Merge branch 'dgtlmoon:master' into transplanting

pull/2351/head
Constantin Hong 4 months ago committed by GitHub
commit 23c6471744
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

@ -2,7 +2,7 @@
# Read more https://github.com/dgtlmoon/changedetection.io/wiki
__version__ = '0.46.02'
__version__ = '0.46.04'
from changedetectionio.strtobool import strtobool
from json.decoder import JSONDecodeError

@ -85,7 +85,8 @@ def construct_blueprint(datastore: ChangeDetectionStore):
browsersteps_start_session['browserstepper'] = browser_steps.browsersteps_live_ui(
playwright_browser=browsersteps_start_session['browser'],
proxy=proxy,
start_url=datastore.data['watching'][watch_uuid].get('url')
start_url=datastore.data['watching'][watch_uuid].get('url'),
headers=datastore.data['watching'][watch_uuid].get('headers')
)
# For test

@ -58,9 +58,9 @@ xpath://body/div/span[contains(@class, 'example-class')]",
{% if '/text()' in field %}
<span class="pure-form-message-inline"><strong>Note!: //text() function does not work where the &lt;element&gt; contains &lt;![CDATA[]]&gt;</strong></span><br>
{% endif %}
<span class="pure-form-message-inline">One rule per line, <i>any</i> rules that matches will be used.<br>
<ul>
<span class="pure-form-message-inline">One CSS, xPath, JSON Path/JQ selector per line, <i>any</i> rules that matches will be used.<br>
<div data-target="#advanced-help-selectors" class="toggle-show pure-button button-tag button-xsmall">Show advanced help and tips</div>
<ul id="advanced-help-selectors">
<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 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> (if installed).
<ul>

@ -65,8 +65,8 @@ class Fetcher():
def __init__(self):
import importlib.resources
self.xpath_element_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('xpath_element_scraper.js').read_text()
self.instock_data_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('stock-not-in-stock.js').read_text()
self.xpath_element_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('xpath_element_scraper.js').read_text(encoding='utf-8')
self.instock_data_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('stock-not-in-stock.js').read_text(encoding='utf-8')
@abstractmethod
def get_error(self):

@ -75,6 +75,7 @@ function isItemInStock() {
'vergriffen',
'vorbestellen',
'vorbestellung ist bald möglich',
'we don\'t currently have any',
'we couldn\'t find any products that match',
'we do not currently have an estimate of when this product will be back in stock.',
'we don\'t know when or if this item will be back in stock.',
@ -173,7 +174,8 @@ function isItemInStock() {
const element = elementsToScan[i];
// outside the 'fold' or some weird text in the heading area
// .getBoundingClientRect() was causing a crash in chrome 119, can only be run on contentVisibility != hidden
if (element.getBoundingClientRect().top + window.scrollY >= vh + 150 || element.getBoundingClientRect().top + window.scrollY <= 100) {
// Note: theres also an automated test that places the 'out of stock' text fairly low down
if (element.getBoundingClientRect().top + window.scrollY >= vh + 250 || element.getBoundingClientRect().top + window.scrollY <= 100) {
continue
}
elementText = "";
@ -187,7 +189,7 @@ function isItemInStock() {
// and these mean its out of stock
for (const outOfStockText of outOfStockTexts) {
if (elementText.includes(outOfStockText)) {
console.log(`Selected 'Out of Stock' - found text "${outOfStockText}" - "${elementText}"`)
console.log(`Selected 'Out of Stock' - found text "${outOfStockText}" - "${elementText}" - offset top ${element.getBoundingClientRect().top}, page height is ${vh}`)
return outOfStockText; // item is out of stock
}
}

@ -164,6 +164,15 @@ visibleElementsArray.forEach(function (element) {
}
}
let label = "not-interesting" // A placeholder, the actual labels for training are done by hand for now
let text = element.textContent.trim().slice(0, 30).trim();
while (/\n{2,}|\t{2,}/.test(text)) {
text = text.replace(/\n{2,}/g, '\n').replace(/\t{2,}/g, '\t')
}
// Try to identify any possible currency amounts "Sale: 4000" or "Sale now 3000 Kc", can help with the training.
const hasDigitCurrency = (/\d/.test(text.slice(0, 6)) || /\d/.test(text.slice(-6)) ) && /([€£$¥₩₹]|USD|AUD|EUR|Kč|kr|SEK|,)/.test(text) ;
size_pos.push({
xpath: xpath_result,
@ -171,9 +180,16 @@ visibleElementsArray.forEach(function (element) {
height: Math.round(bbox['height']),
left: Math.floor(bbox['left']),
top: Math.floor(bbox['top']) + scroll_y,
// tagName used by Browser Steps
tagName: (element.tagName) ? element.tagName.toLowerCase() : '',
// tagtype used by Browser Steps
tagtype: (element.tagName.toLowerCase() === 'input' && element.type) ? element.type.toLowerCase() : '',
isClickable: window.getComputedStyle(element).cursor == "pointer"
isClickable: window.getComputedStyle(element).cursor === "pointer",
// Used by the keras trainer
fontSize: window.getComputedStyle(element).getPropertyValue('font-size'),
fontWeight: window.getComputedStyle(element).getPropertyValue('font-weight'),
hasDigitCurrency: hasDigitCurrency,
label: label,
});
});

