}
@ -657,15 +661,16 @@ def changedetection_app(config=None, datastore_o=None):
default = deepcopy(datastore.data['settings'])
if datastore.proxy_list is not None:
available_proxies = list(datastore.proxy_list.keys())
# When enabled
system_proxy = datastore.data['settings']['requests']['proxy']
# In the case it doesnt exist anymore
if not any([system_proxy in tup for tup in datastore.proxy_list]):
if not system_proxy in available_proxies:
system_proxy = None
default['requests']['proxy'] = system_proxy if system_proxy is not None else datastore.proxy_list[0][0]
default['requests']['proxy'] = system_proxy if system_proxy is not None else available_proxies[0]
# Used by the form handler to keep or remove the proxy settings
default['proxy_list'] = datastore.proxy_list
default['proxy_list'] = available_proxies[0]
# Don't use form.data on POST so that it doesnt overrid the checkbox status from the POST status
@ -680,7 +685,10 @@ def changedetection_app(config=None, datastore_o=None):
# @todo - Couldn't get setattr() etc dynamic addition working, so remove it instead
del form.requests.form.proxy
else:
form.requests.form.proxy.choices = datastore.proxy_list
form.requests.form.proxy.choices = []
for p in datastore.proxy_list:
form.requests.form.proxy.choices.append(tuple((p, datastore.proxy_list[p]['label'])))
if request.method == 'POST':
# Password unset is a GET, but we can lock the session to a salted env password to always need the password
@ -1189,7 +1197,7 @@ def changedetection_app(config=None, datastore_o=None):
datastore.delete(uuid.strip())
flash("{} watches deleted".format(len(uuids)))
if (op == 'pause'):
elif (op == 'pause'):
for uuid in uuids:
uuid = uuid.strip()
if datastore.data['watching'].get(uuid):
@ -1197,13 +1205,40 @@ def changedetection_app(config=None, datastore_o=None):
flash("{} watches paused".format(len(uuids)))
if (op == 'unpause'):
elif (op == 'unpause'):
for uuid in uuids:
uuid = uuid.strip()
if datastore.data['watching'].get(uuid):
datastore.data['watching'][uuid.strip()]['paused'] = False
flash("{} watches unpaused".format(len(uuids)))
elif (op == 'mute'):
for uuid in uuids:
uuid = uuid.strip()
if datastore.data['watching'].get(uuid):
datastore.data['watching'][uuid.strip()]['notification_muted'] = True
flash("{} watches muted".format(len(uuids)))
elif (op == 'unmute'):
for uuid in uuids:
uuid = uuid.strip()
if datastore.data['watching'].get(uuid):
datastore.data['watching'][uuid.strip()]['notification_muted'] = False
flash("{} watches un-muted".format(len(uuids)))
elif (op == 'notification-default'):
from changedetectionio.notification import (
default_notification_format_for_watch
)
for uuid in uuids:
uuid = uuid.strip()
if datastore.data['watching'].get(uuid):
datastore.data['watching'][uuid.strip()]['notification_title'] = None
datastore.data['watching'][uuid.strip()]['notification_body'] = None
datastore.data['watching'][uuid.strip()]['notification_urls'] = []
datastore.data['watching'][uuid.strip()]['notification_format'] = default_notification_format_for_watch
flash("{} watches set to use default notification settings".format(len(uuids)))
return redirect(url_for('index'))
@app.route("/api/share-url", methods=['GET'])
@ -1341,6 +1376,8 @@ def ticker_thread_check_time_launch_checks():
import random
from changedetectionio import update_worker
proxy_last_called_time = {}
recheck_time_minimum_seconds = int(os.getenv('MINIMUM_SECONDS_RECHECK_TIME', 20))
print("System env MINIMUM_SECONDS_RECHECK_TIME", recheck_time_minimum_seconds)
@ -1401,10 +1438,30 @@ def ticker_thread_check_time_launch_checks():
if watch.jitter_seconds == 0:
watch.jitter_seconds = random.uniform(-abs(jitter), jitter)
seconds_since_last_recheck = now - watch['last_checked']
if seconds_since_last_recheck >= (threshold + watch.jitter_seconds) and seconds_since_last_recheck >= recheck_time_minimum_seconds:
if not uuid in running_uuids and uuid not in [q_uuid for p,q_uuid in update_q.queue]:
# Proxies can be set to have a limit on seconds between which they can be called
watch_proxy = datastore.get_preferred_proxy_for_watch(uuid=uuid)
if watch_proxy and watch_proxy in list(datastore.proxy_list.keys()):
# Proxy may also have some threshold minimum
proxy_list_reuse_time_minimum = int(datastore.proxy_list.get(watch_proxy, {}).get('reuse_time_minimum', 0))
if proxy_list_reuse_time_minimum:
proxy_last_used_time = proxy_last_called_time.get(watch_proxy, 0)
time_since_proxy_used = int(time.time() - proxy_last_used_time)
if time_since_proxy_used < proxy_list_reuse_time_minimum:
# Not enough time difference reached, skip this watch
print("> Skipped UUID {} using proxy '{}', not enough time between proxy requests {}s/{}s".format(uuid,
watch_proxy,
time_since_proxy_used,
proxy_list_reuse_time_minimum))
continue
else:
# Record the last used time
proxy_last_called_time[watch_proxy] = int(time.time())
# Use Epoch time as priority, so we get a "sorted" PriorityQueue, but we can still push a priority 1 into it.
priority = int(time.time())
print(

