fuck

Dockerfile

@@ -1,7 +1,7 @@
-FROM python:3-alpine
+FROM python:3.12-alpine
 WORKDIR /app
 COPY requirements.txt /app
 RUN pip install --no-cache-dir -r requirements.txt
 COPY . /app
 EXPOSE 8000
-CMD ["gunicorn", "--bind", "0.0.0.0:8000", "ayta:create_app()"]
+CMD ["gunicorn", "--bind", "0.0.0.0:8000", "--workers", "1", "ayta:create_app()"]

README.md

@@ -6,7 +6,7 @@ current cronjob yt-dlp archive service.
 Partially inspired by [hobune](https://github.com/rebane2001/hobune). While that project is amazingby it's own, it's just not scaleable.
 
 ## The idea
-Having over 250k videos, scaling the current cronjob yt-dlp archive task is just really hard. Filetypes change, things get partially downloaded and such.
+Having over 350k videos, scaling the current cronjob yt-dlp archive task is just really hard. Filetypes change, things get partially downloaded and such.
 Partially yt-dlp is to blame because it's a package that needs to change all the time. But with this some changes are not accounted for.
 yt-dlp will still do the downloads. But a flask frontend will be developed to make all downloaded videos easily indexable.
 For it to be quick (unlike hobune) a database has to be implemented. This could get solved by a static site generator type of software, but that is not my choice.
@@ -54,7 +54,7 @@ Extra functionality for further development of features.
 Mainly focused on retiring the cronjob based scripts and moving it to celery based tasks
 - [ ] manage videos by ID's instead of per channel basis
 - [ ] download videos from queue
-- [ ] Manage websub callbacks
+- [x] Manage websub callbacks
 
 ### Stage 4
 Mongodb finally has it's limitations.

ayta/__init__.py

@@ -26,6 +26,15 @@ def create_app(test_config=None):
     config['CELERY']['beat_schedule'] = {}
     config['CELERY']['beat_schedule']['Renew WebSub endpoints'] = {'task': 'ayta.tasks.websub_renew_expiring', 'schedule': 4000}
     config['CELERY']['beat_schedule']['Process WebSub data'] = {'task': 'ayta.tasks.websub_process_data', 'schedule': 100}
+    config['CELERY']['beat_schedule']['Queue up new videos in static channel playlists'] = {'task': 'ayta.tasks.playlist_to_queue', 'schedule': 50000}
+    
+    # Celery task routing
+    # Tasks not defined in this configuration will be routed to the default queue "celery"
+    
+    config['CELERY']['task_routes'] = {
+        'ayta.tasks.video_download': {'queue': 'download'},
+        'ayta.tasks.video_queue': {'queue': 'download'}
+    }
     
     app = Flask(__name__)
     app.config.from_mapping(config)
@@ -41,7 +50,8 @@ def create_app(test_config=None):
     app.jinja_env.filters['pretty_time'] = filters.pretty_time
     app.jinja_env.filters['current_time'] = filters.current_time
     app.jinja_env.filters['epoch_time'] = filters.epoch_time
-
+    app.jinja_env.filters['epoch_date'] = filters.epoch_date
+    
     from .blueprints import watch
     from .blueprints import index
     from .blueprints import admin

ayta/blueprints/admin.py

@@ -1,8 +1,8 @@
-from flask import Blueprint, render_template, request, redirect, url_for, flash
+from flask import Blueprint, render_template, request, redirect, url_for, flash, current_app
 from ..nosql import get_nosql
 from ..dlp import checkChannelId, getChannelInfo
 from ..decorators import login_required
-from ..tasks import websub_subscribe_callback, websub_unsubscribe_callback
+from ..tasks import test_sleep, websub_subscribe_callback, websub_unsubscribe_callback, video_download, video_queue, playlist_to_queue
 from datetime import datetime
 from secrets import token_urlsafe
 
@@ -30,28 +30,35 @@ def channels():
     generic = {}
 
     if request.method == 'POST':
-        channelId = request.form.get('channel_id', None)
-        originalName = request.form.get('original_name', None)
-        addedDate = request.form.get('added_date', None)
+        task = request.form.get('task', None)
+        
+        if task == 'add_channel':
+            channelId = request.form.get('channel_id', None)
+            originalName = request.form.get('original_name', None)
+            addedDate = request.form.get('added_date', None)
 
-        ### add some validation
-        
-        addedDate = datetime.strptime(addedDate, '%Y-%m-%d')
-        
-        if checkChannelId(channelId) is False:
-            channelId, originalName = getChannelInfo(channelId, ('channel_id', 'uploader'))
-        
-        if not get_nosql().insert_new_channel(channelId, originalName, addedDate):
-            flash('Error inserting new channel, you probably made a mistake somewhere')
-            return redirect(url_for('admin.channels'))
+            ### add some validation
+            
+            addedDate = datetime.strptime(addedDate, '%Y-%m-%d')
+            
+            if checkChannelId(channelId) is False:
+                channelId, originalName = getChannelInfo(channelId, ('channel_id', 'uploader'))
+            
+            if not get_nosql().insert_new_channel(channelId, originalName, addedDate):
+                flash('Error inserting new channel, you probably made a mistake somewhere')
+                return redirect(url_for('admin.channels'))
 