@ -1377,17 +1377,19 @@ def changedetection_app(config=None, datastore_o=None):
import brotli
watch = datastore.data['watching'].get(uuid)
if watch and os.path.isdir(watch.watch_data_dir):
latest_filename = list(watch.history.keys())[0]
if watch and watch.history.keys() and os.path.isdir(watch.watch_data_dir):
latest_filename = list(watch.history.keys())[-1]
html_fname = os.path.join(watch.watch_data_dir, f"{latest_filename}.html.br")
if html_fname.endswith('.br'):
# Read and decompress the Brotli file
with open(html_fname, 'rb') as f:
with open(html_fname, 'rb') as f:
if html_fname.endswith('.br'):
# Read and decompress the Brotli file
decompressed_data = brotli.decompress(f.read())
else:
decompressed_data = f.read()
buffer = BytesIO(decompressed_data)
buffer = BytesIO(decompressed_data)
return send_file(buffer, as_attachment=True, download_name=f"{latest_filename}.html", mimetype='text/html')
return send_file(buffer, as_attachment=True, download_name=f"{latest_filename}.html", mimetype='text/html')
# Return a 500 error

@ -1,11 +1,12 @@
from babel.numbers import parse_decimal
from changedetectionio.model.Watch import model as BaseWatch
from typing import Union
import re
from babel.numbers import parse_decimal
class Restock(dict):
def parse_currency(self, raw_value: str) -> float:
def parse_currency(self, raw_value: str) -> Union[float, None]:
# Clean and standardize the value (ie 1,400.00 should be 1400.00), even better would be store the whole thing as an integer.
standardized_value = raw_value
@ -21,8 +22,11 @@ class Restock(dict):
# Remove any non-numeric characters except for the decimal point
standardized_value = re.sub(r'[^\d.-]', '', standardized_value)
# Convert to float
return float(parse_decimal(standardized_value, locale='en'))
if standardized_value:
# Convert to float
return float(parse_decimal(standardized_value, locale='en'))
return None
def __init__(self, *args, **kwargs):
# Define default values

@ -40,13 +40,16 @@ def get_itemprop_availability(html_content) -> Restock:
import extruct
logger.trace(f"Imported extruct module in {time.time() - now:.3f}s")
value = {}
now = time.time()
# Extruct is very slow, I'm wondering if some ML is going to be faster (800ms on my i7), 'rdfa' seems to be the heaviest.
# Extruct is very slow, I'm wondering if some ML is going to be faster (800ms on my i7), 'rdfa' seems to be the heaviest.
syntaxes = ['dublincore', 'json-ld', 'microdata', 'microformat', 'opengraph']
try:
data = extruct.extract(html_content, syntaxes=syntaxes)
except Exception as e:
logger.warning(f"Unable to extract data, document parsing with extruct failed with {type(e).__name__} - {str(e)}")
return Restock()
data = extruct.extract(html_content, syntaxes=syntaxes)
logger.trace(f"Extruct basic extract of all metadata done in {time.time() - now:.3f}s")
# First phase, dead simple scanning of anything that looks useful

@ -18,9 +18,11 @@ $(document).ready(function () {
});
$("#notification-token-toggle").click(function (e) {
$(".toggle-show").click(function (e) {
e.preventDefault();
$('#notification-tokens-info').toggle();
let target = $(this).data('target');
$(target).toggle();
});
});

