add XML NFO file writing support, rework media cleanup deletion, resolves #11

tubesync/sync/migrations/0005_auto_20201219_0312.py

@@ -0,0 +1,18 @@
+# Generated by Django 3.1.4 on 2020-12-19 03:12
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('sync', '0004_source_media_format'),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name='source',
+            name='source_type',
+            field=models.CharField(choices=[('c', 'YouTube channel'), ('i', 'YouTube channel by ID'), ('p', 'YouTube playlist')], db_index=True, default='c', help_text='Source type', max_length=1, verbose_name='source type'),
+        ),
+    ]

tubesync/sync/migrations/0006_source_write_nfo.py

@@ -0,0 +1,18 @@
+# Generated by Django 3.1.4 on 2020-12-19 03:12
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('sync', '0005_auto_20201219_0312'),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name='source',
+            name='write_nfo',
+            field=models.BooleanField(default=False, help_text='Write an NFO file with the media, these may be detected and used by some media servers', verbose_name='write nfo'),
+        ),
+    ]

tubesync/sync/models.py

@@ -1,6 +1,7 @@
 import os
 import uuid
 import json
+from xml.etree import ElementTree
 from datetime import datetime
 from pathlib import Path
 from django.conf import settings
@@ -253,6 +254,11 @@ class Source(models.Model):
         default=False,
         help_text=_('Copy thumbnails with the media, these may be detected and used by some media servers')
     )
+    write_nfo = models.BooleanField(
+        _('write nfo'),
+        default=False,
+        help_text=_('Write an NFO file in XML with the media info, these may be detected and used by some media servers')
+    )
     has_failed = models.BooleanField(
         _('has failed'),
         default=False,
@@ -476,7 +482,38 @@ class Media(models.Model):
             Source.SOURCE_TYPE_YOUTUBE_CHANNEL: 'formats',
             Source.SOURCE_TYPE_YOUTUBE_CHANNEL_ID: 'formats',
             Source.SOURCE_TYPE_YOUTUBE_PLAYLIST: 'formats',
-        }
+        },
+        'categories': {
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL: 'categories',
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL_ID: 'categories',
+            Source.SOURCE_TYPE_YOUTUBE_PLAYLIST: 'categories',
+        },
+        'rating': {
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL: 'average_rating',
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL_ID: 'average_rating',
+            Source.SOURCE_TYPE_YOUTUBE_PLAYLIST: 'average_rating',
+        },
+        'age_limit': {
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL: 'age_limit',
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL_ID: 'age_limit',
+            Source.SOURCE_TYPE_YOUTUBE_PLAYLIST: 'age_limit',
+        },
+        'uploader': {
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL: 'uploader',
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL_ID: 'uploader',
+            Source.SOURCE_TYPE_YOUTUBE_PLAYLIST: 'uploader',
+        },
+        'upvotes': {
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL: 'like_count',
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL_ID: 'like_count',
+            Source.SOURCE_TYPE_YOUTUBE_PLAYLIST: 'like_count',
+        },
+        'downvotes': {
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL: 'dislike_count',
+            Source.SOURCE_TYPE_YOUTUBE_CHANNEL_ID: 'dislike_count',
+            Source.SOURCE_TYPE_YOUTUBE_PLAYLIST: 'dislike_count',
+        },
+
     }
     STATE_UNKNOWN = 'unknown'
     STATE_SCHEDULED = 'scheduled'
@@ -883,6 +920,34 @@ class Media(models.Model):
             return seconds_to_timestr(duration)
         return '??:??:??'
 
+    @property
+    def categories(self):
+        field = self.get_metadata_field('categories')
+        return self.loaded_metadata.get(field, [])
+
+    @property
+    def rating(self):
+        field = self.get_metadata_field('rating')
+        return self.loaded_metadata.get(field, 0)
+
+    @property
+    def votes(self):
+        field = self.get_metadata_field('upvotes')
+        upvotes = self.loaded_metadata.get(field, 0)
+        field = self.get_metadata_field('downvotes')
+        downvotes = self.loaded_metadata.get(field, 0)
+        return upvotes + downvotes
+
+    @property
+    def age_limit(self):
+        field = self.get_metadata_field('age_limit')
+        return self.loaded_metadata.get(field, 0)
+
+    @property
+    def uploader(self):
+        field = self.get_metadata_field('uploader')
+        return self.loaded_metadata.get(field, '')
+
     @property
     def formats(self):
         field = self.get_metadata_field('formats')
@@ -898,6 +963,26 @@ class Media(models.Model):
         media_details = self.format_dict
         return media_format.format(**media_details)
 