-        return redirect(url_for('admin.channel', channelId=channelId))
+            return redirect(url_for('admin.channel', channelId=channelId))
+            
+        elif task == 'playlist-queue':
+            task = playlist_to_queue.delay()
+            flash(f'Task playlist-queue has been queued: {task.id}')
 
     generic['currentDate'] = datetime.utcnow()
     channelIds = get_nosql().list_all_channels()
     
     for channelId in channelIds:
-        channels[channelId] = get_nosql().get_channel_info(channelId)
+        channels[channelId] = get_nosql().get_channel_info(channelId, limited=True)
         channels[channelId]['video_count'] = get_nosql().get_channel_videos_count(channelId)
         
     return render_template('admin/channels.html', channels=channels, generic=generic)
@@ -156,6 +163,10 @@ def reports():
             get_nosql().close_report(value)
             flash(f'Report closed {value}')
             return redirect(url_for('admin.reports'))
+        elif task == 'clean-closed':
+            get_nosql().report_clean()
+            flash(f'Cleaned closed reports older than 30 days')
+            return redirect(url_for('admin.reports'))
 
     reports = get_nosql().list_reports()
    
@@ -190,8 +201,16 @@ def queue():
             flash(f'Cleaned retired endpoints')
             
         elif task == 'manual-queue':
-            get_nosql().queue_insertQueue(value, 'webui')
-            flash(f'Added to queue: {value}')
+            if not get_nosql().check_exists(value):
+                direct = request.form.get('direct', None)
+                if direct:
+                    task = video_download.delay(value)
+                    flash(f"Started task {task.id}")
+                else:
+                    get_nosql().queue_insertQueue(value, 'webui')
+                    flash(f'Added to queue: {value}')
+            else:
+                flash(f'This video ID already exists in the archive: {value}')
             
         elif task == 'delete-queue':
             get_nosql().queue_deleteQueue(value)
@@ -199,7 +218,13 @@ def queue():
             
         elif task == 'empty-queue':
             get_nosql().queue_emptyQueue()
-            flash(f'Queue has  been emptied')
+            flash(f'Queue has been emptied')
+            
+        elif task == 'queue-run-once':
+            value = int(value) if value.isdigit() else 1
+            for x in range(value):
+                task =  video_queue.delay()
+                flash(f'Task has been started on the oldest queued item: {task.id}')
         
         return redirect(url_for('admin.queue'))
         
@@ -234,4 +259,18 @@ def users():
 
     users = get_nosql().list_all_users()
    
-    return render_template('admin/users.html', users=users)
+    return render_template('admin/users.html', users=users)
+    
+@bp.route('/workers', methods=['GET', 'POST'])
+@login_required
+def workers():
+    if request.method == 'POST':
+        task = request.form.get('task', None)
+        if task == 'test-sleep':
+            test_sleep.delay()
+
+    celery = current_app.extensions.get('celery')
+    
+    tasks = celery.control.inspect().active()
+    reserved = celery.control.inspect().reserved()
+    return render_template('admin/workers.html', tasks=tasks, reserved=reserved)

ayta/blueprints/api.py

@@ -42,7 +42,7 @@ def websub(cap):
 @bp.route('/queue/<cap>', methods=['POST'])
 def queue(cap):
     # if endpoint does not exist
-    if not get_nosql().poster_isActive(cap):
+    if not get_nosql().queue_isActive(cap):
         return abort(404)
     
     videoId = request.form.get('v')
@@ -60,7 +60,7 @@ def queue(cap):
         return abort(409)
     
     # try to insert
-    if get_nosql().poster_insertQueue(videoId, cap):
+    if get_nosql().queue_insertQueue(videoId, cap):
         return '', 202
     else:
         return abort(409)

ayta/blueprints/channel.py

@@ -11,7 +11,7 @@ def base():
     channelIds = get_nosql().list_all_channels()
     
     for channelId in channelIds:
-        channel = get_nosql().get_channel_info(channelId)
+        channel = get_nosql().get_channel_info(channelId, limited=True)
         channel['video_count'] = get_nosql().get_channel_videos_count(channelId)
         channels.append(channel)
         

ayta/filters.py

@@ -16,9 +16,15 @@ def pretty_time(time):
         except:
             return time  # return given time
 
-def epoch_time(time):
+def epoch_date(epoch):
     try:
-        return datetime.fromtimestamp(time).strftime('%d %b %Y')
+        return datetime.fromtimestamp(epoch).strftime('%d %b %Y')
+    except:
+        return None
+        
+def epoch_time(epoch):
+    try:
+        return datetime.fromtimestamp(epoch).strftime('%d %b %Y %H:%M:%S')
     except:
         return None
 

ayta/nosql.py

@@ -98,7 +98,7 @@ class Mango:
         
         stats['videos'] = self.info_json.count_documents({})
         stats['channels'] = self.channels.count_documents({})
-        stats['queue'] = self.posters_queue.count_documents({})
+        stats['queue'] = self.download_queue.count_documents({})
         
         return stats
             
@@ -168,8 +168,12 @@ class Mango:
             ids.append(video['id'])
         return tuple(ids)
  
