Fetcher - CPU usage - Skip processing if the previous checksum and the just fetched one was the same (#925)

pull/1232/head
dgtlmoon 1 year ago committed by GitHub
parent 93cc30437f
commit b76148a0f4
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -10,6 +10,7 @@ import threading
import time
import timeago
from changedetectionio import queuedWatchMetaData
from copy import deepcopy
from distutils.util import strtobool
from feedgen.feed import FeedGenerator
@ -404,7 +405,6 @@ def changedetection_app(config=None, datastore_o=None):
sorted_watches.append(watch)
existing_tags = datastore.get_all_tags()
form = forms.quickWatchForm(request.form)
output = render_template("watch-overview.html",
form=form,
@ -416,7 +416,7 @@ def changedetection_app(config=None, datastore_o=None):
# Don't link to hosting when we're on the hosting environment
hosted_sticky=os.getenv("SALTED_PASS", False) == False,
guid=datastore.data['app_guid'],
queued_uuids=[uuid for p,uuid in update_q.queue])
queued_uuids=[q_uuid.item['uuid'] for q_uuid in update_q.queue])
if session.get('share-link'):
@ -596,25 +596,16 @@ def changedetection_app(config=None, datastore_o=None):
using_default_check_time = False
break
# Use the default if its the same as system wide
# Use the default if it's the same as system-wide.
if form.fetch_backend.data == datastore.data['settings']['application']['fetch_backend']:
extra_update_obj['fetch_backend'] = None
# Ignore text
form_ignore_text = form.ignore_text.data
datastore.data['watching'][uuid]['ignore_text'] = form_ignore_text
# Reset the previous_md5 so we process a new snapshot including stripping ignore text.
if form_ignore_text:
if len(datastore.data['watching'][uuid].history):
extra_update_obj['previous_md5'] = get_current_checksum_include_ignore_text(uuid=uuid)
# Reset the previous_md5 so we process a new snapshot including stripping ignore text.
if form.include_filters.data != datastore.data['watching'][uuid].get('include_filters', []):
if len(datastore.data['watching'][uuid].history):
extra_update_obj['previous_md5'] = get_current_checksum_include_ignore_text(uuid=uuid)
# Be sure proxy value is None
if datastore.proxy_list is not None and form.data['proxy'] == '':
extra_update_obj['proxy'] = None
@ -632,7 +623,7 @@ def changedetection_app(config=None, datastore_o=None):
datastore.needs_write_urgent = True
# Queue the watch for immediate recheck, with a higher priority
update_q.put((1, uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
# Diff page [edit] link should go back to diff page
if request.args.get("next") and request.args.get("next") == 'diff':
@ -773,7 +764,7 @@ def changedetection_app(config=None, datastore_o=None):
importer = import_url_list()
importer.run(data=request.values.get('urls'), flash=flash, datastore=datastore)
for uuid in importer.new_uuids:
update_q.put((1, uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
if len(importer.remaining_data) == 0:
return redirect(url_for('index'))
@ -786,7 +777,7 @@ def changedetection_app(config=None, datastore_o=None):
d_importer = import_distill_io_json()
d_importer.run(data=request.values.get('distill-io'), flash=flash, datastore=datastore)
for uuid in d_importer.new_uuids:
update_q.put((1, uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
@ -1151,7 +1142,7 @@ def changedetection_app(config=None, datastore_o=None):
if not add_paused and new_uuid:
# Straight into the queue.
update_q.put((1, new_uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': new_uuid}))
flash("Watch added.")
if add_paused:
@ -1188,7 +1179,7 @@ def changedetection_app(config=None, datastore_o=None):
uuid = list(datastore.data['watching'].keys()).pop()
new_uuid = datastore.clone(uuid)
update_q.put((5, new_uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=5, item={'uuid': new_uuid, 'skip_when_checksum_same': True}))
flash('Cloned.')
return redirect(url_for('index'))
@ -1196,7 +1187,7 @@ def changedetection_app(config=None, datastore_o=None):
@app.route("/api/checknow", methods=['GET'])
@login_required
def form_watch_checknow():
# Forced recheck will skip the 'skip if content is the same' rule (, 'reprocess_existing_data': True})))
tag = request.args.get('tag')
uuid = request.args.get('uuid')
i = 0
@ -1205,11 +1196,9 @@ def changedetection_app(config=None, datastore_o=None):
for t in running_update_threads:
running_uuids.append(t.current_uuid)
# @todo check thread is running and skip
if uuid:
if uuid not in running_uuids:
update_q.put((1, uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
i = 1
elif tag != None:
@ -1217,14 +1206,14 @@ def changedetection_app(config=None, datastore_o=None):
for watch_uuid, watch in datastore.data['watching'].items():
if (tag != None and tag in watch['tag']):
if watch_uuid not in running_uuids and not datastore.data['watching'][watch_uuid]['paused']:
update_q.put((1, watch_uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': watch_uuid, 'skip_when_checksum_same': False}))
i += 1
else:
# No tag, no uuid, add everything.
for watch_uuid, watch in datastore.data['watching'].items():
if watch_uuid not in running_uuids and not datastore.data['watching'][watch_uuid]['paused']:
update_q.put((1, watch_uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': watch_uuid, 'skip_when_checksum_same': False}))
i += 1
flash("{} watches are queued for rechecking.".format(i))
return redirect(url_for('index', tag=tag))
@ -1344,7 +1333,7 @@ def changedetection_app(config=None, datastore_o=None):
app.register_blueprint(browser_steps.construct_blueprint(datastore), url_prefix='/browser-steps')
import changedetectionio.blueprint.price_data_follower as price_data_follower
app.register_blueprint(price_data_follower.construct_blueprint(datastore), url_prefix='/price_data_follower')
app.register_blueprint(price_data_follower.construct_blueprint(datastore, update_q), url_prefix='/price_data_follower')
# @todo handle ctrl break
@ -1492,7 +1481,7 @@ def ticker_thread_check_time_launch_checks():
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]:
if not uuid in running_uuids and uuid not in [q_uuid.item['uuid'] for 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)
@ -1523,8 +1512,9 @@ def ticker_thread_check_time_launch_checks():
priority,
watch.jitter_seconds,
now - watch['last_checked']))
# Into the queue with you
update_q.put((priority, uuid))
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=priority, item={'uuid': uuid, 'skip_when_checksum_same': True}))
# Reset for next time
watch.jitter_seconds = 0