+    @property
+    def thumbname(self):
+        filename = self.filename
+        prefix, ext = os.path.splitext(filename)
+        return f'{prefix}.jpg'
+
+    @property
+    def thumbpath(self):
+        return self.source.directory_path / self.thumbname
+
+    @property
+    def nfoname(self):
+        filename = self.filename
+        prefix, ext = os.path.splitext(filename)
+        return f'{prefix}.nfo'
+    
+    @property
+    def nfopath(self):
+        return self.source.directory_path / self.nfoname
+
     @property
     def directory_path(self):
         # If a media_file has been downloaded use its existing directory
@@ -925,6 +1010,98 @@ class Media(models.Model):
             return False
         return os.path.exists(self.media_file.path)
 
+    @property
+    def nfoxml(self):
+        '''
+            Returns an NFO formatted (prettified) XML string.
+        '''
+        nfo = ElementTree.Element('episodedetails')
+        nfo.text = '\n  '
+        # title = media metadata title
+        title = nfo.makeelement('title', {})
+        title.text = str(self.name).strip()
+        title.tail = '\n  '
+        nfo.append(title)
+        # showtitle = source name
+        showtitle = nfo.makeelement('showtitle', {})
+        showtitle.text = str(self.source.name).strip()
+        showtitle.tail = '\n  '
+        nfo.append(showtitle)
+        # ratings = media metadata youtube rating
+        value = nfo.makeelement('value', {})
+        value.text = str(self.rating)
+        value.tail = '\n      '
+        votes = nfo.makeelement('votes', {})
+        votes.text = str(self.votes)
+        votes.tail = '\n    '
+        rating_attrs = {'name': 'youtube', 'max': '5', 'default': 'True'}
+        rating = nfo.makeelement('rating', rating_attrs)
+        rating.text = '\n      '
+        rating.append(value)
+        rating.append(votes)
+        rating.tail = '\n  '
+        ratings = nfo.makeelement('ratings', {})
+        ratings.text = '\n    '
+        ratings.append(rating)
+        ratings.tail = '\n  '
+        nfo.append(ratings)
+        # plot = media metadata description
+        plot = nfo.makeelement('plot', {})
+        plot.text = str(self.description).strip()
+        plot.tail = '\n  '
+        nfo.append(plot)
+        # thumb = local path to media thumbnail
+        thumb = nfo.makeelement('thumb', {})
+        thumb.text = self.thumbname if self.source.copy_thumbnails else ''
+        thumb.tail = '\n  '
+        nfo.append(thumb)
+        # mpaa = media metadata age requirement
+        mpaa = nfo.makeelement('mpaa', {})
+        mpaa.text = str(self.age_limit)
+        mpaa.tail = '\n  '
+        nfo.append(mpaa)
+        # runtime = media metadata duration in seconds
+        runtime = nfo.makeelement('runtime', {})
+        runtime.text = str(self.duration)
+        runtime.tail = '\n  '
+        nfo.append(runtime)
+        # id = media key
+        idn = nfo.makeelement('id', {})
+        idn.text = str(self.key).strip()
+        idn.tail = '\n  '
+        nfo.append(idn)
+        # uniqueid = media key
+        uniqueid_attrs = {'type': 'youtube', 'default': 'True'}
+        uniqueid = nfo.makeelement('uniqueid', uniqueid_attrs)
+        uniqueid.text = str(self.key).strip()
+        uniqueid.tail = '\n  '
+        nfo.append(uniqueid)
+        # studio = media metadata uploader
+        studio = nfo.makeelement('studio', {})
+        studio.text = str(self.uploader).strip()
+        studio.tail = '\n  '
+        nfo.append(studio)
+        # aired = media metadata uploaded date
+        aired = nfo.makeelement('aired', {})
+        upload_date = self.upload_date
+        aired.text = upload_date.strftime('%Y-%m-%d') if upload_date else ''
+        aired.tail = '\n  '
+        nfo.append(aired)
+        # dateadded = date and time media was created in tubesync
+        dateadded = nfo.makeelement('dateadded', {})
+        dateadded.text = self.created.strftime('%Y-%m-%d %H:%M:%S')
+        dateadded.tail = '\n  '
+        nfo.append(dateadded)
+        # genre = any media metadata categories if they exist
+        for category_str in self.categories:
+            genre = nfo.makeelement('genre', {})
+            genre.text = str(category_str).strip()
+            genre.tail = '\n  '
+            nfo.append(genre)
+        nfo[-1].tail = '\n'
+        # Return XML tree as a prettified string
+        return ElementTree.tostring(nfo, encoding='utf8', method='xml').decode('utf8')
+
     def get_download_state(self, task=None):
         if self.downloaded:
             return self.STATE_DOWNLOADED