@ -316,6 +316,7 @@ class base_html_playwright(Fetcher):
import playwright._impl._api_types
from playwright._impl._api_types import Error, TimeoutError
response = None
with sync_playwright() as p:
browser_type = getattr(p, self.browser_type)
@ -373,8 +374,11 @@ class base_html_playwright(Fetcher):
print("response object was none")
raise EmptyReply(url=url, status_code=None)
# Bug 2(?) Set the viewport size AFTER loading the page
page.set_viewport_size({"width": 1280, "height": 1024})
# Removed browser-set-size, seemed to be needed to make screenshots work reliably in older playwright versions
# Was causing exceptions like 'waiting for page but content is changing' etc
# https://www.browserstack.com/docs/automate/playwright/change-browser-window-size 1280x720 should be the default
extra_wait = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay
time.sleep(extra_wait)
@ -398,6 +402,13 @@ class base_html_playwright(Fetcher):
raise JSActionExceptions(status_code=response.status, screenshot=error_screenshot, message=str(e), url=url)
else:
# JS eval was run, now we also wait some time if possible to let the page settle
if self.render_extract_delay:
page.wait_for_timeout(self.render_extract_delay * 1000)
page.wait_for_timeout(500)
self.content = page.content()
self.status_code = response.status
self.headers = response.all_headers()
@ -514,8 +525,6 @@ class base_html_webdriver(Fetcher):
# Selenium doesn't automatically wait for actions as good as Playwright, so wait again
self.driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
self.screenshot = self.driver.get_screenshot_as_png()
# @todo - how to check this? is it possible?
self.status_code = 200
# @todo somehow we should try to get this working for WebDriver
@ -526,6 +535,8 @@ class base_html_webdriver(Fetcher):
self.content = self.driver.page_source
self.headers = {}
self.screenshot = self.driver.get_screenshot_as_png()
# Does the connection to the webdriver work? run a test connection.
def is_ready(self):
from selenium import webdriver
@ -564,6 +575,11 @@ class html_requests(Fetcher):
ignore_status_codes=False,
current_css_filter=None):
# Make requests use a more modern looking user-agent
if not 'User-Agent' in request_headers:
request_headers['User-Agent'] = os.getenv("DEFAULT_SETTINGS_HEADERS_USERAGENT",
'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.66 Safari/537.36')
proxies = {}
# Allows override the proxy on a per-request basis

@ -21,34 +21,6 @@ class perform_site_check():
super().__init__(*args, **kwargs)
self.datastore = datastore
# If there was a proxy list enabled, figure out what proxy_args/which proxy to use
# if watch.proxy use that
# fetcher.proxy_override = watch.proxy or main config proxy
# Allows override the proxy on a per-request basis
# ALWAYS use the first one is nothing selected
def set_proxy_from_list(self, watch):
proxy_args = None
if self.datastore.proxy_list is None:
return None
# If its a valid one
if any([watch['proxy'] in p for p in self.datastore.proxy_list]):
proxy_args = watch['proxy']
# not valid (including None), try the system one
else:
system_proxy = self.datastore.data['settings']['requests']['proxy']
# Is not None and exists
if any([system_proxy in p for p in self.datastore.proxy_list]):
proxy_args = system_proxy
# Fallback - Did not resolve anything, use the first available
if proxy_args is None:
proxy_args = self.datastore.proxy_list[0][0]
return proxy_args
# Doesn't look like python supports forward slash auto enclosure in re.findall
# So convert it to inline flag "foobar(?i)" type configuration
def forward_slash_enclosed_regex_to_options(self, regex):
@ -69,6 +41,8 @@ class perform_site_check():
stripped_text_from_html = ""
watch = self.datastore.data['watching'].get(uuid)
if not watch:
return
# Protect against file:// access
if re.search(r'^file', watch['url'], re.IGNORECASE) and not os.getenv('ALLOW_FILE_URI', False):
@ -91,7 +65,7 @@ class perform_site_check():
if 'Accept-Encoding' in request_headers and "br" in request_headers['Accept-Encoding']:
request_headers['Accept-Encoding'] = request_headers['Accept-Encoding'].replace(', br', '')
timeout = self.datastore.data['settings']['requests']['timeout']
timeout = self.datastore.data['settings']['requests'].get('timeout')
url = watch.get('url')
request_body = self.datastore.data['watching'][uuid].get('body')
request_method = self.datastore.data['watching'][uuid].get('method')
@ -111,9 +85,13 @@ class perform_site_check():
# If the klass doesnt exist, just use a default
klass = getattr(content_fetcher, "html_requests")
proxy_id = self.datastore.get_preferred_proxy_for_watch(uuid=uuid)
proxy_url = None
if proxy_id:
proxy_url = self.datastore.proxy_list.get(proxy_id).get('url')
print ("UUID {} Using proxy {}".format(uuid, proxy_url))
proxy_args = self.set_proxy_from_list(watch)
fetcher = klass(proxy_override=proxy_args)
fetcher = klass(proxy_override=proxy_url)
# Configurable per-watch or global extra delay before extracting text (for webDriver types)
system_webdriver_delay = self.datastore.data['settings']['application'].get('webdriver_delay', None)
@ -164,8 +142,9 @@ class perform_site_check():
has_filter_rule = True
if has_filter_rule:
if 'json:' in css_filter_rule:
stripped_text_from_html = html_tools.extract_json_as_string(content=fetcher.content, jsonpath_filter=css_filter_rule)
json_filter_prefixes = ['json:', 'jq:']
if any(prefix in css_filter_rule for prefix in json_filter_prefixes):
stripped_text_from_html = html_tools.extract_json_as_string(content=fetcher.content, json_filter=css_filter_rule)
is_html = False
if is_html or is_source:

@ -304,6 +304,21 @@ class ValidateCSSJSONXPATHInput(object):
# Re #265 - maybe in the future fetch the page and offer a
# warning/notice that its possible the rule doesnt yet match anything?
if 'jq:' in line:
if not self.allow_json:
raise ValidationError("jq not permitted in this field!")
import jq
input = line.replace('jq:', '')
try:
jq.compile(input)
except (ValueError) as e:
message = field.gettext('\'%s\' is not a valid jq expression. (%s)')
raise ValidationError(message % (input, str(e)))
except:
raise ValidationError("A system-error occurred when validating your jq expression")
class ValidateDiffFilters(object):
"""
Validates that at least one filter checkbox is selected
@ -316,6 +331,7 @@ class ValidateDiffFilters(object):
message = field.gettext('At least one filter checkbox must be selected')
raise ValidationError(message)
class quickWatchForm(Form):
url = fields.URLField('URL', validators=[validateURL()])
tag = StringField('Group tag', [validators.Optional()])
@ -325,14 +341,14 @@ class quickWatchForm(Form):
# Common to a single watch and the global settings
class commonSettingsForm(Form):
notification_urls = StringListField('Notification URL list', validators=[validators.Optional(), ValidateAppRiseServers()])
notification_title = StringField('Notification title', default=default_notification_title, validators=[validators.Optional(), ValidateTokensList()])
notification_body = TextAreaField('Notification body', default=default_notification_body, validators=[validators.Optional(), ValidateTokensList()])
notification_format = SelectField('Notification format', choices=valid_notification_formats.keys(), default=default_notification_format)
notification_title = StringField('Notification title', validators=[validators.Optional(), ValidateTokensList()])
notification_body = TextAreaField('Notification body', validators=[validators.Optional(), ValidateTokensList()])
notification_format = SelectField('Notification format', choices=valid_notification_formats.keys())
fetch_backend = RadioField(u'Fetch method', choices=content_fetcher.available_fetchers(), validators=[ValidateContentFetcherIsReady()])
extract_title_as_title = BooleanField('Extract <title> from document and use as watch title', default=False)
webdriver_delay = IntegerField('Wait seconds before extracting text', validators=[validators.Optional(), validators.NumberRange(min=1, message="Should contain one or more seconds")] )
webdriver_delay = IntegerField('Wait seconds before extracting text', validators=[validators.Optional(), validators.NumberRange(min=1,
message="Should contain one or more seconds")])
class watchForm(commonSettingsForm):