@ -1,3 +1,4 @@
from changedetectionio import queuedWatchMetaData
from flask_restful import abort, Resource
from flask import request, make_response
import validators
@ -24,7 +25,7 @@ class Watch(Resource):
abort(404, message='No watch exists with the UUID of {}'.format(uuid))
if request.args.get('recheck'):
self.update_q.put((1, uuid))
self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
return "OK", 200
# Return without history, get that via another API call
@ -100,7 +101,7 @@ class CreateWatch(Resource):
extras = {'title': json_data['title'].strip()} if json_data.get('title') else {}
new_uuid = self.datastore.add_watch(url=json_data['url'].strip(), tag=tag, extras=extras)
self.update_q.put((1, new_uuid))
self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': new_uuid, 'skip_when_checksum_same': True}))
return {'uuid': new_uuid}, 201
# Return concise list of available watches and some very basic info
@ -118,7 +119,7 @@ class CreateWatch(Resource):
if request.args.get('recheck_all'):
for uuid in self.datastore.data['watching'].keys():
self.update_q.put((1, uuid))
self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
return {'status': "OK"}, 200
return list, 200

@ -3,11 +3,13 @@ from distutils.util import strtobool
from flask import Blueprint, flash, redirect, url_for
from flask_login import login_required
from changedetectionio.store import ChangeDetectionStore
from changedetectionio import queuedWatchMetaData
from queue import PriorityQueue
PRICE_DATA_TRACK_ACCEPT = 'accepted'
PRICE_DATA_TRACK_REJECT = 'rejected'
def construct_blueprint(datastore: ChangeDetectionStore):
def construct_blueprint(datastore: ChangeDetectionStore, update_q: PriorityQueue):
price_data_follower_blueprint = Blueprint('price_data_follower', __name__)
@ -15,6 +17,7 @@ def construct_blueprint(datastore: ChangeDetectionStore):
@price_data_follower_blueprint.route("/<string:uuid>/accept", methods=['GET'])
def accept(uuid):
datastore.data['watching'][uuid]['track_ldjson_price_data'] = PRICE_DATA_TRACK_ACCEPT
update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
return redirect(url_for("form_watch_checknow", uuid=uuid))