-    def get_channel_info(self, channelId):
-        return self.channels.find_one({'id': channelId})
+    def get_channel_info(self, channelId, limited=False):
+        projection = {}
+        if limited:
+            projection['playlist'] = 0
+            
+        return self.channels.find_one({'id': channelId}, projection)
         
         
     def update_channel_key(self, channelId, key, value):
@@ -195,11 +199,14 @@ class Mango:
     def get_orphaned_videos(self):
         """ Returns a SET of YouTube video ID's which have info_jsons in the collection but no permanent channel is defined. SLOW OPERATION """
         # Ok lemme explain. Perform inner join from channel collection on channel_id key. match only the fields which are empty. return video id
-        pipeline = [{'$lookup': {'from': 'channels', 'localField': 'channel_id', 'foreignField': 'id', 'as': 'channel'}}, {'$match': {'channel': {'$size': 0}}},{'$project': {'id': 1}}]
-
+        pipeline = [{'$match': {'_status': 'available'}},
+                    {'$lookup': {'from': 'channels', 'localField': 'channel_id', 'foreignField': 'id', 'as': 'channel'}},
+                    {'$match': {'channel': {'$size': 0}}},{'$project': {'id': 1}},
+                    {'$project': {'id': 1}}]
+        
         results = self.info_json.aggregate(pipeline)
         ids = [result['id'] for result in results]
-        
+
         return tuple(ids)
     
     def get_recent_videos(self, count=99):
@@ -240,6 +247,10 @@ class Mango:
     def close_report(self, _id):
         _id = ObjectId(_id)
         return self.reports.update_one({'_id': _id}, {'$set': {'status': 'closed', 'closing_time': current_time(object=True)}})
+        
+    def report_clean(self, keep=30):
+        days = self.datetime.utcnow() - self.timedelta(days=keep)
+        self.reports.delete_many({'status': 'closed', 'closing_time': {'$lt': days}})
 
     ##########################################
     #           RUNLOG FUNCTIONS             #
@@ -408,13 +419,16 @@ class Mango:
     ##########################################
         
     def queue_insertQueue(self, videoId, endpointId=None):
-        # if no document exists
-        if not self.download_queue.count_documents({'id': videoId}) >= 1:
-            self.download_queue.insert_one({'id': videoId, 'endpoint': endpointId, 'created_time': current_time(object=True), 'status': 'queued'}).inserted_id
-            return True
-            
-        # key already in queue
-        return False
+        # if already queued
+        if self.download_queue.count_documents({'id': videoId}) >= 1:
+            return False
+        
+        # if already in archive
+        if self.check_exists(videoId):
+            return False
+        
+        # add to queue
+        return self.download_queue.insert_one({'id': videoId, 'endpoint': endpointId, 'created_time': current_time(object=True), 'status': 'queued'}).inserted_id
         
     def queue_deleteQueue(self, videoId):
         if self.download_queue.delete_one({'id': videoId}):
@@ -426,7 +440,22 @@ class Mango:
         
     def queue_emptyQueue(self):
         return self.download_queue.delete_many({})
-
+        
+    def queue_setFailed(self, videoId, reason=None):
+        return self.download_queue.update_one({'id': videoId}, {'$set': {'status': 'failed', 'fail_reason': reason}})
+        
+    def queue_getNext(self):
+        """ Returns a LIST of queue parameters. Function first checks if ID exists, if so deletes and then checks the next queued until queue is empty (None) or queued id does not exist yet."""
+        while True:
+            queueItem = self.download_queue.find_one({'status': 'queued'})
+            if not queueItem:
+                return None
+            elif self.check_exists(queueItem['id']):
+                self.queue_deleteQueue(queueItem['id'])
+                continue
+            self.download_queue.update_one({'id': queueItem['id']}, {'$set': {'status': 'working'}})
+            return queueItem
+        
 ##########################################
 #           HELPER FUNCTIONS             #
 ##########################################
@@ -450,7 +479,4 @@ def clean_info_json(originalInfo, format='dict'):
         return json.dumps(originalInfo)
     else:
         print('The requested output format is not supported!')
-    
-   
-if __name__ == '__main__':
-    mango = Mango('mongodb://root:example@192.168.66.140:27017')
+    

ayta/oidc.py

@@ -4,6 +4,7 @@ class OIDC():
     Additionally this class provides the function to generate redirect url's and check bearer tokens on their validity as well as caching jwt signing keys.
     Fairly barebones and should be 100% secure. (famous last words)
     This is made for form posted JWT's. While not the most secure it is the most easy way to implement. Moving on to a code based solution might be preferred in the future.
+    The nonce and state store is in memory, so only one instance can be used at a time until central key caching is implemented.
     """
     def __init__(self, app=None):
         self.states = {}
@@ -23,15 +24,18 @@ class OIDC():
         self.domain = config['DOMAIN']
         self.window = 120  # the time window to allow states and nonces in seconds
         
+        # Authentication provider url must be HTTPS and end on a TLD
         if self.provider[:8] != 'https://' or self.provider[-1] == '/':
             print('Incorrect OIDC provider URI', flush=True)
             exit()
-            
+        
+        # Get the provider configuration endpoints
         configuration = requests.get(f'{self.provider}/.well-known/openid-configuration').json()
         
         jwks_uri = configuration.get('jwks_uri')
         self.authorize_uri = configuration.get('authorization_endpoint')
         
+        # Start the JWKS management client, it will load the keys and maintain them
         self.jwks_manager = jwt.PyJWKClient(jwks_uri)
 
     #######################################################
@@ -39,10 +43,13 @@ class OIDC():
     def state_maintenance(self):
         from datetime import datetime
         
+        # Current time minus the acceptable window
         pivot = datetime.now().timestamp() - self.window
         