tubesync/sync/signals.py

@@ -145,20 +145,6 @@ def media_pre_delete(sender, instance, **kwargs):
     if thumbnail_url:
         delete_task_by_media('sync.tasks.download_media_thumbnail',
                              (str(instance.pk), thumbnail_url))
-    # Delete media thumbnail if it exists
-    if instance.thumb:
-        log.info(f'Deleting thumbnail for: {instance} path: {instance.thumb.path}')
-        delete_file(instance.thumb.path)
-    # Delete the media file if it exists
-    if instance.media_file:
-        filepath = instance.media_file.path
-        log.info(f'Deleting media for: {instance} path: {filepath}')
-        delete_file(filepath)
-        # Delete thumbnail copy if it exists
-        barefilepath, fileext = os.path.splitext(filepath)
-        thumbpath = f'{barefilepath}.jpg'
-        log.info(f'Deleting thumbnail for: {instance} path: {thumbpath}')
-        delete_file(thumbpath)
 
 
 @receiver(post_delete, sender=Media)

tubesync/sync/tasks.py

@@ -23,7 +23,8 @@ from background_task.models import Task, CompletedTask
 from common.logger import log
 from common.errors import NoMediaException, DownloadFailedException
 from .models import Source, Media, MediaServer
-from .utils import get_remote_image, resize_image_to_height, delete_file
+from .utils import (get_remote_image, resize_image_to_height, delete_file,
+                    write_text_file)
 
 
 def get_hash(task_name, pk):
@@ -317,11 +318,13 @@ def download_media(media_id):
         media.save()
         # If selected, copy the thumbnail over as well
         if media.source.copy_thumbnails and media.thumb:
-            barefilepath, fileext = os.path.splitext(filepath)
-            thumbpath = f'{barefilepath}.jpg'
             log.info(f'Copying media thumbnail from: {media.thumb.path} '
-                     f'to: {thumbpath}')
+                     f'to: {media.thumbpath}')
-            copyfile(media.thumb.path, thumbpath)
+            copyfile(media.thumb.path, media.thumbpath)
+        # If selected, write an NFO file
+        if media.source.write_nfo:
+            log.info(f'Writing media NFO file to: to: {media.nfopath}')
+            write_text_file(media.nfopath, media.nfoxml)
         # Schedule a task to update media servers
         for mediaserver in MediaServer.objects.all():
             log.info(f'Scheduling media server updates')

tubesync/sync/templates/sync/source.html

@@ -97,6 +97,10 @@
         <td class="hide-on-small-only">Copy thumbnails?</td>
         <td><span class="hide-on-med-and-up">Copy thumbnails?<br></span><strong>{% if source.copy_thumbnails %}<i class="fas fa-check"></i>{% else %}<i class="fas fa-times"></i>{% endif %}</strong></td>
       </tr>
+      <tr title="Should an NFO file be written with the media?">
+        <td class="hide-on-small-only">Write NFO?</td>
+        <td><span class="hide-on-med-and-up">Write NFO?<br></span><strong>{% if source.write_nfo %}<i class="fas fa-check"></i>{% else %}<i class="fas fa-times"></i>{% endif %}</strong></td>
+      </tr>
       {% if source.delete_old_media and source.days_to_keep > 0 %}
       <tr title="Days after which your media from this source will be locally deleted">
         <td class="hide-on-small-only">Delete old media</td>