@ -11,8 +11,11 @@
class="notification-urls" )
}}
<div class="pure-form-message-inline">
<ul>
<li>Use <a target=_new href="https://github.com/caronc/apprise">AppRise URLs</a> for notification to just about any service! <i><a target=_new href="https://github.com/dgtlmoon/changedetection.io/wiki/Notification-configuration-notes">Please read the notification services wiki here for important configuration notes</a></i>.</li>
<p>
<strong>Tip:</strong> Use <a target=_new href="https://github.com/caronc/apprise">AppRise Notification URLs</a> for notification to just about any service! <i><a target=_new href="https://github.com/dgtlmoon/changedetection.io/wiki/Notification-configuration-notes">Please read the notification services wiki here for important configuration notes</a></i>.<br>
</p>
<div data-target="#advanced-help-notifications" class="toggle-show pure-button button-tag button-xsmall">Show advanced help and tips</div>
<ul style="display: none" id="advanced-help-notifications">
<li><code><a target=_new href="https://github.com/caronc/apprise/wiki/Notify_discord">discord://</a></code> (or <code>https://discord.com/api/webhooks...</code>)) only supports a maximum <strong>2,000 characters</strong> of notification text, including the title.</li>
<li><code><a target=_new href="https://github.com/caronc/apprise/wiki/Notify_telegram">tgram://</a></code> bots can't send messages to other bots, so you should specify chat ID of non-bot user.</li>
<li><code><a target=_new href="https://github.com/caronc/apprise/wiki/Notify_telegram">tgram://</a></code> only supports very limited HTML and can fail when extra tags are sent, <a href="https://core.telegram.org/bots/api#html-style">read more here</a> (or use plaintext/markdown format)</li>
@ -40,7 +43,7 @@
</div>
<div class="pure-controls">
<div id="notification-token-toggle" class="pure-button button-tag button-xsmall">Show token/placeholders</div>
<div data-target="#notification-tokens-info" class="toggle-show pure-button button-tag button-xsmall">Show token/placeholders</div>
</div>
<div class="pure-controls" style="display: none;" id="notification-tokens-info">
<table class="pure-table" id="token-table">

@ -4,6 +4,7 @@
{% from '_common_fields.html' import render_common_settings_form %}
<script src="{{url_for('static_content', group='js', filename='tabs.js')}}" defer></script>
<script src="{{url_for('static_content', group='js', filename='vis.js')}}" defer></script>
<script src="{{url_for('static_content', group='js', filename='global-settings.js')}}" defer></script>
<script>
const browser_steps_available_screenshots=JSON.parse('{{ watch.get_browsersteps_available_screenshots|tojson }}');
const browser_steps_config=JSON.parse('{{ browser_steps_config|tojson }}');
@ -275,9 +276,9 @@ xpath://body/div/span[contains(@class, 'example-class')]",
{% if '/text()' in field %}
<span class="pure-form-message-inline"><strong>Note!: //text() function does not work where the &lt;element&gt; contains &lt;![CDATA[]]&gt;</strong></span><br>
{% endif %}
<span class="pure-form-message-inline">One rule per line, <i>any</i> rules that matches will be used.<br>
<ul>
<span class="pure-form-message-inline">One CSS, xPath, JSON Path/JQ selector per line, <i>any</i> rules that matches will be used.<br>
<p><div data-target="#advanced-help-selectors" class="toggle-show pure-button button-tag button-xsmall">Show advanced help and tips</div><br></p>
<ul id="advanced-help-selectors" style="display: none;">
<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 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> (if installed).
<ul>
@ -297,9 +298,12 @@ xpath://body/div/span[contains(@class, 'example-class')]",
<li>To use XPath1.0: Prefix with <code>xpath1:</code></li>
</ul>
</li>
</ul>
Please be sure that you thoroughly understand how to write CSS, JSONPath, XPath{% if jq_support %}, or jq selector{%endif%} rules before filing an issue on GitHub! <a
<li>
Please be sure that you thoroughly understand how to write CSS, JSONPath, XPath{% if jq_support %}, or jq selector{%endif%} 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>
</li>
</ul>
</span>
</div>
<fieldset class="pure-control-group">

@ -2,7 +2,7 @@
import os
import time
from flask import url_for
from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client, wait_for_notification_endpoint_output
from changedetectionio.notification import (
default_notification_body,
default_notification_format,
@ -94,7 +94,7 @@ def test_restock_detection(client, live_server, measure_memory_usage):
assert b'not-in-stock' not in res.data
# We should have a notification
time.sleep(2)
wait_for_notification_endpoint_output()
assert os.path.isfile("test-datastore/notification.txt"), "Notification received"
os.unlink("test-datastore/notification.txt")
@ -103,6 +103,7 @@ def test_restock_detection(client, live_server, measure_memory_usage):
set_original_response()
client.get(url_for("form_watch_checknow"), follow_redirects=True)
wait_for_all_checks(client)
time.sleep(5)
assert not os.path.isfile("test-datastore/notification.txt"), "No notification should have fired when it went OUT OF STOCK by default"
# BUT we should see that it correctly shows "not in stock"

