Reimplement orphaned video listing and more accurate stats

Dockerfile

@@ -4,4 +4,4 @@ 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.
@@ -52,13 +52,22 @@ Extra functionality for further development of features.
 
 ### Stage 3
 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 videos by ID's instead of per channel basis
+- [x] download videos from queue
+- [x] Manage websub callbacks
+- [x] Implement yt-dlp proxy servers, as the VPN is blocked
+- [x] Celery tasks based video downloading
+- [x] Manage websub callbacks
+- [x] Celery task queue views
+- [x] More performant statistics
+- [ ] Retire cronjobs
+- [ ] Retire file based configurations
 
 ### Stage 4
 Mongodb finally has it's limitations.
 - [ ] Migrate to postgresql
+- [ ] Retire time based tasks like channel mirroring
+- [ ] A more comprehensive statistics page, uploads per day, downloads per day and such
 
 ### Stage ...
 Since this is my flagship software which I have developed more features will be added.

ad-hoc-test.py

@@ -1,13 +0,0 @@
-#Import os Library
-import os
-import datetime
-import json
-
-def print_current_time(give=False):
-    time = datetime.datetime.now().replace(microsecond=0)
-    print(f'--- It is {time} ---')
-    return time
-
-with open('lockfile', 'w') as file:
-        data = {'time': print_current_time(), 'PID': os.getpid()}
-        file.write(json.dumps(data, default=str))

ayta/__init__.py

@@ -24,8 +24,19 @@ def create_app(test_config=None):
     # Celery Periodic tasks
     
     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']['Renew WebSub endpoints around every hour'] = {'task': 'ayta.tasks.websub_renew_expiring', 'schedule': 4000}
+    config['CELERY']['beat_schedule']['Process WebSub data around every two minutes'] = {'task': 'ayta.tasks.websub_process_data', 'schedule': 100}
+    config['CELERY']['beat_schedule']['Queue up new videos in static channel playlists about 2 times a day'] = {'task': 'ayta.tasks.playlist_to_queue', 'schedule': 50000}
+    config['CELERY']['beat_schedule']['Download around 123 videos spread out through the day'] = {'task': 'ayta.tasks.video_queue', 'schedule': 700}
+    config['CELERY']['beat_schedule']['Generate new statistiscs about every 3 hours'] = {'task': 'ayta.tasks.generate_statistics', 'schedule': 10000}
+    
+    # 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)
@@ -42,6 +53,7 @@ def create_app(test_config=None):
     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
+    app.jinja_env.filters['datetime_date'] = filters.datetime_date
     
     from .blueprints import watch
     from .blueprints import index

ayta/blueprints/admin.py

@@ -2,7 +2,7 @@ from flask import Blueprint, render_template, request, redirect, url_for, flash,
 from ..nosql import get_nosql
 from ..dlp import checkChannelId, getChannelInfo
 from ..decorators import login_required
-from ..tasks import test_sleep, websub_subscribe_callback, websub_unsubscribe_callback, video_download
+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()
    
@@ -190,14 +201,16 @@ def queue():
             flash(f'Cleaned retired endpoints')
             
         elif task == 'manual-queue':
-            direct = request.form.get('direct', None)
-            
-            if direct:
-                task = video_download.delay(value)
-                flash(f"Started task {task.id}")
+            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:
-                get_nosql().queue_insertQueue(value, 'webui')
-                flash(f'Added to queue: {value}')
+                flash(f'This video ID already exists in the archive: {value}')
             
         elif task == 'delete-queue':
             get_nosql().queue_deleteQueue(value)
@@ -205,14 +218,21 @@ 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'))
         
     endpoints = get_nosql().queue_getEndpoints()
     queue = get_nosql().queue_getQueue()
+    count = len(list(queue.clone()))
    
-    return render_template('admin/queue.html', endpoints=endpoints, queue=queue)
+    return render_template('admin/queue.html', endpoints=endpoints, queue=queue, count=count)
 
 @bp.route('/users', methods=['GET', 'POST'])
 @login_required
@@ -243,7 +263,7 @@ def users():
     return render_template('admin/users.html', users=users)
     