+        # List with expired states
         expired_states = [state for state, timestamp in self.states.items() if timestamp <= pivot]
         
+        # Remove expired states from store
         for state in expired_states:
             del self.states[state]
 
@@ -50,22 +57,30 @@ class OIDC():
         import secrets
         from datetime import datetime
         
+        # Clean state store first
         self.state_maintenance()
         
+        # Generate token and paired timestamp
         state = secrets.token_urlsafe(8)
         timestamp = datetime.now().timestamp()
-
+        
+        # Add token to the state store
         self.states[state] = timestamp
-
+        
+        # Return the state
         return state
         
     def state_check(self, state):
+        # Clean state store first
         self.state_maintenance()
 
+        # If given state is actively stored
         if state in self.states:
+            # Delete state and return True
             del self.states[state]
             return True
-            
+        
+        # Given state is not stored
         return False
         
     #######################################################
@@ -116,21 +131,32 @@ class OIDC():
         
     def check_bearer(self, token):
         import jwt
-
+        
+        # Test given JWT
         try:
+            # Get the signed public key from the token
             signing_key = self.jwks_manager.get_signing_key_from_jwt(token).key
+            
+            # Try to decode the token, this will also check the validity in these points:
+            # 1. Token is signed by expected keys
+            # 2. Token is issued by the expected provider
+            # 3. Expected parameters are really in the token
+            # 4. Token is really intended for us
+            # 5. Token is still valid (with 5 sec margin)
             decoded = jwt.decode(token, signing_key,
                                  algorithms=jwt.algorithms.get_default_algorithms(),
                                  issuer=self.provider,
                                  require=['aud', 'client_id', 'exp', 'iat', 'iss', 'rat', 'sub'],
                                  audience=self.client_id,
                                  leeway=5)
+                                 
+        # Any exception (invalid JWT, invalid formatting etc...) must return False
         except Exception as e:
-            print(e, flush=True)
             return False
 
-        # double check if given token is really requested by us
+        # Double check if given token is really requested by us by matching the nonce in the signed key
         if not self.nonce_check(decoded.get('nonce', None)):
             return False
-
+        
+        # Return the unique user identifier
         return decoded.get('sub', False)

ayta/tasks.py

@@ -5,6 +5,48 @@ from flask import current_app
 #              CELERY TASKS              #
 ##########################################
 
+@shared_task()
+def test_sleep(time=60):
+    from time import sleep
+    sleep(time)
+    return True
+    
+@shared_task()
+def video_download(videoId):
+    """ 
+    I do not want to deal with the quirks of native yt-dlp in python, hence the subprocess.
+    """
+    import subprocess
+    
+    process = subprocess.run(['/usr/local/bin/yt-dlp', '--config-location', '/var/www/archive.ventilaar.net/goodstuff/config_video.conf', '--', f'https://www.youtube.com/watch?v={videoId}'], stdout=subprocess.PIPE, stderr=subprocess.STDOUT, text=True)
+    
+    if process.returncode != 0:
+        return (False, process.stdout)
+    return (True, None)
+    
+@shared_task()
+def video_queue():
+    """ 
+    Gets the oldest video ID from the queue and runs video_download() on it.
+    """
+    from .nosql import get_nosql
+    
+    videoId = get_nosql().queue_getNext()
+    
+    if videoId:
+        videoId = videoId['id']
+    else:
+        return None
+    
+    status, reason = video_download(videoId)
+    
+    if status:
+        get_nosql().queue_deleteQueue(videoId)
+        return True
+    else:
+        get_nosql().queue_setFailed(videoId, reason)
+        return False
+
 @shared_task()
 def websub_subscribe_callback(channelId):
     import requests
@@ -85,8 +127,13 @@ def websub_process_data():
             if state == 'added':
                 if not get_nosql().check_exists(videoId):  # if video not exists
                    get_nosql().queue_insertQueue(videoId, 'WebSub')
+                   # note for future me
+                   # the websub notifications report ALL videos, including shorts and livestreams
+                   # so if you are going to work on individual video downloading make sure you filter them!
 
             elif state == 'removed':
+                # we currently do not do anything with removed videos
+                # but the idea is to trigger a full channel mirror in case a creator started to mass delete videos
                 pass
             
         get_nosql().websub_deletePostProcessing(_id)
@@ -114,10 +161,33 @@ def websub_renew_expiring(hours=6):
         websub_subscribe_callback.delay(data.get('channel'))
         
         # limit amount of subscribe requests to spread out the requests over time
+        # with an expiration pivot of 6h and a maximum validity of 5 days we can currently handle 3072 channels
         count = count + 1
-        if count >= 100:
+        if count >= 256:
             break
 