@ -3,6 +3,7 @@ from typing import List
from bs4 import BeautifulSoup
from jsonpath_ng.ext import parse
import jq
import re
from inscriptis import get_text
from inscriptis.model.config import ParserConfig
@ -79,19 +80,26 @@ def extract_element(find='title', html_content=''):
return element_text
#
def _parse_json(json_data, jsonpath_filter):
s=[]
jsonpath_expression = parse(jsonpath_filter.replace('json:', ''))
match = jsonpath_expression.find(json_data)
def _parse_json(json_data, json_filter):
if 'json:' in json_filter:
jsonpath_expression = parse(json_filter.replace('json:', ''))
match = jsonpath_expression.find(json_data)
return _get_stripped_text_from_json_match(match)
if 'jq:' in json_filter:
jq_expression = jq.compile(json_filter.replace('jq:', ''))
match = jq_expression.input(json_data).all()
return _get_stripped_text_from_json_match(match)
def _get_stripped_text_from_json_match(match):
s = []
# More than one result, we will return it as a JSON list.
if len(match) > 1:
for i in match:
s.append(i.value)
s.append(i.value if hasattr(i, 'value') else i)
# Single value, use just the value, as it could be later used in a token in notifications.
if len(match) == 1:
s = match[0].value
s = match[0].value if hasattr(match[0], 'value') else match[0]
# Re #257 - Better handling where it does not exist, in the case the original 's' value was False..
if not match:
@ -103,16 +111,16 @@ def _parse_json(json_data, jsonpath_filter):
return stripped_text_from_html
def extract_json_as_string(content, jsonpath_filter):
def extract_json_as_string(content, json_filter):
stripped_text_from_html = False
# Try to parse/filter out the JSON, if we get some parser error, then maybe it's embedded <script type=ldjson>
try:
stripped_text_from_html = _parse_json(json.loads(content), jsonpath_filter)
stripped_text_from_html = _parse_json(json.loads(content), json_filter)
except json.JSONDecodeError:
# Foreach <script json></script> blob.. just return the first that matches jsonpath_filter
# Foreach <script json></script> blob.. just return the first that matches json_filter
s = []
soup = BeautifulSoup(content, 'html.parser')
bs_result = soup.findAll('script')
@ -131,7 +139,7 @@ def extract_json_as_string(content, jsonpath_filter):
# Just skip it
continue
else:
stripped_text_from_html = _parse_json(json_data, jsonpath_filter)
stripped_text_from_html = _parse_json(json_data, json_filter)
if stripped_text_from_html:
break

@ -13,10 +13,6 @@ class model(dict):
'watching': {},
'settings': {
'headers': {
'User-Agent': getenv("DEFAULT_SETTINGS_HEADERS_USERAGENT", 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.66 Safari/537.36'),
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9',
'Accept-Encoding': 'gzip, deflate', # No support for brolti in python requests yet.
'Accept-Language': 'en-GB,en-US;q=0.9,en;'
},
'requests': {
'timeout': int(getenv("DEFAULT_SETTINGS_REQUESTS_TIMEOUT", "45")), # Default 45 seconds

@ -49,3 +49,47 @@ pytest tests/visualselector/test_fetch_data.py
unset PLAYWRIGHT_DRIVER_URL
docker kill $$-test_browserless
# Test proxy list handling, starting two squids on different ports
# Each squid adds a different header to the response, which is the main thing we test for.
docker run -d --name $$-squid-one --rm -v `pwd`/tests/proxy_list/squid.conf:/etc/squid/conf.d/debian.conf -p 3128:3128 ubuntu/squid:4.13-21.10_edge
docker run -d --name $$-squid-two --rm -v `pwd`/tests/proxy_list/squid.conf:/etc/squid/conf.d/debian.conf -p 3129:3128 ubuntu/squid:4.13-21.10_edge
# So, basic HTTP as env var test
export HTTP_PROXY=http://localhost:3128
export HTTPS_PROXY=http://localhost:3128
pytest tests/proxy_list/test_proxy.py
docker logs $$-squid-one 2>/dev/null|grep one.changedetection.io
if [ $? -ne 0 ]
then
echo "Did not see a request to one.changedetection.io in the squid logs (while checking env vars HTTP_PROXY/HTTPS_PROXY)"
fi
unset HTTP_PROXY
unset HTTPS_PROXY
# 2nd test actually choose the preferred proxy from proxies.json
cp tests/proxy_list/proxies.json-example ./test-datastore/proxies.json
# Makes a watch use a preferred proxy
pytest tests/proxy_list/test_multiple_proxy.py
# Should be a request in the default "first" squid
docker logs $$-squid-one 2>/dev/null|grep chosen.changedetection.io
if [ $? -ne 0 ]
then
echo "Did not see a request to chosen.changedetection.io in the squid logs (while checking preferred proxy)"
fi
# And one in the 'second' squid (user selects this as preferred)
docker logs $$-squid-two 2>/dev/null|grep chosen.changedetection.io
if [ $? -ne 0 ]
then
echo "Did not see a request to chosen.changedetection.io in the squid logs (while checking preferred proxy)"
fi
# @todo - test system override proxy selection and watch defaults, setup a 3rd squid?
docker kill $$-squid-one
docker kill $$-squid-two

@ -30,4 +30,11 @@ $(document).ready(function() {
});
toggle();
$('#notification-setting-reset-to-default').click(function (e) {
$('#notification_title').val('');
$('#notification_body').val('');
$('#notification_format').val('System default');
$('#notification_urls').val('');
e.preventDefault();
});
});

