Compare commits

..

2 Commits

Author SHA1 Message Date
dgtlmoon 8f83223f26 Improve job names in graph 2025-06-03 16:48:26 +02:00
dgtlmoon 0215e91991 Build test - Build test for platforms in parallel 2025-06-03 16:43:30 +02:00
13 changed files with 35 additions and 178 deletions
@@ -228,8 +228,10 @@ def construct_blueprint(datastore: ChangeDetectionStore):
watch.save_screenshot(screenshot=screenshot)
watch.save_xpath_data(data=xpath_data)
except playwright._impl._api_types.Error as e:
return make_response("Browser session ran out of time :( Please reload this page."+str(e), 401)
except Exception as e:
return make_response(f"Error fetching screenshot and element data - {str(e)}", 401)
return make_response("Error fetching screenshot and element data - " + str(e), 401)
# SEND THIS BACK TO THE BROWSER
output = {
@@ -1,6 +1,8 @@
import os
import time
import re
import sys
import traceback
from random import randint
from loguru import logger
@@ -90,32 +92,8 @@ class steppable_browser_interface():
if optional_value and ('{%' in optional_value or '{{' in optional_value):
optional_value = jinja_render(template_str=optional_value)
# Trigger click and cautiously handle potential navigation
# This means the page redirects/reloads/changes JS etc etc
if call_action_name.startswith('click_'):
try:
# Set up navigation expectation before the click (like sync version)
async with self.page.expect_event("framenavigated", timeout=3000) as navigation_info:
await action_handler(selector, optional_value)
# Check if navigation actually occurred
try:
await navigation_info.value # This waits for the navigation promise
logger.debug(f"Navigation occurred on {call_action_name}.")
except Exception:
logger.debug(f"No navigation occurred within timeout when calling {call_action_name}, that's OK, continuing.")
except Exception as e:
# If expect_event itself times out, that means no navigation occurred - that's OK
if "framenavigated" in str(e) and "exceeded" in str(e):
logger.debug(f"No navigation occurred within timeout when calling {call_action_name}, that's OK, continuing.")
else:
raise e
else:
# Some other action that probably a navigation is not expected
await action_handler(selector, optional_value)
await action_handler(selector, optional_value)
# Safely wait for timeout
await self.page.wait_for_timeout(1.5 * 1000)
logger.debug(f"Call action done in {time.time()-now:.2f}s")
@@ -132,7 +110,7 @@ class steppable_browser_interface():
# Incase they request to go back to the start
async def action_goto_site(self, selector=None, value=None):
return await self.action_goto_url(value=re.sub(r'^source:', '', self.start_url, flags=re.IGNORECASE))
return await self.action_goto_url(value=self.start_url)
async def action_click_element_containing_text(self, selector=None, value=''):
logger.debug("Clicking element containing text")
@@ -450,9 +428,6 @@ class browsersteps_live_ui(steppable_browser_interface):
try:
# Get screenshot first
screenshot = await capture_full_page_async(page=self.page)
if not screenshot:
logger.error("No screenshot was retrieved :((")
logger.debug(f"Time to get screenshot from browser {time.time() - now:.2f}s")
# Then get interactive elements
@@ -475,12 +450,6 @@ class browsersteps_live_ui(steppable_browser_interface):
except Exception as e:
logger.error(f"Error getting current state: {str(e)}")
# If the page has navigated (common with logins) then the context is destroyed on navigation, continue
# I'm not sure that this is required anymore because we have the "expect navigation wrapper" at the top
if "Execution context was destroyed" in str(e):
logger.debug("Execution context was destroyed, most likely because of navigation, continuing...")
pass
# Attempt recovery - force garbage collection
try:
await self.page.request_gc()
+10 -43
View File
@@ -51,15 +51,7 @@ async def capture_full_page(page):
await page.setViewport({'width': page.viewport['width'], 'height': step_size})
while y < min(page_height, SCREENSHOT_MAX_TOTAL_HEIGHT):
# better than scrollTo incase they override it in the page
await page.evaluate(
"""(y) => {
document.documentElement.scrollTop = y;
document.body.scrollTop = y;
}""",
y
)
await page.evaluate(f"window.scrollTo(0, {y})")
screenshot_chunks.append(await page.screenshot(type_='jpeg',
fullPage=False,
quality=int(os.getenv("SCREENSHOT_QUALITY", 72))))
@@ -157,11 +149,7 @@ class fetcher(Fetcher):
):
import re
self.delete_browser_steps_screenshots()
n = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay
extra_wait = min(n, 15)
logger.debug(f"Extra wait set to {extra_wait}s, requested was {n}s.")