 @bp.route('/workers', methods=['GET', 'POST'])
-#@login_required
+@login_required
 def workers():
     if request.method == 'POST':
         task = request.form.get('task', None)
@@ -253,4 +273,5 @@ def workers():
     celery = current_app.extensions.get('celery')
     
     tasks = celery.control.inspect().active()
-    return render_template('admin/workers.html', tasks=tasks)
+    reserved = celery.control.inspect().reserved()
+    return render_template('admin/workers.html', tasks=tasks, reserved=reserved)

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/blueprints/search.py

@@ -22,5 +22,4 @@ def base():
             
             return render_template('search/index.html', results=results, query=query)
                 
-        
-    return render_template('search/index.html', stats=get_nosql().gen_stats())
+    return render_template('search/index.html', stats=get_nosql().statistics_get())

ayta/filters.py

@@ -27,6 +27,12 @@ def epoch_time(epoch):
         return datetime.fromtimestamp(epoch).strftime('%d %b %Y %H:%M:%S')
     except:
         return None
+        
+def datetime_date(obj):
+    try:
+        return obj.strftime('%d %b %Y %H:%M')
+    except:
+        return None
 
 def current_time(null=None, object=False):
     if object:

ayta/nosql.py

@@ -44,6 +44,7 @@ class Mango:
         self.reports = self.db['reports']
         self.queue_endpoints = self.db['queue_endpoints']
         self.users = self.db['users']
+        self.statistics = self.db['statistics']
             
         self.ensure_indexes()
    
@@ -55,7 +56,11 @@ class Mango:
             'info_json': [
                 ('id_1', True),
                 ('channel_id_1', False),
-                ('uploader_1', False)
+                ('uploader_1', False),
+                ('timestamp', False),
+                ('upload_date', False),
+                ('filesize_approx', False),
+                ('_status', False)
             ],
             'websub_callbacks': [
                 ('id', True)
@@ -92,16 +97,6 @@ class Mango:
     #             general functions          #
     ##########################################
     
-    def gen_stats(self):
-        """ Returns DICT; Channel statistics given the dict key """
-        stats = {}
-        
-        stats['videos'] = self.info_json.count_documents({})
-        stats['channels'] = self.channels.count_documents({})
-        stats['queue'] = self.download_queue.count_documents({})
-        
-        return stats
-            
     def search_videos(self, query):
         # search the index for the requested query. return limited keys
         results = self.info_json.find({"$text": {"$search": query}}, 
@@ -168,8 +163,12 @@ 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):
@@ -194,7 +193,9 @@ 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
+        
+        # The following code I have commented out because the query took too long to proccess, timing the operation out
+        """# Ok lemme explain. Perform inner join from channel collection on channel_id key. match only the fields which are empty. return video id
         pipeline = [{'$match': {'_status': 'available'}},
                     {'$lookup': {'from': 'channels', 'localField': 'channel_id', 'foreignField': 'id', 'as': 'channel'}},
                     {'$match': {'channel': {'$size': 0}}},{'$project': {'id': 1}},
@@ -203,7 +204,20 @@ class Mango:
         results = self.info_json.aggregate(pipeline)
         ids = [result['id'] for result in results]
 
-        return tuple(ids)
+        return tuple(ids)"""
+        
+        # Reimplementing the query but in python, as I do not care about memory usage or data transfer
+        channels = self.channels.find({}, {'_id': 0, 'id': 1})
+        videos = self.info_json.find({'_status': 'available'}, {'_id': 0, 'channel_id': 1, 'id': 1})
+        
+        channels = set([x['id'] for x in channels])
+        orphaned = []
+        
+        for item in videos:
+            if item['channel_id'] not in channels:
+                orphaned.append(item['id'])
+        
+        return tuple(orphaned)
     
     def get_recent_videos(self, count=99):
         """ Returns a SET of YouTube video ID's which have been added last to the info_json collection """
@@ -243,6 +257,10 @@ 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             #
@@ -411,13 +429,16 @@ 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}):
@@ -429,7 +450,60 @@ 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
+            
+    ##########################################
+    #          STATISTICS FUNCTIONS          #
+    ##########################################
+    
+    def statistics_counts(self):
+        counts = {}
+        
+        counts['videos'] = f"{self.info_json.count_documents({'_status': 'available'})} videos in the archive"
+        counts['channels'] = f'{self.channels.count_documents({})} channels in the system'
+        counts['download_queue'] = f"{self.download_queue.count_documents({'status': 'queued'})} queued videos for download"
+        
+        return counts
+        
+    def statistics_sizes(self):
+        sizes = {}
+        data = self.info_json.find({'_status': 'available'}, {'filesize_approx': 1})
+        
+        total = 0
+        
+        for x in data:
+            size = x.get('filesize_approx')
+            if size:
+                total = total + int(size)
 