+@shared_task()
+def playlist_to_queue():
+    from .nosql import get_nosql
+    import random
+    from datetime import datetime, timedelta
+    
+    pivot = datetime.utcnow() - timedelta(days=3)  # calculates 3 days before now
+    channels = list(get_nosql().list_all_channels(active=True))
+    random.shuffle(channels)  # randomize channelId order because otherwise the queue will follow the channel order as well
+    
+    for channel in channels:
+        info = get_nosql().get_channel_info(channel)
+        
+        # if last_run not set or last_run is older than the pivot (indicating it has not been updated)
+        if not info.get('last_run') or info.get('last_run') < pivot:
+            # skip channel
+            continue
+        
+        for item in info['playlist']['entries']:
+            videoId = item['id']
+            get_nosql().queue_insertQueue(videoId, 'Playlist mirroring')
+
 ##########################################
 #              TASK MODULES              #
 ##########################################

ayta/templates/admin/channels.html

@@ -15,6 +15,18 @@
   </div>
 </div>
 <div class="row">
+  <div class="col s12 l4 m-4">
+    <div class="card">
+      <div class="card-content">
+        <span class="card-title">Direct actions</span>
+        <form class="mt-4" method="post">
+          <button class="btn mb-2 green" type="submit" name="task" value="playlist-queue">Playlist to Queue</button>
+          <br>
+          <span class="supporting-text">Forcerun playlist to queue task</span>
+        </form>
+      </div>
+    </div>
+  </div>
   <div class="col s12 l4 m-4">
 	<div class="card">
       <div class="card-content">
@@ -38,7 +50,7 @@
               });
             </script>
 		  </div>
-          <button class="btn mt-4" type="submit" name="action" value="add_channel">Add</button>
+          <button class="btn mt-4" type="submit" name="task" value="add_channel">Add</button>
 	    </form>
       </div>
     </div>

ayta/templates/admin/index.html

@@ -11,79 +11,89 @@
 <div class="divider"></div>
 <div class="row">
   <div class="col s12">
-	<h5>Global channel options</h5>
+    <h5>Global channel options</h5>
   </div>
 </div>
 <div class="row">
   <div class="col s6 l4 m-4">
-	<a href="{{ url_for('admin.system') }}">
-	  <div class="card black-text">
+    <a href="{{ url_for('admin.system') }}">
+      <div class="card black-text">
         <div class="card-content">
           <span class="card-title">System</span>
-		  <p class="grey-text">Internal system settings</p>
+          <p class="grey-text">Internal system settings</p>
         </div>
       </div>
-	</a>
+    </a>
   </div>
   <div class="col s6 l4 m-4">
-	<a href="{{ url_for('admin.channels') }}">
-	  <div class="card black-text">
+    <a href="{{ url_for('admin.channels') }}">
+      <div class="card black-text">
         <div class="card-content">
           <span class="card-title">Channels</span>
-		  <p class="grey-text">Manage channels in the system</p>
+          <p class="grey-text">Manage channels in the system</p>
         </div>
       </div>
-	</a>
+    </a>
   </div>
   <div class="col s6 l4 m-4">
-	<a href="{{ url_for('admin.runs') }}">
-	  <div class="card black-text">
+    <a href="{{ url_for('admin.runs') }}">
+      <div class="card black-text">
         <div class="card-content">
           <span class="card-title">Archive runs</span>
-		  <p class="grey-text">Look at the cron run logs</p>
+          <p class="grey-text">Look at the cron run logs</p>
         </div>
       </div>
-	</a>
+    </a>
   </div>
   <div class="col s6 l4 m-4">
-	<a href="{{ url_for('admin.websub') }}">
-	  <div class="card black-text">
+    <a href="{{ url_for('admin.websub') }}">
+      <div class="card black-text">
         <div class="card-content">
           <span class="card-title">WebSub</span>
-		  <p class="grey-text">Edit WebSub YouTube links</p>
+          <p class="grey-text">Edit WebSub YouTube links</p>
         </div>
       </div>
-	</a>
+    </a>
   </div>
   <div class="col s6 l4 m-4">
-	<a href="{{ url_for('admin.reports') }}">
-	  <div class="card black-text">
+    <a href="{{ url_for('admin.reports') }}">
+      <div class="card black-text">
         <div class="card-content">
           <span class="card-title">Reports</span>
-		  <p class="grey-text">View user reports</p>
+          <p class="grey-text">View user reports</p>
         </div>
       </div>
-	</a>
+    </a>
   </div>
   <div class="col s6 l4 m-4">
-	<a href="{{ url_for('admin.queue') }}">
-	  <div class="card black-text">
+    <a href="{{ url_for('admin.queue') }}">
+      <div class="card black-text">
         <div class="card-content">
           <span class="card-title">Queue</span>
-		  <p class="grey-text">Video download queue and API access</p>
+          <p class="grey-text">Video download queue and API access</p>
         </div>
       </div>
-	</a>
+    </a>
   </div>
   <div class="col s6 l4 m-4">
-	<a href="{{ url_for('admin.users') }}">
-	  <div class="card black-text">
+    <a href="{{ url_for('admin.users') }}">
+      <div class="card black-text">
         <div class="card-content">
           <span class="card-title">Users</span>
-		  <p class="grey-text">Authenticated users</p>
+          <p class="grey-text">Authenticated users</p>
         </div>
       </div>
-	</a>
+    </a>
+  </div>
+  <div class="col s6 l4 m-4">
+    <a href="{{ url_for('admin.workers') }}">
+      <div class="card black-text">
+        <div class="card-content">
+          <span class="card-title">Workers</span>
+          <p class="grey-text">Worker and task management</p>
+        </div>
+      </div>
+    </a>
   </div>
 </div>
 {% endblock %}