extra_wait = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay
from pyppeteer import Pyppeteer
pyppeteer_instance = Pyppeteer()
@@ -177,7 +165,7 @@ class fetcher(Fetcher):
except websockets.exceptions.InvalidURI:
raise BrowserConnectError(msg=f"Error connecting to the browser, check your browser connection address (should be ws:// or wss://")
except Exception as e:
raise BrowserConnectError(msg=f"Error connecting to the browser - Exception '{str(e)}'")
raise BrowserConnectError(msg=f"Error connecting to the browser {str(e)}")
# Better is to launch chrome with the URL as arg
# non-headless - newPage() will launch an extra tab/window, .browser should already contain 1 page/tab
@@ -239,35 +227,13 @@ class fetcher(Fetcher):
# browsersteps_interface = steppable_browser_interface()
# browsersteps_interface.page = self.page
async def handle_frame_navigation(event):
logger.debug(f"Frame navigated: {event}")
w = extra_wait - 2 if extra_wait > 4 else 2
logger.debug(f"Waiting {w} seconds before calling Page.stopLoading...")
await asyncio.sleep(w)
logger.debug("Issuing stopLoading command...")
await self.page._client.send('Page.stopLoading')
logger.debug("stopLoading command sent!")
response = await self.page.goto(url, waitUntil="load")
self.page._client.on('Page.frameStartedNavigating', lambda event: asyncio.create_task(handle_frame_navigation(event)))
self.page._client.on('Page.frameStartedLoading', lambda event: asyncio.create_task(handle_frame_navigation(event)))
self.page._client.on('Page.frameStoppedLoading', lambda event: logger.debug(f"Frame stopped loading: {event}"))
response = None
attempt=0
while not response:
logger.debug(f"Attempting page fetch {url} attempt {attempt}")
response = await self.page.goto(url)
await asyncio.sleep(1 + extra_wait)
if response:
break
if not response:
logger.warning("Page did not fetch! trying again!")
if response is None and attempt>=2:
await self.page.close()
await browser.close()
logger.warning(f"Content Fetcher > Response object was none (as in, the response from the browser was empty, not just the content) exiting attmpt {attempt}")
raise EmptyReply(url=url, status_code=None)
attempt+=1
if response is None:
await self.page.close()
await browser.close()
logger.warning("Content Fetcher > Response object was none (as in, the response from the browser was empty, not just the content)")
raise EmptyReply(url=url, status_code=None)
self.headers = response.headers
@@ -310,6 +276,7 @@ class fetcher(Fetcher):
# if self.browser_steps_get_valid_steps():
# self.iterate_browser_steps()
await asyncio.sleep(1 + extra_wait)
# So we can find an element on the page where its selector was entered manually (maybe not xPath etc)
# Setup the xPath/VisualSelector scraper
@@ -18,7 +18,6 @@ async () => {
'back-order or out of stock',
'backordered',
'benachrichtigt mich', // notify me
'binnenkort leverbaar', // coming soon
'brak na stanie',
'brak w magazynie',
'coming soon',
@@ -86,7 +85,6 @@ async () => {
'tidak tersedia',
'tijdelijk uitverkocht',
'tiket tidak tersedia',
'to subscribe to back in stock',
'tükendi',
'unavailable nearby',
'unavailable tickets',
@@ -121,7 +119,8 @@ async () => {
return text.toLowerCase().trim();
}
const negateOutOfStockRegex = new RegExp('^([0-9] in stock|add to cart|in stock|arrives approximately)', 'ig');
const negateOutOfStockRegex = new RegExp('^([0-9] in stock|add to cart|in stock)', 'ig');
// The out-of-stock or in-stock-text is generally always above-the-fold
// and often below-the-fold is a list of related products that may or may not contain trigger text
// so it's good to filter to just the 'above the fold' elements
-35
View File
@@ -3,41 +3,6 @@ import asyncio
from blinker import signal
from loguru import logger
class NotificationQueue(queue.Queue):
"""
Extended Queue that sends a 'notification_event' signal when notifications are added.