+        sizes['storage'] = f'{int(total / 1000000000000 + 5)} TB of storage'  # the 5 is the amount of TB of unjust approximation (updated feb 2025)
+
+        return sizes
+        
+    def statistics_generate(self):
+        data = self.statistics_sizes() | self.statistics_counts()
+        data['last_updated'] = self.datetime.utcnow()
+        
+        self.statistics.delete_many({})  # drop existing documents
+        self.statistics.insert_one(data)
+        
+    def statistics_get(self):
+        return self.statistics.find_one({}, {'_id': 0})
+        
 ##########################################
 #           HELPER FUNCTIONS             #
 ##########################################
@@ -453,7 +527,4 @@ 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 = {}
@@ -151,7 +152,6 @@ class OIDC():
                                  
         # 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 by matching the nonce in the signed key

ayta/tasks.py

@@ -21,9 +21,31 @@ def video_download(videoId):
     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
+        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):
@@ -144,6 +166,39 @@ def websub_renew_expiring(hours=6):
         if count >= 256:
             break
 
+@shared_task()
+def playlist_to_queue():
+    """
+    As there is still one cronjob based task running daily in the background, we have to make sure that gets hooked as well into the system.
+    The cronjob task gets the last 50 uploads for all channels and commits the playlist json into the database
+    This task makes sure we append the ID's that we got from the playlist into the download queue.
+    Should idealy be run after the cronjob completes, but I don't want to implement an API that does that, so this gets run twice a day.
+    """
+    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')
+            
+@shared_task()
+def generate_statistics():
+    from .nosql import get_nosql
+    get_nosql().statistics_generate()
+
 ##########################################
 #              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/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="direct"><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>
@@ -91,41 +96,44 @@
     <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>
 </div>
 <div class="divider"></div>
 <div class="row">
-  <div class="col s6 l9">
-	<h5>Queued ID's</h5>
+  <div class="col s4 l8">
+    <h5>Queued ID's</h5>
   </div>
-  <div class="col s6 l3 m-4 input-field">
-	<input id="filter_query" type="text">
+  <div class="col s4 l1">
+    <p>{{ count }} items</p>
+  </div>
+  <div class="col s4 l3 m-4 input-field">
+    <input id="filter_query" type="text">
     <label for="filter_query">Filter results</label>
   </div>
 </div>
@@ -134,28 +142,35 @@
     <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

@@ -19,8 +19,40 @@
 </form>
 <div class="divider"></div>
 <div class="row">
-  <div class="col s12">
-    <h6>Current workers</h6>
+  <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;">
@@ -35,13 +67,14 @@
         {% for task in tasks[worker] %}
         <tr>
           <td>{{ task.get('id') }}</td>
-          <td>{{ task.get('type') }}</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">
@@ -45,6 +57,18 @@
 	  </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>