@ -23,6 +23,9 @@ class Non200ErrorCodeReceived(Exception):
self.page_text = html_tools.html_to_text(page_html)
return
class checksumFromPreviousCheckWasTheSame(Exception):
def __init__(self):
return
class JSActionExceptions(Exception):
def __init__(self, status_code, url, screenshot, message=''):

@ -6,6 +6,7 @@ import urllib3
from changedetectionio import content_fetcher, html_tools
from changedetectionio.blueprint.price_data_follower import PRICE_DATA_TRACK_ACCEPT, PRICE_DATA_TRACK_REJECT
from copy import deepcopy
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
@ -38,8 +39,7 @@ class perform_site_check():
return regex
def run(self, uuid):
from copy import deepcopy
def run(self, uuid, skip_when_checksum_same=True):
changed_detected = False
screenshot = False # as bytes
stripped_text_from_html = ""
@ -122,6 +122,14 @@ class perform_site_check():
self.screenshot = fetcher.screenshot
self.xpath_data = fetcher.xpath_data
# Watches added automatically in the queue manager will skip if its the same checksum as the previous run
# Saves a lot of CPU
update_obj['previous_md5_before_filters'] = hashlib.md5(fetcher.content.encode('utf-8')).hexdigest()
if skip_when_checksum_same:
if update_obj['previous_md5_before_filters'] == watch.get('previous_md5_before_filters'):
raise content_fetcher.checksumFromPreviousCheckWasTheSame()
# Fetching complete, now filters
# @todo move to class / maybe inside of fetcher abstract base?

@ -14,51 +14,52 @@ from changedetectionio.notification import (
class model(dict):
__newest_history_key = None
__history_n=0
__history_n = 0
__base_config = {
#'history': {}, # Dict of timestamp and output stripped filename (removed)
#'newest_history_key': 0, (removed, taken from history.txt index)
'body': None,
'check_unique_lines': False, # On change-detected, compare against all history if its something new
'check_count': 0,
'consecutive_filter_failures': 0, # Every time the CSS/xPath filter cannot be located, reset when all is fine.
'extract_text': [], # Extract text by regex after filters
'extract_title_as_title': False,
'fetch_backend': None,
'filter_failure_notification_send': strtobool(os.getenv('FILTER_FAILURE_NOTIFICATION_SEND_DEFAULT', 'True')),
'has_ldjson_price_data': None,
'track_ldjson_price_data': None,
'headers': {}, # Extra headers to send
'ignore_text': [], # List of text to ignore when calculating the comparison checksum
'include_filters': [],
'last_checked': 0,
'last_error': False,
'last_viewed': 0, # history key value of the last viewed via the [diff] link
'method': 'GET',
# Custom notification content
'notification_body': None,
'notification_format': default_notification_format_for_watch,
'notification_muted': False,
'notification_title': None,
'notification_screenshot': False, # Include the latest screenshot if available and supported by the apprise URL
'notification_urls': [], # List of URLs to add to the notification Queue (Usually AppRise)
'paused': False,
'previous_md5': False,
'proxy': None, # Preferred proxy connection
'subtractive_selectors': [],
'tag': None,
'text_should_not_be_present': [], # Text that should not present
# Re #110, so then if this is set to None, we know to use the default value instead
# Requires setting to None on submit if it's the same as the default
# Should be all None by default, so we use the system default in this case.
'time_between_check': {'weeks': None, 'days': None, 'hours': None, 'minutes': None, 'seconds': None},
'title': None,
'trigger_text': [], # List of text or regex to wait for until a change is detected
'url': None,
'uuid': str(uuid.uuid4()),
'webdriver_delay': None,
'webdriver_js_execute_code': None, # Run before change-detection
}
# 'history': {}, # Dict of timestamp and output stripped filename (removed)
# 'newest_history_key': 0, (removed, taken from history.txt index)
'body': None,
'check_unique_lines': False, # On change-detected, compare against all history if its something new
'check_count': 0,
'consecutive_filter_failures': 0, # Every time the CSS/xPath filter cannot be located, reset when all is fine.
'extract_text': [], # Extract text by regex after filters
'extract_title_as_title': False,
'fetch_backend': None,
'filter_failure_notification_send': strtobool(os.getenv('FILTER_FAILURE_NOTIFICATION_SEND_DEFAULT', 'True')),
'has_ldjson_price_data': None,
'track_ldjson_price_data': None,
'headers': {}, # Extra headers to send
'ignore_text': [], # List of text to ignore when calculating the comparison checksum
'include_filters': [],
'last_checked': 0,
'last_error': False,
'last_viewed': 0, # history key value of the last viewed via the [diff] link
'method': 'GET',
# Custom notification content
'notification_body': None,
'notification_format': default_notification_format_for_watch,
'notification_muted': False,
'notification_title': None,
'notification_screenshot': False, # Include the latest screenshot if available and supported by the apprise URL
'notification_urls': [], # List of URLs to add to the notification Queue (Usually AppRise)
'paused': False,
'previous_md5': False,
'previous_md5_before_filters': False, # Used for skipping changedetection entirely
'proxy': None, # Preferred proxy connection
'subtractive_selectors': [],
'tag': None,
'text_should_not_be_present': [], # Text that should not present
# Re #110, so then if this is set to None, we know to use the default value instead
# Requires setting to None on submit if it's the same as the default
# Should be all None by default, so we use the system default in this case.
'time_between_check': {'weeks': None, 'days': None, 'hours': None, 'minutes': None, 'seconds': None},
'title': None,
'trigger_text': [], # List of text or regex to wait for until a change is detected
'url': None,
'uuid': str(uuid.uuid4()),
'webdriver_delay': None,
'webdriver_js_execute_code': None, # Run before change-detection
}
jitter_seconds = 0
def __init__(self, *arg, **kw):