This class extends the standard Queue and adds a signal emission after a notification
is put into the queue. The signal includes the watch UUID if available.
"""
def __init__(self, maxsize=0):
super().__init__(maxsize)
try:
self.notification_event_signal = signal('notification_event')
except Exception as e:
logger.critical(f"Exception creating notification_event signal: {e}")
def put(self, item, block=True, timeout=None):
# Call the parent's put method first
super().put(item, block, timeout)
# After putting the notification in the queue, emit signal with watch UUID
try:
if self.notification_event_signal and isinstance(item, dict):
watch_uuid = item.get('uuid')
if watch_uuid:
# Send the notification_event signal with the watch UUID
self.notification_event_signal.send(watch_uuid=watch_uuid)
logger.trace(f"NotificationQueue: Emitted notification_event signal for watch UUID {watch_uuid}")
else:
# Send signal without UUID for system notifications
self.notification_event_signal.send()
logger.trace("NotificationQueue: Emitted notification_event signal for system notification")
except Exception as e:
logger.error(f"Exception emitting notification_event signal: {e}")
class SignalPriorityQueue(queue.PriorityQueue):
"""
Extended PriorityQueue that sends a signal when items with a UUID are added.
+2 -2
View File
@@ -12,7 +12,7 @@ from blinker import signal
from changedetectionio.strtobool import strtobool
from threading import Event
from changedetectionio.custom_queue import SignalPriorityQueue, AsyncSignalPriorityQueue, NotificationQueue
from changedetectionio.custom_queue import SignalPriorityQueue, AsyncSignalPriorityQueue
from changedetectionio import worker_handler
from flask import (
@@ -52,7 +52,7 @@ extra_stylesheets = []
# Use async queue by default, keep sync for backward compatibility
update_q = AsyncSignalPriorityQueue() if worker_handler.USE_ASYNC_WORKERS else SignalPriorityQueue()
notification_q = NotificationQueue()
notification_q = queue.Queue()
MAX_QUEUE_SIZE = 2000
app = Flask(__name__,
@@ -29,11 +29,6 @@ class SignalHandler:
watch_delete_signal = signal('watch_deleted')
watch_delete_signal.connect(self.handle_deleted_signal, weak=False)
# Connect to the notification_event signal
notification_event_signal = signal('notification_event')
notification_event_signal.connect(self.handle_notification_event, weak=False)
logger.info("SignalHandler: Connected to notification_event signal")
# Create and start the queue update thread using standard threading
import threading
self.polling_emitter_thread = threading.Thread(
@@ -94,23 +89,6 @@ class SignalHandler:
except Exception as e:
logger.error(f"Socket.IO error in handle_queue_length: {str(e)}")
def handle_notification_event(self, *args, **kwargs):
"""Handle notification_event signal and emit to all clients"""
try:
watch_uuid = kwargs.get('watch_uuid')
logger.debug(f"SignalHandler: Notification event received for watch UUID: {watch_uuid}")
# Emit the notification event to all connected clients
self.socketio_instance.emit("notification_event", {
"watch_uuid": watch_uuid,
"event_timestamp": time.time()
})
logger.trace(f"Socket.IO: Emitted notification_event for watch UUID {watch_uuid}")
except Exception as e:
logger.error(f"Socket.IO error in handle_notification_event: {str(e)}")
def polling_emit_running_or_queued_watches_threaded(self):
"""Threading version of polling for Windows compatibility"""
+1 -7
View File
@@ -55,7 +55,6 @@ $(document).ready(function () {
// Connection status logging
socket.on('connect', function () {
$('#realtime-conn-error').hide();