tubesync/sync/testdata/metadata.json

@@ -3,15 +3,22 @@
     "upload_date":"20170911",
     "license":null,
     "creator":null,
-    "title":"no fancy stuff",
+    "title":"no fancy stuff title",
     "alt_title":null,
-    "description":"no fancy stuff",
+    "description":"no fancy stuff desc",
-    "categories":[],
+    "average_rating": 1.2345,
+    "dislike_count": 123,
+    "like_count": 456,
+    "uploader": "test uploader",
+    "categories":[
+        "test category 1",
+        "test category 2"
+    ],
     "tags":[],
     "subtitles":{},
     "automatic_captions":{},
     "duration":401.0,
-    "age_limit":0,
+    "age_limit":50,
     "annotations":null,
     "chapters":null,
     "formats":[

tubesync/sync/tests.py

@@ -6,6 +6,7 @@
 
 
 import logging
+from datetime import datetime
 from urllib.parse import urlsplit
 from django.conf import settings
 from django.test import TestCase, Client
@@ -430,7 +431,6 @@ class FrontEndTestCase(TestCase):
         response = c.get('/tasks-completed')
         self.assertEqual(response.status_code, 200)
 
-
     def test_mediasevrers(self):
         # Media servers overview page
         c = Client()
@@ -588,7 +588,71 @@ class FilepathTestCase(TestCase):
         self.source.media_format = ('{title}_{key}_{resolution}-{height}x{width}-'
                                     '{acodec}-{vcodec}-{fps}fps-{hdr}.{ext}')
         self.assertEqual(test_media.filename,
-                         'no-fancy-stuff_test_720p-720x1280-opus-vp9-30fps-hdr.mkv')
+                         ('no-fancy-stuff-title_test_720p-720x1280-opus'
+                          '-vp9-30fps-hdr.mkv'))
+
+
+class MediaTestCase(TestCase):
+
+    def setUp(self):
+        # Disable general logging for test case
+        logging.disable(logging.CRITICAL)
+        # Add a test source
+        self.source = Source.objects.create(
+            source_type=Source.SOURCE_TYPE_YOUTUBE_CHANNEL,
+            key='testkey',
+            name='testname',
+            directory='testdirectory',
+            media_format=settings.MEDIA_FORMATSTR_DEFAULT,
+            index_schedule=3600,
+            delete_old_media=False,
+            days_to_keep=14,
+            source_resolution=Source.SOURCE_RESOLUTION_1080P,
+            source_vcodec=Source.SOURCE_VCODEC_VP9,
+            source_acodec=Source.SOURCE_ACODEC_OPUS,
+            prefer_60fps=False,
+            prefer_hdr=False,
+            fallback=Source.FALLBACK_FAIL
+        )
+        # Add some test media
+        self.media = Media.objects.create(
+            key='mediakey',
+            source=self.source,
+            metadata=metadata,
+        )
+        # Fix a created datetime for predictable testing
+        self.media.created = datetime(year=2020, month=1, day=1, hour=1,
+                                      minute=1, second=1)
+
+    def test_nfo(self):
+        expected_nfo = [
+            "<?xml version='1.0' encoding='utf8'?>",
+            '<episodedetails>',
+            '  <title>no fancy stuff title</title>',
+            '  <showtitle>testname</showtitle>',
+            '  <ratings>',
+            '    <rating default="True" max="5" name="youtube">',
+            '      <value>1.2345</value>',
+            '      <votes>579</votes>',
+            '    </rating>',
+            '  </ratings>',
+            '  <plot>no fancy stuff desc</plot>',
+            '  <thumb />',  # media.thumbfile is empty without media existing
+            '  <mpaa>50</mpaa>',
+            '  <runtime>401</runtime>',
+            '  <id>mediakey</id>',
+            '  <uniqueid default="True" type="youtube">mediakey</uniqueid>',
+            '  <studio>test uploader</studio>',
+            '  <aired>2017-09-11</aired>',
+            '  <dateadded>2020-01-01 01:01:01</dateadded>',
+            '  <genre>test category 1</genre>',
+            '  <genre>test category 2</genre>',
+            '</episodedetails>',
+        ]
+        # Compare it line by line
+        test_nfo = self.media.nfoxml.split('\n')
+        for i, line in enumerate(test_nfo):
+            self.assertEqual(line, expected_nfo[i])
 
 
 class FormatMatchingTestCase(TestCase):

