Tidy up thread logic and version check

pull/15/head
Leigh Morresi 4 years ago
parent 187523d8d6
commit e0578acca2

@ -17,6 +17,8 @@ import os
import timeago import timeago
import threading import threading
from threading import Event
import queue import queue
from flask import Flask, render_template, request, send_file, send_from_directory, abort, redirect, url_for from flask import Flask, render_template, request, send_file, send_from_directory, abort, redirect, url_for
@ -42,7 +44,9 @@ app = Flask(__name__, static_url_path="/var/www/change-detection/backen/static")
# Stop browser caching of assets # Stop browser caching of assets
app.config['SEND_FILE_MAX_AGE_DEFAULT'] = 0 app.config['SEND_FILE_MAX_AGE_DEFAULT'] = 0
app.config['STOP_THREADS'] = False app.config.exit = Event()
app.config['NEW_VERSION_AVAILABLE'] = False
# Disables caching of the templates # Disables caching of the templates
app.config['TEMPLATES_AUTO_RELOAD'] = True app.config['TEMPLATES_AUTO_RELOAD'] = True
@ -79,10 +83,12 @@ def _jinja2_filter_datetimestamp(timestamp, format="%Y-%m-%d %H:%M:%S"):
def changedetection_app(config=None, datastore_o=None): def changedetection_app(config=None, datastore_o=None):
global datastore global datastore
datastore = datastore_o datastore = datastore_o
# Hmm
app.config.update(dict(DEBUG=True)) app.config.update(dict(DEBUG=True))
app.config.update(config or {}) app.config.update(config or {})
# Setup cors headers to allow all domains # Setup cors headers to allow all domains
# https://flask-cors.readthedocs.io/en/latest/ # https://flask-cors.readthedocs.io/en/latest/
# CORS(app) # CORS(app)
@ -90,6 +96,7 @@ def changedetection_app(config=None, datastore_o=None):
# https://github.com/pallets/flask/blob/93dd1709d05a1cf0e886df6223377bdab3b077fb/examples/tutorial/flaskr/__init__.py#L39 # https://github.com/pallets/flask/blob/93dd1709d05a1cf0e886df6223377bdab3b077fb/examples/tutorial/flaskr/__init__.py#L39
# You can divide up the stuff like this # You can divide up the stuff like this
@app.route("/", methods=['GET']) @app.route("/", methods=['GET'])
def index(): def index():
global messages global messages
@ -318,12 +325,14 @@ def changedetection_app(config=None, datastore_o=None):
if len(remaining_urls) == 0: if len(remaining_urls) == 0:
return redirect(url_for('index')) return redirect(url_for('index'))
#@todo repair
else: else:
output = render_template("import.html", output = render_template("import.html",
messages=messages, messages=messages,
remaining="\n".join(remaining_urls) remaining="\n".join(remaining_urls)
) )
messages = [] messages = []
return output return output
# Clear all statuses, so we do not see the 'unviewed' class # Clear all statuses, so we do not see the 'unviewed' class
@ -501,9 +510,27 @@ def changedetection_app(config=None, datastore_o=None):
# @todo handle ctrl break # @todo handle ctrl break
ticker_thread = threading.Thread(target=ticker_thread_check_time_launch_checks).start() ticker_thread = threading.Thread(target=ticker_thread_check_time_launch_checks).start()
# Check for new release version
threading.Thread(target=check_for_new_version).start()
return app return app
# Check for new version and anonymous stats
def check_for_new_version():
import requests
app.config['NEW_VERSION_AVAILABLE'] = True
import urllib3
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
while not app.config.exit.is_set():
r = requests.post("https://changedetection.io/check-ver.php",
data={'version': datastore.data['version_tag'],
'app_guid': datastore.data['app_guid']},
verify=False)
app.config.exit.wait(10)
# Requests for checking on the site use a pool of thread Workers managed by a Queue. # Requests for checking on the site use a pool of thread Workers managed by a Queue.
class Worker(threading.Thread): class Worker(threading.Thread):
current_uuid = None current_uuid = None
@ -517,16 +544,13 @@ class Worker(threading.Thread):
update_handler = fetch_site_status.perform_site_check(datastore=datastore) update_handler = fetch_site_status.perform_site_check(datastore=datastore)
while True: while not app.config.exit.is_set():
try: try:
uuid = self.q.get(block=True, timeout=1) uuid = self.q.get(block=False)
except queue.Empty: except queue.Empty:
# We have a chance to kill this thread that needs to monitor for new jobs.. pass
# Delays here would be caused by a current response object pending
# @todo switch to threaded response handler
if app.config['STOP_THREADS']:
return
else: else:
self.current_uuid = uuid self.current_uuid = uuid
@ -549,9 +573,13 @@ class Worker(threading.Thread):
self.current_uuid = None # Done self.current_uuid = None # Done
self.q.task_done() self.q.task_done()
app.config.exit.wait(1)
# Thread runner to check every minute, look for new watches to feed into the Queue. # Thread runner to check every minute, look for new watches to feed into the Queue.
def ticker_thread_check_time_launch_checks(): def ticker_thread_check_time_launch_checks():
# Spin up Workers. # Spin up Workers.
for _ in range(datastore.data['settings']['requests']['workers']): for _ in range(datastore.data['settings']['requests']['workers']):
new_worker = Worker(update_q) new_worker = Worker(update_q)
@ -559,23 +587,19 @@ def ticker_thread_check_time_launch_checks():
new_worker.start() new_worker.start()
# Every minute check for new UUIDs to follow up on # Every minute check for new UUIDs to follow up on
while True: minutes = datastore.data['settings']['requests']['minutes_between_check']
if app.config['STOP_THREADS']:
return
while not app.config.exit.is_set():
running_uuids = [] running_uuids = []
for t in running_update_threads: for t in running_update_threads:
running_uuids.append(t.current_uuid) running_uuids.append(t.current_uuid)
# Look at the dataset, find a stale watch to process # Look at the dataset, find a stale watch to process
minutes = datastore.data['settings']['requests']['minutes_between_check'] threshold = time.time() - (minutes * 60)
for uuid, watch in datastore.data['watching'].items(): for uuid, watch in datastore.data['watching'].items():
if watch['last_checked'] <= time.time() - (minutes * 60): if watch['last_checked'] <= threshold:
# @todo maybe update_q.queue is enough?
if not uuid in running_uuids and uuid not in update_q.queue: if not uuid in running_uuids and uuid not in update_q.queue:
update_q.put(uuid) update_q.put(uuid)
# Should be low so we can break this out in testing # Should be low so we can break this out in testing
time.sleep(1) app.config.exit.wait(1)