@ -81,8 +81,6 @@ class ChangeDetectionStore:
except (FileNotFoundError, json.decoder.JSONDecodeError):
if include_default_watches:
print("Creating JSON store at", self.datastore_path)
self.add_watch(url='http://www.quotationspage.com/random.php', tag='test')
self.add_watch(url='https://news.ycombinator.com/', tag='Tech news')
self.add_watch(url='https://changedetection.io/CHANGELOG.txt', tag='changedetection.io')
@ -113,9 +111,7 @@ class ChangeDetectionStore:
self.__data['settings']['application']['api_access_token'] = secret
# Proxy list support - available as a selection in settings when text file is imported
# CSV list
# "name, address", or just "name"
proxy_list_file = "{}/proxies.txt".format(self.datastore_path)
proxy_list_file = "{}/proxies.json".format(self.datastore_path)
if path.isfile(proxy_list_file):
self.import_proxy_list(proxy_list_file)
@ -437,19 +433,41 @@ class ChangeDetectionStore:
unlink(item)
def import_proxy_list(self, filename):
import csv
with open(filename, newline='') as f:
reader = csv.reader(f, skipinitialspace=True)
# @todo This loop can could be improved
l = []
for row in reader:
if len(row):
if len(row)>=2:
l.append(tuple(row[:2]))
else:
l.append(tuple([row[0], row[0]]))
self.proxy_list = l if len(l) else None
with open(filename) as f:
self.proxy_list = json.load(f)
print ("Registered proxy list", list(self.proxy_list.keys()))
def get_preferred_proxy_for_watch(self, uuid):
"""
Returns the preferred proxy by ID key
:param uuid: UUID
:return: proxy "key" id
"""
proxy_id = None
if self.proxy_list is None:
return None
# If its a valid one
watch = self.data['watching'].get(uuid)
if watch.get('proxy') and watch.get('proxy') in list(self.proxy_list.keys()):
return watch.get('proxy')
# not valid (including None), try the system one
else:
system_proxy_id = self.data['settings']['requests'].get('proxy')
# Is not None and exists
if self.proxy_list.get(system_proxy_id):
return system_proxy_id
# Fallback - Did not resolve anything, use the first available
if system_proxy_id is None:
first_default = list(self.proxy_list)[0]
return first_default
return None
# Run all updates
# IMPORTANT - Each update could be run even when they have a new install and the schema is correct
@ -541,8 +559,37 @@ class ChangeDetectionStore:
continue
return
# Generate a previous.txt for all watches that do not have one and contain history
def update_5(self):
# If the watch notification body, title look the same as the global one, unset it, so the watch defaults back to using the main settings
# In other words - the watch notification_title and notification_body are not needed if they are the same as the default one
current_system_body = self.data['settings']['application']['notification_body'].translate(str.maketrans('', '', "\r\n "))
current_system_title = self.data['settings']['application']['notification_body'].translate(str.maketrans('', '', "\r\n "))
for uuid, watch in self.data['watching'].items():
try:
watch_body = watch.get('notification_body', '')
if watch_body and watch_body.translate(str.maketrans('', '', "\r\n ")) == current_system_body:
# Looks the same as the default one, so unset it
watch['notification_body'] = None
watch_title = watch.get('notification_title', '')
if watch_title and watch_title.translate(str.maketrans('', '', "\r\n ")) == current_system_title:
# Looks the same as the default one, so unset it
watch['notification_title'] = None
except Exception as e:
continue
return
# We incorrectly used common header overrides that should only apply to Requests
# These are now handled in content_fetcher::html_requests and shouldnt be passed to Playwright/Selenium
def update_6(self):
# These were hard-coded in early versions
for v in ['User-Agent', 'Accept', 'Accept-Encoding', 'Accept-Language']:
if self.data['settings']['headers'].get(v):
del self.data['settings']['headers'][v]
# Generate a previous.txt for all watches that do not have one and contain history
def update_7(self):
for uuid, watch in self.data['watching'].items():
# Make sure we actually have history
if (watch.history_n == 0):
@ -559,3 +606,4 @@ class ChangeDetectionStore:
with open(latest_file_name, "rb") as f2:
f.write(f2.read())