console.log('Socket.IO connected with path:', socketio_url);
console.log('Socket transport:', socket.io.engine.transport.name);
bindSocketHandlerButtonsEvents(socket);
@@ -75,8 +74,7 @@ $(document).ready(function () {
socket.on('disconnect', function (reason) {
console.log('Socket.IO disconnected, reason:', reason);
$('.ajax-op').off('.socketHandlerNamespace');
$('#realtime-conn-error').show();
$('.ajax-op').off('.socketHandlerNamespace')
});
socket.on('queue_size', function (data) {
@@ -94,10 +92,6 @@ $(document).ready(function () {
}
});
socket.on('notification_event', function (data) {
console.log(`Stub handler for notification_event ${data.watch_uuid}`)
});
// Listen for periodically emitted watch data
console.log('Adding watch_update event listener');
@@ -17,13 +17,11 @@
&.title-col {
word-break: break-all;
white-space: normal;
a::after {
content: url(data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAoAAAAKCAYAAACNMs+9AAAAQElEQVR42qXKwQkAIAxDUUdxtO6/RBQkQZvSi8I/pL4BoGw/XPkh4XigPmsUgh0626AjRsgxHTkUThsG2T/sIlzdTsp52kSS1wAAAABJRU5ErkJggg==);
margin: 0 3px 0 5px;
}
}
a.external::after {
content: url(data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAoAAAAKCAYAAACNMs+9AAAAQElEQVR42qXKwQkAIAxDUUdxtO6/RBQkQZvSi8I/pL4BoGw/XPkh4XigPmsUgh0626AjRsgxHTkUThsG2T/sIlzdTsp52kSS1wAAAABJRU5ErkJggg==);
margin: 0 3px 0 5px;
}
}
@@ -1190,12 +1190,3 @@ ul {
vertical-align: middle;
}
#realtime-conn-error {
position: absolute;
bottom: 0;
left: 30px;
background: var(--color-warning);
padding: 10px;
font-size: 0.8rem;
color: #fff;
}
+3 -12
View File
@@ -537,9 +537,9 @@ body.preview-text-enabled {
.watch-table td.title-col {
word-break: break-all;
white-space: normal; }
.watch-table td a.external::after {
content: url(data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAoAAAAKCAYAAACNMs+9AAAAQElEQVR42qXKwQkAIAxDUUdxtO6/RBQkQZvSi8I/pL4BoGw/XPkh4XigPmsUgh0626AjRsgxHTkUThsG2T/sIlzdTsp52kSS1wAAAABJRU5ErkJggg==);
margin: 0 3px 0 5px; }
.watch-table td.title-col a::after {
content: url(data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAoAAAAKCAYAAACNMs+9AAAAQElEQVR42qXKwQkAIAxDUUdxtO6/RBQkQZvSi8I/pL4BoGw/XPkh4XigPmsUgh0626AjRsgxHTkUThsG2T/sIlzdTsp52kSS1wAAAABJRU5ErkJggg==);
margin: 0 3px 0 5px; }
.watch-table th {
white-space: nowrap; }
.watch-table th a {
@@ -1535,12 +1535,3 @@ ul {
height: 21px;
padding: 2px;
vertical-align: middle; }
#realtime-conn-error {
position: absolute;
bottom: 0;
left: 30px;
background: var(--color-warning);
padding: 10px;
font-size: 0.8rem;
color: #fff; }
+6 -2
View File
@@ -412,8 +412,12 @@ class ChangeDetectionStore:
# system was out of memory, out of RAM etc
with open(self.json_store_path+".tmp", 'w') as json_file:
# Use compact JSON in production for better performance
json.dump(data, json_file, indent=2)
os.replace(self.json_store_path+".tmp", self.json_store_path)
debug_mode = os.environ.get('CHANGEDETECTION_DEBUG', 'false').lower() == 'true'
if debug_mode:
json.dump(data, json_file, indent=4)
else:
json.dump(data, json_file, separators=(',', ':'))
os.replace(self.json_store_path+".tmp", self.json_store_path)
except Exception as e:
logger.error(f"Error writing JSON!! (Main JSON file save was skipped) : {str(e)}")
-1
View File
@@ -236,7 +236,6 @@
<script src="{{url_for('static_content', group='js', filename='toggle-theme.js')}}" defer></script>
<div id="checking-now-fixed-tab" style="display: none;"><span class="spinner"></span><span>&nbsp;Checking now</span></div>
<div id="realtime-conn-error" style="display:none">Offline</div>
</body>
</html>