@ -2,7 +2,8 @@ import time
import requests import requests
import hashlib import hashlib
from inscriptis import get_text from inscriptis import get_text
import urllib3
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
# Some common stuff here that can be moved to a base class # Some common stuff here that can be moved to a base class
class perform_site_check(): class perform_site_check():

@ -22,6 +22,7 @@ class ChangeDetectionStore:
self.datastore_path = datastore_path self.datastore_path = datastore_path
self.json_store_path = "{}/url-watches.json".format(self.datastore_path) self.json_store_path = "{}/url-watches.json".format(self.datastore_path)
self.stop_thread = False self.stop_thread = False
self.__data = { self.__data = {
'note': "Hello! If you change this file manually, please be sure to restart your changedetection.io instance!", 'note': "Hello! If you change this file manually, please be sure to restart your changedetection.io instance!",
'watching': {}, 'watching': {},
@ -79,8 +80,6 @@ class ChangeDetectionStore:
if 'requests' in from_disk['settings']: if 'requests' in from_disk['settings']:
self.__data['settings']['requests'].update(from_disk['settings']['requests']) self.__data['settings']['requests'].update(from_disk['settings']['requests'])
self.__data['tag'] = "0.27"
# Reinitialise each `watching` with our generic_definition in the case that we add a new var in the future. # Reinitialise each `watching` with our generic_definition in the case that we add a new var in the future.
# @todo pretty sure theres a python we todo this with an abstracted(?) object! # @todo pretty sure theres a python we todo this with an abstracted(?) object!
@ -101,6 +100,14 @@ class ChangeDetectionStore:
self.add_watch(url='https://www.gov.uk/coronavirus', tag='Covid') self.add_watch(url='https://www.gov.uk/coronavirus', tag='Covid')
self.add_watch(url='https://changedetection.io', tag='Tech news') self.add_watch(url='https://changedetection.io', tag='Tech news')
self.__data['version_tag'] = "0.27"
if not 'app_guid' in self.__data:
self.__data['app_guid'] = str(uuid_builder.uuid4())
self.needs_write = True
# Finally start the thread that will manage periodic data saves to JSON # Finally start the thread that will manage periodic data saves to JSON
save_data_thread = threading.Thread(target=self.save_datastore).start() save_data_thread = threading.Thread(target=self.save_datastore).start()

@ -21,7 +21,9 @@
{% if current_diff_url %} {% if current_diff_url %}
<a class=current-diff-url href="{{ current_diff_url }}"><span style="max-width: 30%; overflow: hidden;">{{ current_diff_url }}</a> <a class=current-diff-url href="{{ current_diff_url }}"><span style="max-width: 30%; overflow: hidden;">{{ current_diff_url }}</a>
{% else %} {% else %}
<span id="version-text" class="pure-menu-heading">Version {{ version }}</span> {% if new_version_available %}
<span id="version-text" class="pure-menu-heading">A new version is available</span>
{% endif %}
{% endif %} {% endif %}
<ul class="pure-menu-list"> <ul class="pure-menu-list">

@ -30,10 +30,11 @@ def app(request):
app_config = {'datastore_path': datastore_path} app_config = {'datastore_path': datastore_path}
datastore = store.ChangeDetectionStore(datastore_path=app_config['datastore_path'], include_default_watches=False) datastore = store.ChangeDetectionStore(datastore_path=app_config['datastore_path'], include_default_watches=False)
app = changedetection_app(app_config, datastore) app = changedetection_app(app_config, datastore)
app.config['STOP_THREADS'] = True
def teardown(): def teardown():
datastore.stop_thread = True datastore.stop_thread = True
app.config['STOP_THREADS'] = True app.config.exit.set()
try: try:
os.unlink("{}/url-watches.json".format(datastore_path)) os.unlink("{}/url-watches.json".format(datastore_path))
except FileNotFoundError: except FileNotFoundError:

Loading…
Cancel
Save