@ -77,6 +77,7 @@
<span class="pure-form-message-inline">
<p>Use the <strong>Basic</strong> method (default) where your watched site doesn't need Javascript to render.</p>
<p>The <strong>Chrome/Javascript</strong> method requires a network connection to a running WebDriver+Chrome server, set by the ENV var 'WEBDRIVER_URL'. </p>
Tip: <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Proxy-configuration#brightdata-proxy-support">Connect using BrightData Proxies, find out more here.</a>
</span>
</div>
{% if form.proxy %}
@ -146,6 +147,8 @@ User-Agent: wonderbra 1.0") }}
There are <a href="{{ url_for('settings_page')}}#notifications">system-wide notification URLs enabled</a>, this form will override notification settings for this watch only &dash; an empty Notification URL list here will still send notifications.
</div>
{% endif %}
<a href="#notifications" id="notification-setting-reset-to-default" class="pure-button button-xsmall" style="right: 20px; top: 20px; position: absolute; background-color: #5f42dd; border-radius: 4px; font-size: 70%; color: #fff">Use system defaults</a>
{{ render_common_settings_form(form, emailprefix, settings_application) }}
</div>
</fieldset>
@ -191,8 +194,12 @@ User-Agent: wonderbra 1.0") }}
<span class="pure-form-message-inline">
<ul>
<li>CSS - Limit text to this CSS rule, only text matching this CSS rule is included.</li>
<li>JSON - Limit text to this JSON rule, using <a href="https://pypi.org/project/jsonpath-ng/">JSONPath</a>, prefix with <code>"json:"</code>, use <code>json:$</code> to force re-formatting if required, <a
href="https://jsonpath.com/" target="new">test your JSONPath here</a></li>
<li>JSON - Limit text to this JSON rule, using either <a href="https://pypi.org/project/jsonpath-ng/" target="new">JSONPath</a> or <a href="https://stedolan.github.io/jq/" target="new">jq</a>.
<ul>
<li>JSONPath: Prefix with <code>json:</code>, use <code>json:$</code> to force re-formatting if required, <a href="https://jsonpath.com/" target="new">test your JSONPath here</a>.</li>
<li>jq: Prefix with <code>jq:</code> and <a href="https://jqplay.org/" target="new">test your jq here</a>. Using <a href="https://stedolan.github.io/jq/" target="new">jq</a> allows for complex filtering and processing of JSON data with built-in functions, regex, filtering, and more. See examples and documentation <a href="https://stedolan.github.io/jq/manual/" target="new">here</a>.</li>
</ul>
</li>
<li>XPath - Limit text to this XPath rule, simply start with a forward-slash,
<ul>
<li>Example: <code>//*[contains(@class, 'sametext')]</code> or <code>xpath://*[contains(@class, 'sametext')]</code>, <a
@ -201,7 +208,7 @@ User-Agent: wonderbra 1.0") }}
</ul>
</li>
</ul>
Please be sure that you thoroughly understand how to write CSS or JSONPath, XPath selector rules before filing an issue on GitHub! <a
Please be sure that you thoroughly understand how to write CSS, JSONPath, XPath, or jq selector rules before filing an issue on GitHub! <a
href="https://github.com/dgtlmoon/changedetection.io/wiki/CSS-Selector-help">here for more CSS selector help</a>.<br/>
</span>
</div>

@ -99,6 +99,8 @@
<p>Use the <strong>Basic</strong> method (default) where your watched sites don't need Javascript to render.</p>
<p>The <strong>Chrome/Javascript</strong> method requires a network connection to a running WebDriver+Chrome server, set by the ENV var 'WEBDRIVER_URL'. </p>
</span>
<br/>
Tip: <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Proxy-configuration#brightdata-proxy-support">Connect using BrightData Proxies, find out more here.</a>
</div>
<fieldset class="pure-group" id="webdriver-override-options">
<div class="pure-form-message-inline">

@ -30,6 +30,9 @@
<div id="checkbox-operations">
<button class="pure-button button-secondary button-xsmall" style="font-size: 70%" name="op" value="pause">Pause</button>
<button class="pure-button button-secondary button-xsmall" style="font-size: 70%" name="op" value="unpause">UnPause</button>
<button class="pure-button button-secondary button-xsmall" style="font-size: 70%" name="op" value="mute">Mute</button>
<button class="pure-button button-secondary button-xsmall" style="font-size: 70%" name="op" value="unmute">UnMute</button>
<button class="pure-button button-secondary button-xsmall" style="font-size: 70%" name="op" value="notification-default">Use default notification</button>
<button class="pure-button button-secondary button-xsmall" style="background: #dd4242; font-size: 70%" name="op" value="delete">Delete</button>
</div>
<div>

@ -0,0 +1,14 @@
#!/usr/bin/python3
from .. import conftest
#def pytest_addoption(parser):
# parser.addoption("--url_suffix", action="store", default="identifier for request")
#def pytest_generate_tests(metafunc):
# # This is called for every test. Only get/set command line arguments
# # if the argument is specified in the list of test "fixturenames".
# option_value = metafunc.config.option.url_suffix
# if 'url_suffix' in metafunc.fixturenames and option_value is not None:
# metafunc.parametrize("url_suffix", [option_value])

@ -0,0 +1,10 @@
{
"proxy-one": {
"label": "One",
"url": "http://127.0.0.1:3128"
},
"proxy-two": {
"label": "two",
"url": "http://127.0.0.1:3129"
}
}