ayta/templates/search/index.html

@@ -5,68 +5,72 @@
 {% block content %}
 <div class="row">
   <div class="col s12 l3 m-4">
-	<h4>Search the archive</h4>
-	<p>Searching is currently partially working and will probably not work optimally for a long time until the database and backend is fully reworked.</p>
-	<p>In the meantime if you know the channel name and video title you can use local search on <a href="{{ url_for('channel.base') }}">this</a> page</p>
-	<img class="responsive-img" src="{{ url_for('static', filename='img/mongo_meme.png') }}">
-	{% if stats is defined %}
-	<div class="divider"></div>
-	<h5>Stats of the archive</h5>
+    <h4>Search the archive</h4>
+    <p>Searching is currently partially working and will probably not work optimally for a long time until the database and backend is fully reworked.</p>
+    <p>In the meantime if you know the channel name and video title you can use local search on <a href="{{ url_for('channel.base') }}">this</a> page</p>
+    <img class="responsive-img" src="{{ url_for('static', filename='img/mongo_meme.png') }}">
+    {% if stats is not none and stats is defined %}
+    <div class="divider"></div>
+    <h5>Stats of the archive</h5>
     <ul class="collection">
-	  {% for stat in stats %}
+      {% for stat in stats %}
       <li class="collection-item">
-	    <span class="title">{{ stat }}</span>
-		<p>{{ stats[stat] }}</p>
-	  </li>
-	  {% endfor %}
+        <!--<span class="title">{{ stat }}</span>-->
+        {% if stat == 'last_updated' %}
+        Last updated {{ stats[stat]|datetime_date }} UTC
+        {% else %}
+        {{ stats[stat] }}
+        {% endif %}
+      </li>
+      {% endfor %}
     </ul>
-	{% endif %}
+    {% endif %}
   </div>
   <div class="col s12 l9 m-4">
     <div class="row">
-	  <div class="col s6 offset-s3">
+      <div class="col s6 offset-s3">
         <img class="responsive-img" src="{{ url_for('static', filename='img/bing_chilling.png') }}">
-	  </div>
+      </div>
       <div class="col s12 center-align">
         <h5>"A big archive needs a search function." -Sun Tzu</h5>
       </div>
-	</div>
-	<div class="divider"></div>
-	<form method="post" class="">
-	  <div class="row">
-	     <div class="col s12 m-4 input-field">
-	       <input id="first_name" name="query" type="text" placeholder='Search the archive!' maxlength="64" value="{{ query }}">
+    </div>
+    <div class="divider"></div>
+    <form method="post" class="">
+      <div class="row">
+         <div class="col s12 m-4 input-field">
+           <input id="first_name" name="query" type="text" placeholder='Search the archive!' maxlength="64" value="{{ query }}">
            <label for="first_name">Searching in video titles, uploader names and tags.</label>
            <span class="supporting-text">Input will be interpreted as keywords. You can search for literal text by using quotes("). Or exclude by prepending minus (-).</span>
-		 </div>
-		 <div class="col s12 m-4">
+         </div>
+         <div class="col s12 m-4">
          <button class="btn icon-right waves-effect waves-light" type="submit" name="task" value="search">Search</button>
-	     </div>
-	  </div>
-	</form>
-	{% if results is defined %}
-	<div class="divider"></div>
-	
+         </div>
+      </div>
+    </form>
+    {% if results is defined %}
+    <div class="divider"></div>
+    
     <table class="striped highlight responsive-table">
       <thead>
         <tr>
           <th>Title</th>
           <th>Uploader</th>
-		  <th>Date</th>
+          <th>Date</th>
         </tr>
       </thead>
       <tbody>
-		{% for result in results %}
+        {% for result in results %}
         <tr>
           <td><a href="{{ url_for('watch.base') }}?v={{ result.get('id') }}">{{ result.get('title') }}</a></td>
           <td><a href="{{ url_for('channel.channel', channelId=result.get('channel_id')) }}">{{ result.get('uploader') }}</a></td>
-		  <td>{{ result.get('upload_date')|pretty_time }}</td>
+          <td>{{ result.get('upload_date')|pretty_time }}</td>
         </tr>
-		{% endfor %}
+        {% endfor %}
       </tbody>
     </table>
-	{% if results|length == 0 %}<h6>No results. Relax the search terms more please!</h6>{% else %}<p>Not the results you were looking for? Try adding quotes ("") around important words.</p>{% endif %}
-	{% endif %}
+    {% if results|length == 0 %}<h6>No results. Relax the search terms more please!</h6>{% else %}<p>Not the results you were looking for? Try adding quotes ("") around important words.</p>{% endif %}
+    {% endif %}
   </div>
 </div>
 {% endblock %}

ayta/templates/watch/index.html

@@ -6,7 +6,7 @@
 <meta property="og:title" content="{{ render.get('info').get('title') }}" />
 <meta property="og:type" content="website" />
 <meta property="og:url" content="{{ url_for('watch.base') }}?v={{ render.get('info').get('id') }}" />
-<meta property="og:image" content="https://archive.ventilaar.net/videos/automatic/{{ render.get('info').get('channel_id') }}/{{ render.get('info').get('id') }}/{{ render.get('info').get('title') }}.jpg" />
+<meta property="og:image" content="https://archive.ventilaar.net/videos/automatic/{{ render.get('info').get('channel_id') }}/{{ render.get('info').get('id') }}/{{ render.get('info').get('_title_slug') }}.jpg" />
 <meta property="og:description" content="{{ render.get('info').get('description', '')|truncate(100) }}" />
 {% endblock %}
 