@ -2,7 +2,7 @@
import os.path
import time
from flask import url_for
from .util import live_server_setup, wait_for_all_checks
from .util import live_server_setup, wait_for_all_checks, wait_for_notification_endpoint_output
from changedetectionio import html_tools
@ -165,7 +165,7 @@ def test_check_add_line_contains_trigger(client, live_server, measure_memory_usa
assert b'unviewed' in res.data
# Takes a moment for apprise to fire
time.sleep(3)
wait_for_notification_endpoint_output()
assert os.path.isfile("test-datastore/notification.txt"), "Notification fired because I can see the output file"
with open("test-datastore/notification.txt", 'rb') as f:
response = f.read()

@ -69,6 +69,12 @@ def test_check_basic_change_detection_functionality(client, live_server, measure
wait_for_all_checks(client)
uuid = extract_UUID_from_client(client)
# Check the 'get latest snapshot works'
res = client.get(url_for("watch_get_latest_html", uuid=uuid))
assert b'which has this one new line' in res.data
# Now something should be ready, indicated by having a 'unviewed' class
res = client.get(url_for("index"))
assert b'unviewed' in res.data
@ -86,7 +92,7 @@ def test_check_basic_change_detection_functionality(client, live_server, measure
assert expected_url.encode('utf-8') in res.data
# Following the 'diff' link, it should no longer display as 'unviewed' even after we recheck it a few times
res = client.get(url_for("diff_history_page", uuid="first"))
res = client.get(url_for("diff_history_page", uuid=uuid))
assert b'selected=""' in res.data, "Confirm diff history page loaded"
# Check the [preview] pulls the right one
@ -143,18 +149,12 @@ def test_check_basic_change_detection_functionality(client, live_server, measure
assert b'unviewed' not in res.data
# #2458 "clear history" should make the Watch object update its status correctly when the first snapshot lands again
uuid = extract_UUID_from_client(client)
client.get(url_for("clear_watch_history", uuid=uuid))
client.get(url_for("form_watch_checknow"), follow_redirects=True)
wait_for_all_checks(client)
res = client.get(url_for("index"))
assert b'preview/' in res.data
# Check the 'get latest snapshot works'
res = client.get(url_for("watch_get_latest_html", uuid=uuid))
assert b'<head><title>head title</title></head>' in res.data
#
# Cleanup everything
res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)

@ -4,7 +4,7 @@
import os
import time
from flask import url_for
from .util import set_original_response, live_server_setup
from .util import set_original_response, live_server_setup, wait_for_notification_endpoint_output
from changedetectionio.model import App
@ -102,14 +102,15 @@ def test_filter_doesnt_exist_then_exists_should_get_notification(client, live_se
follow_redirects=True
)
assert b"Updated watch." in res.data
time.sleep(3)
wait_for_notification_endpoint_output()
# Shouldn't exist, shouldn't have fired
assert not os.path.isfile("test-datastore/notification.txt")
# Now the filter should exist
set_response_with_filter()
client.get(url_for("form_watch_checknow"), follow_redirects=True)
time.sleep(3)
wait_for_notification_endpoint_output()
assert os.path.isfile("test-datastore/notification.txt")

@ -1,7 +1,8 @@
import os
import time
from flask import url_for
from .util import set_original_response, live_server_setup, extract_UUID_from_client, wait_for_all_checks
from .util import set_original_response, live_server_setup, extract_UUID_from_client, wait_for_all_checks, \
wait_for_notification_endpoint_output
from changedetectionio.model import App
@ -107,7 +108,8 @@ def run_filter_test(client, live_server, content_filter):
# One more check should trigger the _FILTER_FAILURE_THRESHOLD_ATTEMPTS_DEFAULT threshold
client.get(url_for("form_watch_checknow"), follow_redirects=True)
wait_for_all_checks(client)
time.sleep(2) # delay for apprise to fire
wait_for_notification_endpoint_output()
# Now it should exist and contain our "filter not found" alert
assert os.path.isfile("test-datastore/notification.txt")
@ -127,6 +129,7 @@ def run_filter_test(client, live_server, content_filter):
client.get(url_for("form_watch_checknow"), follow_redirects=True)
wait_for_all_checks(client)
wait_for_notification_endpoint_output()
# It should have sent a notification, but..
assert os.path.isfile("test-datastore/notification.txt")
# but it should not contain the info about a failed filter (because there was none in this case)