@ -0,0 +1,41 @@
acl localnet src 0.0.0.1-0.255.255.255 # RFC 1122 "this" network (LAN)
acl localnet src 10.0.0.0/8 # RFC 1918 local private network (LAN)
acl localnet src 100.64.0.0/10 # RFC 6598 shared address space (CGN)
acl localnet src 169.254.0.0/16 # RFC 3927 link-local (directly plugged) machines
acl localnet src 172.16.0.0/12 # RFC 1918 local private network (LAN)
acl localnet src 192.168.0.0/16 # RFC 1918 local private network (LAN)
acl localnet src fc00::/7 # RFC 4193 local private network range
acl localnet src fe80::/10 # RFC 4291 link-local (directly plugged) machines
acl localnet src 159.65.224.174
acl SSL_ports port 443
acl Safe_ports port 80 # http
acl Safe_ports port 21 # ftp
acl Safe_ports port 443 # https
acl Safe_ports port 70 # gopher
acl Safe_ports port 210 # wais
acl Safe_ports port 1025-65535 # unregistered ports
acl Safe_ports port 280 # http-mgmt
acl Safe_ports port 488 # gss-http
acl Safe_ports port 591 # filemaker
acl Safe_ports port 777 # multiling http
acl CONNECT method CONNECT
http_access deny !Safe_ports
http_access deny CONNECT !SSL_ports
http_access allow localhost manager
http_access deny manager
http_access allow localhost
http_access allow localnet
http_access deny all
http_port 3128
coredump_dir /var/spool/squid
refresh_pattern ^ftp: 1440 20% 10080
refresh_pattern ^gopher: 1440 0% 1440
refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
refresh_pattern \/(Packages|Sources)(|\.bz2|\.gz|\.xz)$ 0 0% 0 refresh-ims
refresh_pattern \/Release(|\.gpg)$ 0 0% 0 refresh-ims
refresh_pattern \/InRelease$ 0 0% 0 refresh-ims
refresh_pattern \/(Translation-.*)(|\.bz2|\.gz|\.xz)$ 0 0% 0 refresh-ims
refresh_pattern . 0 20% 4320
logfile_rotate 0

@ -0,0 +1,38 @@
#!/usr/bin/python3
import time
from flask import url_for
from ..util import live_server_setup
def test_preferred_proxy(client, live_server):
time.sleep(1)
live_server_setup(live_server)
time.sleep(1)
url = "http://chosen.changedetection.io"
res = client.post(
url_for("import_page"),
# Because a URL wont show in squid/proxy logs due it being SSLed
# Use plain HTTP or a specific domain-name here
data={"urls": url},
follow_redirects=True
)
assert b"1 Imported" in res.data
time.sleep(2)
res = client.post(
url_for("edit_page", uuid="first"),
data={
"css_filter": "",
"fetch_backend": "html_requests",
"headers": "",
"proxy": "proxy-two",
"tag": "",
"url": url,
},
follow_redirects=True
)
assert b"Updated watch." in res.data
time.sleep(2)
# Now the request should appear in the second-squid logs

@ -0,0 +1,19 @@
#!/usr/bin/python3
import time
from flask import url_for
from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
# just make a request, we will grep in the docker logs to see it actually got called
def test_check_basic_change_detection_functionality(client, live_server):
live_server_setup(live_server)
res = client.post(
url_for("import_page"),
# Because a URL wont show in squid/proxy logs due it being SSLed
# Use plain HTTP or a specific domain-name here
data={"urls": "http://one.changedetection.io"},
follow_redirects=True
)
assert b"1 Imported" in res.data
time.sleep(3)