ayta/templates/admin/queue.html

@@ -11,67 +11,72 @@
 <div class="divider"></div>
 <div class="row">
   <div class="col s12">
-	<h5>Options</h5>
+    <h5>Options</h5>
   </div>
 </div>
 <div class="row">
   <div class="col s12 l4 m-4">
-	<div class="card">
+    <div class="card">
       <div class="card-content">
         <span class="card-title">Direct actions</span>
-		<form class="mt-4" method="post" onsubmit="return confirm('Are you sure?');">
+        <form class="mt-4" method="post" onsubmit="return confirm('Are you sure?');">
           <button class="btn mb-2 red" type="submit" name="task" value="empty-queue">Empty Queue</button>
-		  <br>
-		  <span class="supporting-text">Removes all queued ids</span>
-	    </form>
-		<form class="mt-4" method="post" onsubmit="return confirm('Are you sure?');">
+          <br>
+          <span class="supporting-text">Removes all queued ids</span>
+        </form>
+        <form class="mt-4" method="post" onsubmit="return confirm('Are you sure?');">
           <button class="btn mb-2" type="submit" name="task" value="clean-retired">Clean retired</button>
-		  <br>
-		  <span class="supporting-text">Prunes all deactivated endpoints, but keeps last 3 days</span>
-	    </form>
-		
+          <br>
+          <span class="supporting-text">Prunes all deactivated endpoints, but keeps last 3 days</span>
+        </form>
+        <form class="mt-4 input-field" method="post" onsubmit="return confirm('Are you sure?');">
+          <input type="number" style="width: 80px" value="1" name="value" min="1" max="99">
+          <button class="btn mb-2 green" type="submit" name="task" value="queue-run-once">Download oldest queued</button>
+          <br>
+          <span class="supporting-text">Will download the oldest queued video ID</span>
+        </form>
       </div>
     </div>
   </div>
   <div class="col s12 l4 m-4">
-	<div class="card">
+    <div class="card">
       <div class="card-content">
         <span class="card-title">Create new endpoint</span>
-	    <form method="post">
-		  <div class="row">
+        <form method="post">
+          <div class="row">
             <div class="col s12 input-field">
               <input placeholder="Custom endpoint" name="value" type="text" class="validate" minlength="12">
-			  <span class="supporting-text">Leaving this empty will create a random secure string</span>
-	        </div>
-			<div class="col s12 input-field">
+              <span class="supporting-text">Leaving this empty will create a random secure string</span>
+            </div>
+            <div class="col s12 input-field">
               <input placeholder="Description" name="description" type="text" class="validate" minlength="8" maxlength="64" required>
-			  <span class="supporting-text">Description for the endpoint for better administration</span>
-	        </div>
+              <span class="supporting-text">Description for the endpoint for better administration</span>
+            </div>
           <button class="btn mt-4" type="submit" name="task" value="add-endpoint">Create</button>
-		  </div>
-	    </form>
+          </div>
+        </form>
       </div>
     </div>
   </div>
   <div class="col s12 l4 m-4">
-	<div class="card">
+    <div class="card">
       <div class="card-content">
         <span class="card-title">Queue manually</span>
-	    <form method="post">
-		  <div class="row">
+        <form method="post">
+          <div class="row">
             <div class="col s12 input-field">
               <input placeholder="Youtube video ID" name="value" type="text" class="validate" minlength="11" maxlength="11" required>
-			  <span class="supporting-text">Must be a valid Youtube video ID</span>
-	        </div>
-			<div class="col s12 mt-5 input-field">
+              <span class="supporting-text">Must be a valid Youtube video ID</span>
+            </div>
+            <div class="col s12 mt-5 input-field">
               <div class="switch">
-		        <label>Queue<input type="checkbox" value="direct" name="value" disabled><span class="lever"></span>Direct</label>
-				<span class="supporting-text">Queue up or start directly</span>
-		      </div>
-	        </div>
+                <label>Queue<input type="checkbox" value="direct" name="direct"><span class="lever"></span>Direct</label>
+                <span class="supporting-text">Queue up or start directly</span>
+              </div>
+            </div>
           <button class="btn mt-4" type="submit" name="task" value="manual-queue">Queue</button>
-		  </div>
-	    </form>
+          </div>
+        </form>
       </div>
     </div>
   </div>
@@ -79,10 +84,10 @@
 <div class="divider"></div>
 <div class="row">
   <div class="col s6 l9">
-	<h5>Registered endpoints</h5>
+    <h5>Registered endpoints</h5>
   </div>
   <div class="col s6 l3 m-4 input-field">
-	<input id="filter_query" type="text">
+    <input id="filter_query" type="text">
     <label for="filter_query">Filter results</label>
   </div>
 </div>
@@ -91,30 +96,30 @@
     <table class="striped highlight responsive-table">
       <thead>
         <tr>
-		  <th>Actions</th>
-		  <th>id</th>
-		  <th>description</th>
-		  <th>status</th>
-		  <th>created_time</th>
-		  <th>retired_time</th>
+          <th>Actions</th>
+          <th>id</th>
+          <th>description</th>
+          <th>status</th>
+          <th>created_time</th>
+          <th>retired_time</th>
         </tr>
       </thead>
       <tbody>
-	    {% for endpoint in endpoints %}
+        {% for endpoint in endpoints %}
         <tr class="filterable">
-		  <td>
-		    <form method="post">
-			  <input type="text" value="{{ endpoint.get('id') }}" name="value" hidden>
-		      <button class="btn-small waves-effect waves-light" type="submit" name="task" value="retire" title="Retire endpoint" {% if endpoint.get('status') != 'active' %}disabled{% endif %}>🗑️</button>
-			</form>
-		  </td>
-		  <td>{{ endpoint.get('id') }}</td>
-		  <td>{{ endpoint.get('description') }}</td>
-		  <td>{{ endpoint.get('status') }}</td>
-		  <td>{{ endpoint.get('created_time') }}</td>
-		  <td>{{ endpoint.get('retired_time') }}</td>
+          <td>
+            <form method="post">
+              <input type="text" value="{{ endpoint.get('id') }}" name="value" hidden>
+              <button class="btn-small waves-effect waves-light" type="submit" name="task" value="retire" title="Retire endpoint" {% if endpoint.get('status') != 'active' %}disabled{% endif %}>🗑️</button>
+            </form>
+          </td>
+          <td>{{ endpoint.get('id') }}</td>
+          <td>{{ endpoint.get('description') }}</td>
+          <td>{{ endpoint.get('status') }}</td>
+          <td>{{ endpoint.get('created_time') }}</td>
+          <td>{{ endpoint.get('retired_time') }}</td>
         </tr>
-		{% endfor %}
+        {% endfor %}
       </tbody>
     </table>
   </div>
@@ -122,10 +127,10 @@
 <div class="divider"></div>
 <div class="row">
   <div class="col s6 l9">
-	<h5>Queued ID's</h5>
+    <h5>Queued ID's</h5>
   </div>
   <div class="col s6 l3 m-4 input-field">
-	<input id="filter_query" type="text">
+    <input id="filter_query" type="text">
     <label for="filter_query">Filter results</label>
   </div>
 </div>
@@ -134,28 +139,35 @@
     <table class="striped highlight responsive-table">
       <thead>
         <tr>
-		  <th>Actions</th>
-		  <th>id</th>
-		  <th>endpoint</th>
-		  <th>status</th>
-		  <th>created_time</th>
+          <th>Actions</th>
+          <th>id</th>
+          <th>endpoint</th>
+          <th>status</th>
+          <th>created_time</th>
+          <th>fail_reason</th>
         </tr>
       </thead>
       <tbody>
-	    {% for id in queue %}
+        {% for id in queue %}
         <tr class="filterable">
-		  <td>
-		    <form method="post">
-			  <input type="text" value="{{ id.get('id') }}" name="value" hidden>
-		      <button class="btn-small waves-effect waves-light" type="submit" name="task" value="delete-queue" title="Delete from queue" {% if id.get('status') != 'queued' %}disabled{% endif %}>🗑️</button>
-			</form>
-		  </td>
-		  <td>{{ id.get('id') }}</td>
-		  <td>{{ id.get('endpoint') }}</td>
-		  <td>{{ id.get('status') }}</td>
-		  <td>{{ id.get('created_time') }}</td>
+          <td>
+            <form method="post">
+              <input type="text" value="{{ id.get('id') }}" name="value" hidden>
+              <button class="btn-small waves-effect waves-light" type="submit" name="task" value="delete-queue" title="Delete from queue" {% if id.get('status') == 'working' %}disabled{% endif %}>🗑️</button>
+            </form>
+            <form method="post">
+              <input type="text" value="{{ id.get('id') }}" name="value" hidden>
+              <button class="btn-small waves-effect waves-light" type="submit" name="task" value="run-download" title="Run download task" disabled>⏩</button>
+			  <!-- This function fill not work until the download queue and video download process is rewritten -->
+            </form>
+          </td>
+          <td>{{ id.get('id') }}</td>
+          <td>{{ id.get('endpoint') }}</td>
+          <td>{{ id.get('status') }}</td>
+          <td>{{ id.get('created_time') }}</td>
+          <td><textarea class="info">{{ id.get('fail_reason') }}</textarea></td>
         </tr>
-		{% endfor %}
+        {% endfor %}
       </tbody>
     </table>
   </div>