@ -2,6 +2,8 @@
from flask import url_for
from .util import set_original_response, set_modified_response, live_server_setup, wait_for_all_checks
import time
def set_nonrenderable_response():
test_return_data = """<html>
@ -11,17 +13,16 @@ def set_nonrenderable_response():
</body>
</html>
"""
with open("test-datastore/endpoint-content.txt", "w") as f:
f.write(test_return_data)
time.sleep(1)
return None
def set_zero_byte_response():
with open("test-datastore/endpoint-content.txt", "w") as f:
f.write("")
time.sleep(1)
return None
def test_check_basic_change_detection_functionality(client, live_server, measure_memory_usage):

@ -3,7 +3,7 @@ import os
import time
from flask import url_for
from .util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
from .util import live_server_setup, wait_for_all_checks, extract_UUID_from_client, wait_for_notification_endpoint_output
from ..notification import default_notification_format
instock_props = [
@ -182,7 +182,8 @@ def _run_test_minmax_limit(client, extra_watch_edit_form):
# price changed to something LESS than min (900), SHOULD be a change
set_original_response(props_markup=instock_props[0], price='890.45')
# let previous runs wait
time.sleep(1)
time.sleep(2)
res = client.get(url_for("form_watch_checknow"), follow_redirects=True)
assert b'1 watches queued for rechecking.' in res.data
wait_for_all_checks(client)
@ -362,7 +363,7 @@ def test_change_with_notification_values(client, live_server):
set_original_response(props_markup=instock_props[0], price='1950.45')
client.get(url_for("form_watch_checknow"))
wait_for_all_checks(client)
time.sleep(3)
wait_for_notification_endpoint_output()
assert os.path.isfile("test-datastore/notification.txt"), "Notification received"
with open("test-datastore/notification.txt", 'r') as f:
notification = f.read()

@ -76,6 +76,17 @@ def set_more_modified_response():
return None
def wait_for_notification_endpoint_output():
'''Apprise can take a few seconds to fire'''
from os.path import isfile
for i in range(1, 20):
time.sleep(1)
if isfile("test-datastore/notification.txt"):
return True
return False
# kinda funky, but works for now
def extract_api_key_from_UI(client):
import re

@ -18,7 +18,7 @@ services:
#
# Log levels are in descending order. (TRACE is the most detailed one)
# Log output levels: TRACE, DEBUG(default), INFO, SUCCESS, WARNING, ERROR, CRITICAL
# - LOGGER_LEVEL=DEBUG
# - LOGGER_LEVEL=TRACE
#
# Alternative WebDriver/selenium URL, do not use "'s or 's!
# - WEBDRIVER_URL=http://browser-chrome:4444/wd/hub
@ -29,8 +29,9 @@ services:
#
# https://selenium-python.readthedocs.io/api.html#module-selenium.webdriver.common.proxy
#
# Alternative Playwright URL, do not use "'s or 's!
# - PLAYWRIGHT_DRIVER_URL=ws://playwright-chrome:3000
# Alternative target "Chrome" Playwright URL, do not use "'s or 's!
# "Playwright" is a driver/librarythat allows changedetection to talk to a Chrome or similar browser.
# - PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000
#
# Playwright proxy settings playwright_proxy_server, playwright_proxy_bypass, playwright_proxy_username, playwright_proxy_password
#
@ -73,10 +74,10 @@ services:
# condition: service_started
# Used for fetching pages via Playwright+Chrome where you need Javascript support.
# Sockpuppetbrowser is basically chrome wrapped in an API for allowing fast fetching of web-pages.
# RECOMMENDED FOR FETCHING PAGES WITH CHROME
# playwright-chrome:
# hostname: playwright-chrome
# sockpuppetbrowser:
# hostname: sockpuppetbrowser
# image: dgtlmoon/sockpuppetbrowser:latest
# cap_add:
# - SYS_ADMIN

@ -79,8 +79,9 @@ pyppeteerstealth>=0.0.4
pytest ~=7.2
pytest-flask ~=1.2
# Pin jsonschema version to prevent build errors on armv6 while rpds-py wheels aren't available (1708)
jsonschema==4.17.3
# Anything 4.0 and up but not 5.0
jsonschema ~= 4.0
loguru

Loading…
Cancel
Save