@ -2,7 +2,7 @@
# coding=utf-8
import time
from flask import url_for
from flask import url_for, escape
from . util import live_server_setup
import pytest
@ -36,16 +36,26 @@ and it can also be repeated
from .. import html_tools
# See that we can find the second <script> one, which is not broken, and matches our filter
text = html_tools.extract_json_as_string(content, "$.offers.price")
text = html_tools.extract_json_as_string(content, "json:$.offers.price")
assert text == "23.5"
text = html_tools.extract_json_as_string('{"id":5}', "$.id")
# also check for jq
text = html_tools.extract_json_as_string(content, "jq:.offers.price")
assert text == "23.5"
text = html_tools.extract_json_as_string('{"id":5}', "json:$.id")
assert text == "5"
text = html_tools.extract_json_as_string('{"id":5}', "jq:.id")
assert text == "5"
# When nothing at all is found, it should throw JSONNOTFound
# Which is caught and shown to the user in the watch-overview table
with pytest.raises(html_tools.JSONNotFound) as e_info:
html_tools.extract_json_as_string('COMPLETE GIBBERISH, NO JSON!', "$.id")
html_tools.extract_json_as_string('COMPLETE GIBBERISH, NO JSON!', "json:$.id")
with pytest.raises(html_tools.JSONNotFound) as e_info:
html_tools.extract_json_as_string('COMPLETE GIBBERISH, NO JSON!', "jq:.id")
def set_original_ext_response():
data = """
@ -66,6 +76,7 @@ def set_original_ext_response():
with open("test-datastore/endpoint-content.txt", "w") as f:
f.write(data)
return None
def set_modified_ext_response():
data = """
@ -86,6 +97,7 @@ def set_modified_ext_response():
with open("test-datastore/endpoint-content.txt", "w") as f:
f.write(data)
return None
def set_original_response():
test_return_data = """
@ -184,10 +196,10 @@ def test_check_json_without_filter(client, live_server):
assert b'&#34;&lt;b&gt;' in res.data
assert res.data.count(b'{\n') >= 2
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
assert b'Deleted' in res.data
def test_check_json_filter(client, live_server):
json_filter = 'json:boss.name'
def check_json_filter(json_filter, client, live_server):
set_original_response()
# Give the endpoint time to spin up
@ -226,7 +238,7 @@ def test_check_json_filter(client, live_server):
res = client.get(
url_for("edit_page", uuid="first"),
)
assert bytes(json_filter.encode('utf-8')) in res.data
assert bytes(escape(json_filter).encode('utf-8')) in res.data
# Trigger a check
client.get(url_for("form_watch_checknow"), follow_redirects=True)
@ -252,10 +264,16 @@ def test_check_json_filter(client, live_server):
# And #462 - check we see the proper utf-8 string there
assert "Örnsköldsvik".encode('utf-8') in res.data
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
assert b'Deleted' in res.data
def test_check_jsonpath_filter(client, live_server):
check_json_filter('json:boss.name', client, live_server)
def test_check_json_filter_bool_val(client, live_server):
json_filter = "json:$['available']"
def test_check_jq_filter(client, live_server):
check_json_filter('jq:.boss.name', client, live_server)
def check_json_filter_bool_val(json_filter, client, live_server):
set_original_response()
# Give the endpoint time to spin up
@ -304,14 +322,21 @@ def test_check_json_filter_bool_val(client, live_server):
# But the change should be there, tho its hard to test the change was detected because it will show old and new versions
assert b'false' in res.data
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
assert b'Deleted' in res.data
def test_check_jsonpath_filter_bool_val(client, live_server):
check_json_filter_bool_val("json:$['available']", client, live_server)
def test_check_jq_filter_bool_val(client, live_server):
check_json_filter_bool_val("jq:.available", client, live_server)
# Re #265 - Extended JSON selector test
# Stuff to consider here
# - Selector should be allowed to return empty when it doesnt match (people might wait for some condition)
# - The 'diff' tab could show the old and new content
# - Form should let us enter a selector that doesnt (yet) match anything
def test_check_json_ext_filter(client, live_server):
json_filter = 'json:$[?(@.status==Sold)]'
def check_json_ext_filter(json_filter, client, live_server):
set_original_ext_response()
# Give the endpoint time to spin up
@ -350,7 +375,7 @@ def test_check_json_ext_filter(client, live_server):
res = client.get(
url_for("edit_page", uuid="first"),
)
assert bytes(json_filter.encode('utf-8')) in res.data
assert bytes(escape(json_filter).encode('utf-8')) in res.data
# Trigger a check
client.get(url_for("form_watch_checknow"), follow_redirects=True)
@ -376,3 +401,11 @@ def test_check_json_ext_filter(client, live_server):
assert b'ForSale' not in res.data
assert b'Sold' in res.data
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
assert b'Deleted' in res.data
def test_check_jsonpath_ext_filter(client, live_server):
check_json_ext_filter('json:$[?(@.status==Sold)]', client, live_server)
def test_check_jq_ext_filter(client, live_server):
check_json_ext_filter('jq:.[] | select(.status | contains("Sold"))', client, live_server)

@ -6,6 +6,8 @@ services:
hostname: changedetection
volumes:
- changedetection-data:/datastore
# Configurable proxy list support, see https://github.com/dgtlmoon/changedetection.io/wiki/Proxy-configuration#proxy-list-support
# - ./proxies.json:/datastore/proxies.json
# environment:
# Default listening port, can also be changed with the -p option
@ -30,7 +32,7 @@ services:
#
# https://playwright.dev/python/docs/api/class-browsertype#browser-type-launch-option-proxy
#
# Plain requsts - proxy support example.
# Plain requests - proxy support example.
# - HTTP_PROXY=socks5h://10.10.1.10:1080
# - HTTPS_PROXY=socks5h://10.10.1.10:1080
#

Binary file not shown.

After

Width:  |  Height:  |  Size: 46 KiB

@ -10,15 +10,19 @@ flask_restful
pytz
# Set these versions together to avoid a RequestsDependencyWarning
requests[socks] ~= 2.26
# >= 2.26 also adds Brotli support if brotli is installed
brotli ~= 1.0
requests[socks] ~= 2.28
urllib3 > 1.26
chardet > 2.3.0
wtforms ~= 3.0
jsonpath-ng ~= 1.5.3
jq ~= 1.3.0
# Notification library
apprise ~= 1.0.0
apprise ~= 1.1.0
# apprise mqtt https://github.com/dgtlmoon/changedetection.io/issues/315
paho-mqtt

Loading…
Cancel
Save