one_offs/add_failed_queue_to_unavailable.py

@@ -0,0 +1,20 @@
+from ayta.nosql import Mango
+#import ayta
+#app = ayta.create_app()
+mango = Mango('mongodb://root:example@192.168.66.140:27017')
+
+data = mango.download_queue.find({'status': 'failed'})
+for x in data:
+    vId = x['id']
+    lines = x['fail_reason'].splitlines()
+    error = lines[-1]
+    check = "This video has been removed for violating YouTube's Terms of Service"
+    
+    if check in error:
+        print(vId)
+        mango.info_json.insert_one({'id': vId, '_status': 'unavailable', 
+        '_status_description': f'Video is unavailable because YouTube said: {check}'})
+        mango.queue_deleteQueue(vId)
+    else:
+        print(error)
+print('done')

one_offs/archive_size.py

@@ -0,0 +1,18 @@
+from ayta.nosql import Mango
+#import ayta
+#app = ayta.create_app()
+mango = Mango('mongodb://root:example@192.168.66.140:27017')
+
+data = mango.info_json.find({'_status': 'available'}, {'filesize_approx': 1})
+
+total = 0
+
+for x in data:
+    size = x.get('filesize_approx')
+    if size:
+        total = total + int(size)
+
+# the 5000 is the amount of GB of unjust approximation
+total = int(total / 1000000000 + 5000)
+
+print(f'Approximate size: {total} GB')

one_offs/stats_downloads_per_day.py

@@ -0,0 +1,37 @@
+from ayta.nosql import Mango
+import matplotlib.pyplot as plt
+from datetime import datetime, timedelta
+#import ayta
+#app = ayta.create_app()
+mango = Mango('mongodb://root:example@192.168.66.140:27017')
+
+pivot = datetime.utcnow() - timedelta(days=90)
+pivot = int(pivot.timestamp())
+
+data = mango.info_json.find({'_status': 'available', 'timestamp': {'$gt': pivot}}, {'epoch': 1})
+
+stat = {}
+
+for x in data:
+    epoch = x['epoch']
+    day = datetime.fromtimestamp(epoch).strftime('%Y%m%d')
+    
+    if day not in stat:
+        stat[day] = 1
+    else:
+        stat[day] = stat[day] + 1
+
+dates = list(stat.keys())
+values = list(stat.values())
+
+plt.figure(figsize=(16, 8))  # Set the figure size
+plt.bar(dates, values)       # Create the bar chart
+
+# Customize the x-axis labels to be vertical
+plt.xticks(rotation=45, ha='right')  # Rotate xticklabels by 45 degrees and align them to the right
+plt.xlabel('Date')           # Label for x-axis
+plt.ylabel('Counts')         # Label for y-axis
+plt.title('Bar Graph of Counts by Date')  # Title of the graph
+
+# Display the graph
+plt.show()

one_offs/stats_uploads_per_day.py

@@ -0,0 +1,35 @@
+from ayta.nosql import Mango
+import matplotlib.pyplot as plt
+from datetime import datetime, timedelta
+#import ayta
+#app = ayta.create_app()
+mango = Mango('mongodb://root:example@192.168.66.140:27017')
+
+pivot = '20220101'
+
+data = mango.info_json.find({'_status': 'available', 'upload_date': {'$gt': pivot}}, {'upload_date': 1})
+
+stat = {}
+
+for x in data:
+    day = x['upload_date']
+    
+    if day not in stat:
+        stat[day] = 1
+    else:
+        stat[day] = stat[day] + 1
+
+dates = list(stat.keys())
+values = list(stat.values())
+
+plt.figure(figsize=(16, 8))  # Set the figure size
+plt.bar(dates, values)       # Create the bar chart
+
+# Customize the x-axis labels to be vertical
+plt.xticks(rotation=45, ha='right')  # Rotate xticklabels by 45 degrees and align them to the right
+plt.xlabel('Date')           # Label for x-axis
+plt.ylabel('Counts')         # Label for y-axis
+plt.title('Bar Graph of Counts by Date')  # Title of the graph
+
+# Display the graph
+plt.show()

requirements.txt

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