@ -0,0 +1,10 @@
from dataclasses import dataclass, field
from typing import Any
# So that we can queue some metadata in `item`
# https://docs.python.org/3/library/queue.html#queue.PriorityQueue
#
@dataclass(order=True)
class PrioritizedItem:
priority: int
item: Any=field(compare=False)

@ -4,6 +4,7 @@ import queue
import time
from changedetectionio import content_fetcher
from changedetectionio import queuedWatchMetaData
from changedetectionio.fetch_site_status import FilterNotFoundInResponse
# A single update worker
@ -157,11 +158,12 @@ class update_worker(threading.Thread):
while not self.app.config.exit.is_set():
try:
priority, uuid = self.q.get(block=False)
queued_item_data = self.q.get(block=False)
except queue.Empty:
pass
else:
uuid = queued_item_data.item.get('uuid')
self.current_uuid = uuid
if uuid in list(self.datastore.data['watching'].keys()):
@ -171,11 +173,11 @@ class update_worker(threading.Thread):
update_obj= {}
xpath_data = False
process_changedetection_results = True
print("> Processing UUID {} Priority {} URL {}".format(uuid, priority, self.datastore.data['watching'][uuid]['url']))
print("> Processing UUID {} Priority {} URL {}".format(uuid, queued_item_data.priority, self.datastore.data['watching'][uuid]['url']))
now = time.time()
try:
changed_detected, update_obj, contents = update_handler.run(uuid)
changed_detected, update_obj, contents = update_handler.run(uuid, skip_when_checksum_same=queued_item_data.item.get('skip_when_checksum_same'))
# Re #342
# In Python 3, all strings are sequences of Unicode characters. There is a bytes type that holds raw bytes.
# We then convert/.decode('utf-8') for the notification etc
@ -241,6 +243,10 @@ class update_worker(threading.Thread):
process_changedetection_results = True
except content_fetcher.checksumFromPreviousCheckWasTheSame as e:
# Yes fine, so nothing todo
pass
except content_fetcher.BrowserStepsStepTimout as e:
if not self.datastore.data['watching'].get(uuid):

Loading…
Cancel
Save