tubesync/sync/utils.py

@@ -108,6 +108,14 @@ def file_is_editable(filepath):
     return False
 
 
+def write_text_file(filepath, filedata):
+    if not isinstance(filedata, str):
+        raise ValueError(f'filedata must be a str, got "{type(filedata)}"')
+    with open(filepath, 'wt') as f:
+        bytes_written = f.write(filedata)
+    return bytes_written
+
+
 def delete_file(filepath):
     if file_is_editable(filepath):
         return os.remove(filepath)

tubesync/sync/views.py

@@ -276,7 +276,7 @@ class AddSourceView(CreateView):
     fields = ('source_type', 'key', 'name', 'directory', 'media_format',
               'index_schedule', 'delete_old_media', 'days_to_keep',
               'source_resolution', 'source_vcodec', 'source_acodec', 'prefer_60fps',
-              'prefer_hdr', 'fallback', 'copy_thumbnails')
+              'prefer_hdr', 'fallback', 'copy_thumbnails', 'write_nfo')
     errors = {
         'invalid_media_format': _('Invalid media format, the media format contains '
                                   'errors or is empty. Check the table at the end of '
@@ -334,8 +334,8 @@ class SourceView(DetailView):
     messages = {
         'source-created': _('Your new source has been created. If you have added a '
                             'very large source such as a channel with hundreds of '
-                            'videos it can take several minutes for media to start '
+                            'videos it can take several minutes or up to an hour '
-                            'to appear.'),
+                            'for media to start to appear.'),
         'source-updated': _('Your source has been updated.'),
     }
 
@@ -367,7 +367,7 @@ class UpdateSourceView(UpdateView):
     fields = ('source_type', 'key', 'name', 'directory', 'media_format',
               'index_schedule', 'delete_old_media', 'days_to_keep',
               'source_resolution', 'source_vcodec', 'source_acodec', 'prefer_60fps',
-              'prefer_hdr', 'fallback', 'copy_thumbnails')
+              'prefer_hdr', 'fallback', 'copy_thumbnails', 'write_nfo')
     errors = {
         'invalid_media_format': _('Invalid media format, the media format contains '
                                   'errors or is empty. Check the table at the end of '
@@ -411,7 +411,12 @@ class DeleteSourceView(DeleteView, FormMixin):
             source = self.get_object()
             for media in Media.objects.filter(source=source):
                 if media.media_file:
+                    # Delete the media file
                     delete_file(media.media_file.name)
+                    # Delete thumbnail copy if it exists
+                    delete_file(media.thumbpath)
+                    # Delete NFO file if it exists
+                    delete_file(media.nfopath)
         return super().post(request, *args, **kwargs)
 
     def get_success_url(self):
@@ -556,13 +561,12 @@ class MediaRedownloadView(FormView, SingleObjectMixin):
             self.object.thumb = None
         # If the media file exists on disk, delete it
         if self.object.media_file_exists:
-            filepath = self.object.media_file.path
+            delete_file(self.object.media_file.path)
-            delete_file(filepath)
             self.object.media_file = None
             # If the media has an associated thumbnail copied, also delete it
-            barefilepath, fileext = os.path.splitext(filepath)
+            delete_file(self.object.thumbpath)
-            thumbpath = f'{barefilepath}.jpg'
+            # If the media has an associated NFO file with it, also delete it
-            delete_file(thumbpath)
+            delete_file(self.object.nfopath)
         # Reset all download data
         self.object.downloaded = False
         self.object.downloaded_audio_codec = None
@@ -602,13 +606,12 @@ class MediaSkipView(FormView, SingleObjectMixin):
         delete_task_by_media('sync.tasks.download_media', (str(self.object.pk),))
         # If the media file exists on disk, delete it
         if self.object.media_file_exists:
-            filepath = self.object.media_file.path
             delete_file(self.object.media_file.path)
             self.object.media_file = None
             # If the media has an associated thumbnail copied, also delete it
-            barefilepath, fileext = os.path.splitext(filepath)
+            delete_file(self.object.thumbpath)
-            thumbpath = f'{barefilepath}.jpg'
+            # If the media has an associated NFO file with it, also delete it
-            delete_file(thumbpath)
+            delete_file(self.object.nfopath)
         # Reset all download data
         self.object.downloaded = False
         self.object.downloaded_audio_codec = None