ayta/templates/admin/workers.html

@@ -0,0 +1,80 @@
+{% extends 'material_base.html' %}
+{% block title %}Workers administration page{% endblock %}
+{% block description %}Workers administration page of the AYTA system{% endblock %}
+
+{% block content %}
+<div class="row">
+  <div class="col s12">
+    <h4>Workers administration page</h4>
+  </div>
+</div>
+<div class="divider"></div>
+<div class="row">
+  <div class="col s12">
+    <h5>Options</h5>
+  </div>
+</div>
+<form method="POST">
+      <input title="test-sleep" type="submit" value="test-sleep" name="task">
+</form>
+<div class="divider"></div>
+<div class="row">
+  <div class="col s12 m-4">
+    <h5>Reserved tasks per worker</h5>
+    <p>Usually 4 tasks per worker</p>
+	{% if reserved is none %}
+	<h6>No workers with reserved tasks, are there any workers with stuck tasks or are they even online?</h6>
+	{% else %}
+    {% for worker in reserved %}
+    <span>{{ worker }}</span>
+    <table class="striped highlight responsive-table" style=" border: 1px solid black;">
+      <thead>
+        <tr>
+          <th>ID</th>
+          <th>Task</th>
+          <th>Arguments</th>
+        </tr>
+      </thead>
+      <tbody>
+        {% for task in reserved[worker] %}
+        <tr>
+          <td>{{ task.get('id') }}</td>
+          <td>{{ task.get('name') }}</td>
+          <td>{{ task.get('args') }} {{ task.get('kwargs') }}</td>
+        </tr>
+        {% endfor %}
+      </tbody>
+    </table>
+    {% endfor %}
+	{% endif %}
+  </div>
+  <div class="col s12 m-4">
+    <h5>Current workers and processing tasks</h5>
+	{% if tasks is none %}
+	<h6>No workers with running tasks, are there any workers with stuck tasks or are they even online?</h6>
+	{% else %}
+    {% for worker in tasks %}
+    <span>{{ worker }}</span>
+    <table class="striped highlight responsive-table" style=" border: 1px solid black;">
+      <thead>
+        <tr>
+          <th>ID</th>
+          <th>Task</th>
+          <th>Time started</th>
+        </tr>
+      </thead>
+      <tbody>
+        {% for task in tasks[worker] %}
+        <tr>
+          <td>{{ task.get('id') }}</td>
+          <td>{{ task.get('name') }}</td>
+          <td>{{ task.get('time_start')|epoch_time }}</td>
+        </tr>
+        {% endfor %}
+      </tbody>
+    </table>
+    {% endfor %}
+	{% endif %}
+  </div>
+</div>
+{% endblock %}

