pymongo requires gcc now?

Dockerfile

@@ -1,4 +1,5 @@
 FROM python:3-alpine
+RUN apk update && apk add python3-dev gcc libc-dev libffi-dev
 WORKDIR /app
 COPY requirements.txt /app
 RUN pip install --no-cache-dir -r requirements.txt

ayta/__init__.py

@@ -41,7 +41,8 @@ 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
 from datetime import datetime
 from secrets import token_urlsafe
 
@@ -190,8 +190,14 @@ def queue():
             flash(f'Cleaned retired endpoints')
             
         elif task == 'manual-queue':
-            get_nosql().queue_insertQueue(value, 'webui')
-            flash(f'Added to queue: {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}')
             
         elif task == 'delete-queue':
             get_nosql().queue_deleteQueue(value)
@@ -234,4 +240,17 @@ 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()
+    return render_template('admin/workers.html', tasks=tasks)

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/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
             
@@ -195,11 +195,14 @@ 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):

ayta/oidc.py

@@ -23,15 +23,18 @@ 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 +42,13 @@ 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 +56,30 @@ 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 +130,33 @@ 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 +5,26 @@ 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
+    return True
+    
+
 @shared_task()
 def websub_subscribe_callback(channelId):
     import requests
@@ -85,8 +105,13 @@ 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)
@@ -114,8 +139,9 @@ 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
 
 ##########################################

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

@@ -65,7 +65,7 @@
 	        </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>
+		        <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>

ayta/templates/admin/workers.html

@@ -0,0 +1,47 @@
+{% 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">
+    <h6>Current workers</h6>
+    {% 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('type') }}</td>
+          <td>{{ task.get('time_start')|epoch_time }}</td>
+        </tr>
+        {% endfor %}
+      </tbody>
+    </table>
+    {% endfor %}
+  </div>
+</div>
+{% endblock %}

ayta/templates/help.html

@@ -43,6 +43,10 @@
 	    <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>
 	</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">