ayta/templates/help.html

@@ -25,12 +25,24 @@
 	  </li>
 	  <li class="collection-item">
 	    <a href="{{ url_for('channel.channel', channelId='UCIcgBZ9hEJxHv6r_jDYOMqg') }}"><span class="title">Unus Annus</span></a>
-        <p>Reason: This channel does not exist. (Self removed)</p>
+        <p>Reason: This channel does not exist.</p>
 	  </li>
 	  <li class="collection-item">
 	    <a href="{{ url_for('channel.channel', channelId='UCz1s8aJYSQuaXJCtEi-VWRA') }}"><span class="title">Dutch Legion</span></a>
         <p>Reason: This account has been terminated due to multiple or severe violations of YouTube's policy prohibiting hate speech.</p>
 	  </li>
+	  <li class="collection-item">
+	    <a href="{{ url_for('channel.channel', channelId='UC91-8aNaRbp71UMEb_34ryg') }}"><span class="title">RBMK5000</span></a>
+        <p>Reason: This channel does not exist.</p>
+	  </li>
+	  <li class="collection-item">
+	    <a href="{{ url_for('channel.channel', channelId='UCoPSAT64vfXlulyWd_dPE3Q') }}"><span class="title">Evilfisher2</span></a>
+        <p>Reason: This channel was removed because it violated our Community Guidelines.</p>
+	  </li>
+	  <li class="collection-item">
+	    <a href="{{ url_for('channel.channel', channelId='UCZXkvavD2YKnFCzCkZ-bNPw') }}"><span class="title">mrabhy</span></a>
+        <p>Reason: This channel was removed because it violated our Community Guidelines.</p>
+	  </li>
     </ul>
   </div>
   <div class="col s12 l6 center-align">
@@ -43,6 +55,22 @@
 	    <a href="{{ url_for('channel.channel', channelId='UCzGdxkzULCa9RlD-Q2EZPXQ') }}"><span class="title">Kalashnikov Group</span></a>
         <p>Reason: This account has been terminated for a violation of YouTube's Terms of Service.</p>
 	  </li>
+	  <li class="collection-item">
+	    <a href="{{ url_for('channel.channel', channelId='UCtfg1tENiu3SgGMZVduFmTg') }}"><span class="title">FiberNinja</span></a>
+        <p>Reason: This channel was removed because it violated our Community Guidelines.</p>
+	  </li>
+	  <li class="collection-item">
+	    <a href="{{ url_for('channel.channel', channelId='UCv4VkfbX8YfqodF-4coEEfQ') }}"><span class="title">James Somerton</span></a>
+        <p>Reason: This channel does not exist.</p>
+	  </li>
+	  <li class="collection-item">
+	    <a href="{{ url_for('channel.channel', channelId='UC8XH9kpilkuss4bVeRZD1kw') }}"><span class="title">Plagued Moth</span></a>
+        <p>Reason: This channel was removed because it violated our Community Guidelines.</p>
+	  </li>
+	  <li class="collection-item">
+	    <a href="{{ url_for('channel.channel', channelId='UCxZTTWP0QN7-ch2wW1QeFwg') }}"><span class="title">CowOfTheSea</span></a>
+        <p>Reason: This channel was removed because it violated our Community Guidelines.</p>
+	  </li>
 	</ul>
   </div>
 </div>

ayta/templates/watch/index.html

@@ -27,7 +27,7 @@
   <div class="col s12 l3">
     <p><b>Video by:</b> <a href="{{ url_for('channel.channel', channelId=render.get('info').get('channel_id')) }}">{{ render.get('info').get('uploader') }}</a></p>
     <p><b>Upload date:</b> {{ render.get('info').get('upload_date')|pretty_time }}</p>
-    <p><b>Archive date:</b> {{ render.get('info').get('epoch')|epoch_time }}</p>
+    <p><b>Archive date:</b> {{ render.get('info').get('epoch')|epoch_date }}</p>
     <p><b>Video length:</b> {{ render.get('info').get('duration')|pretty_duration }}</p>
   </div>
   <div class="col s4 l3 center-align">

requirements.txt

@@ -3,9 +3,12 @@
 flask
 flask-caching
 flask-limiter
+flask-sqlalchemy
+flask-migrate
 pymongo
 yt-dlp
 gunicorn
 celery
 sqlalchemy
+requests
 pyjwt[crypto]