Add support for regex video title filtering

.github/workflows/ci.yaml

@@ -4,6 +4,7 @@ env:
   IMAGE_NAME: tubesync
 
 on:
+  workflow_dispatch:
   push:
     branches:
       - main

tubesync/common/templates/pagination.html

@@ -3,7 +3,7 @@
   <div class="col s12">
     <div class="pagination">
       {% for i in paginator.page_range %}
-        <a class="pagenum{% if i == page_obj.number %} currentpage{% endif %}" href="?{% if filter %}filter={{ filter }}&{% endif %}page={{ i }}{% if show_skipped %}&show_skipped=yes{% endif %}">{{ i }}</a>
+        <a class="pagenum{% if i == page_obj.number %} currentpage{% endif %}" href="?{% if filter %}filter={{ filter }}&{% endif %}page={{ i }}{% if show_skipped %}&show_skipped=yes{% endif %}{% if only_skipped %}&only_skipped=yes{% endif %}">{{ i }}</a>
       {% endfor %}
     </div>
   </div>

tubesync/sync/migrations/0020_auto_20231024_1825.py

@@ -0,0 +1,29 @@
+# Generated by Django 3.2.22 on 2023-10-24 17:25
+
+import django.core.validators
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('sync', '0019_add_delete_removed_media'),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name='source',
+            name='filter_text',
+            field=models.CharField(blank=True, default='', help_text='Regex compatible filter string for video titles', max_length=100, verbose_name='filter string'),
+        ),
+        migrations.AlterField(
+            model_name='source',
+            name='auto_subtitles',
+            field=models.BooleanField(default=False, help_text='Accept auto-generated subtitles', verbose_name='accept auto-generated subs'),
+        ),
+        migrations.AlterField(
+            model_name='source',
+            name='sub_langs',
+            field=models.CharField(default='en', help_text='List of subtitles langs to download, comma-separated. Example: en,fr or all,-fr,-live_chat', max_length=30, validators=[django.core.validators.RegexValidator(message='Subtitle langs must be a comma-separated list of langs. example: en,fr or all,-fr,-live_chat', regex='^(\\-?[\\_\\.a-zA-Z]+,)*(\\-?[\\_\\.a-zA-Z]+){1}$')], verbose_name='subs langs'),
+        ),
+    ]

tubesync/sync/models.py

@@ -1,6 +1,7 @@
 import os
 import uuid
 import json
+import re
 from xml.etree import ElementTree
 from collections import OrderedDict
 from datetime import datetime, timedelta
@@ -287,6 +288,13 @@ class IndexSchedule(models.IntegerChoices):
         help_text=_('If "delete old media" is ticked, the number of days after which '
                     'to automatically delete media')
     )
+    filter_text = models.CharField(
+        _('filter string'),
+        max_length=100,
+        default='',
+        blank=True,
+        help_text=_('Regex compatible filter string for video titles')
+    )
     delete_removed_media = models.BooleanField(
         _('delete removed media'),
         default=False,
@@ -538,6 +546,9 @@ def get_example_media_format(self):
         except Exception as e:
             return ''
 
+    def is_regex_match(self, media_item_title):
+        return bool(re.search(self.filter_text,media_item_title))       
+
     def index_media(self):
         '''
             Index the media source returning a list of media metadata as dicts.

tubesync/sync/signals.py

@@ -104,36 +104,57 @@ def media_post_save(sender, instance, created, **kwargs):
     # already been downloaded
     if not instance.downloaded:
         max_cap_age = instance.source.download_cap_date
-        published = instance.published
-        if not published:
-            if not instance.skip:
-                log.warn(f'Media: {instance.source} / {instance} has no published date '
-                         f'set, marking to be skipped')
-                instance.skip = True
-                cap_changed = True
-            else:
+        filter_text = instance.source.filter_text
+        published = instance.published 
+
+        if instance.skip:
+            #currently marked to be skipped, check if skip conditions still apply
+            if not published:
                 log.debug(f'Media: {instance.source} / {instance} has no published date '
-                          f'set but is already marked to be skipped')
+                        f'set but is already marked to be skipped')
+            else:            
+                if max_cap_age and filter_text:
+                    if (published > max_cap_age) and (source.is_regex_match(instance.title)):
+                        # Media was published after the cap date and matches the filter text, but is set to be skipped
+                        print('Has a valid publishing date and matches filter, marking unskipped')
+                        instance.skip = False
+                        cap_changed = True
+                    else:
+                        print('does not have a valid publishing date or filter string, already marked skipped')
+                        log.info(f'Media: {instance.source} / {instance} has no published date '
+                                f'set but is already marked to be skipped')
+                elif max_cap_age:
+                    if published > max_cap_age:
+                        # Media was published after the cap date but is set to be skipped
+                        log.info(f'Media: {instance.source} / {instance} has a valid '
+                                f'publishing date, marking to be unskipped')
+                        instance.skip = False
+                        cap_changed = True
+                elif filter_text:
+                    if source.is_regex_match(instance.title):
+                        # Media matches the filter text but is set to be skipped
+                        log.info(f'Media: {instance.source} / {instance} matches the filter text, marking to be unskipped')
+                        instance.skip = False
+                        cap_changed = True
         else:
-            if max_cap_age:
-                if published > max_cap_age and instance.skip:
-                    # Media was published after the cap date but is set to be skipped
-                    log.info(f'Media: {instance.source} / {instance} has a valid '
-                            f'publishing date, marking to be unskipped')
-                    instance.skip = False
-                    cap_changed = True
-                elif published <= max_cap_age and not instance.skip:
-                    log.info(f'Media: {instance.source} / {instance} is too old for '
-                            f'the download cap date, marking to be skipped')
-                    instance.skip = True
-                    cap_changed = True
+            if not published:
+                log.info(f'Media: {instance.source} / {instance} has no published date, marking to be skipped')
+                instance.skip = True
+                cap_changed = True
             else:
-                if instance.skip:
-                    # Media marked to be skipped but source download cap removed
-                    log.info(f'Media: {instance.source} / {instance} has a valid '
-                            f'publishing date, marking to be unskipped')
-                    instance.skip = False
-                    cap_changed = True
+                if max_cap_age:
+                    if published <= max_cap_age:            
+                        log.info(f'Media: {instance.source} / {instance} is too old for '
+                                f'the download cap date, marking to be skipped')
+                        instance.skip = True
+                        cap_changed = True
+                if filter_text:
+                    if not re.search(filter_text,instance.title):
+                        #media doesn't match the filter text but is not marked to be skipped
+                        log.info(f'Media: {instance.source} / {instance} does not match the filter text')
+                        instance.skip = True
+                        cap_changed = True
+
     # Recalculate the "can_download" flag, this may
     # need to change if the source specifications have been changed
     if instance.metadata:

tubesync/sync/tasks.py

@@ -254,6 +254,11 @@ def download_media_metadata(media_id):
             log.warn(f'Media: {source} / {media} is older than cap age '
                      f'{max_cap_age}, skipping')
             media.skip = True
+    # If the source has a search filter, check the video title matches the filter
+    if source.filter_text and not source.is_regex_match(media.title):
+        # Filter text not found in the media title. Accepts regex string, blank search filter results in this returning false
+        log.warn(f'Media: {source} / {media} does not match {source.filter_text}, skipping')
+        media.skip = True
     # If the source has a cut-off check the upload date is within the allowed delta
     if source.delete_old_media and source.days_to_keep > 0:
         if not isinstance(media.published, datetime):

tubesync/sync/templates/sync/media.html

@@ -64,5 +64,5 @@ <h1 class="truncate">Media</h1>
   </div>
   {% endfor %}
 </div>
-{% include 'pagination.html' with pagination=sources.paginator filter=source.pk show_skipped=show_skipped %}
+{% include 'pagination.html' with pagination=sources.paginator filter=source.pk show_skipped=show_skipped only_skipped=only_skipped%}
 {% endblock %}

tubesync/sync/templates/sync/source.html

@@ -43,6 +43,10 @@ <h1 class="truncate">Source <strong>{{ source.name }}</strong></h1>
         <td class="hide-on-small-only">Directory</td>
         <td><span class="hide-on-med-and-up">Directory<br></span><strong>{{ source.directory }}</strong></td>
       </tr>
+      <tr title="Filter text">
+        <td class="hide-on-small-only">Filter text</td>
+        <td><span class="hide-on-med-and-up">Filter text<br></span><strong>{{ source.filter_text }}</strong></td>
+      </tr>
       <tr title="Media file name format to use for saving files">
         <td class="hide-on-small-only">Media format</td>
         <td><span class="hide-on-med-and-up">Media format<br></span><strong>{{ source.media_format }}</strong></td>

tubesync/sync/tests.py

@@ -175,6 +175,7 @@ def test_source(self):
             'directory': 'testdirectory',
             'media_format': settings.MEDIA_FORMATSTR_DEFAULT,
             'download_cap': 0,
+            'filter_text':'.*',
             'index_schedule': 3600,
             'delete_old_media': False,
             'days_to_keep': 14,
@@ -217,6 +218,7 @@ def test_source(self):
             'directory': 'testdirectory',
             'media_format': settings.MEDIA_FORMATSTR_DEFAULT,
             'download_cap': 0,
+            'filter_text':'.*',
             'index_schedule': Source.IndexSchedule.EVERY_HOUR,
             'delete_old_media': False,
             'days_to_keep': 14,
@@ -247,6 +249,7 @@ def test_source(self):
             'directory': 'testdirectory',
             'media_format': settings.MEDIA_FORMATSTR_DEFAULT,
             'download_cap': 0,
+            'filter_text':'.*',
             'index_schedule': Source.IndexSchedule.EVERY_2_HOURS,  # changed
             'delete_old_media': False,
             'days_to_keep': 14,
@@ -1468,6 +1471,29 @@ def test_metadata_20230629(self):
             self.media.get_best_video_format()
             self.media.get_best_audio_format()
 
+    def test_is_regex_match(self):
+
+        self.media.metadata = all_test_metadata['boring']
+        expected_matches = {
+            ('.*'): (True),
+            ('no fancy stuff'): (True),
+            ('No fancy stuff'): (False),
+            ('(?i)No fancy stuff'): (True), #set case insensitive flag
+            ('no'): (True),
+            ('Foo'): (False),
+            ('^(?!.*fancy).*$'): (False),
+            ('^(?!.*funny).*$'): (True),
+            ('(?=.*f.*)(?=.{0,2}|.{4,})'): (True),
+            ('f{4,}'): (False),
+            ('^[^A-Z]*$'): (True),
+            ('^[^a-z]*$'): (False),
+            ('^[^\\s]*$'): (False)
+        }
+
+        for params, expected in expected_matches.items():
+            self.source.filter_text = params
+            expected_match_result = expected
+            self.assertEqual(self.source.is_regex_match(self.media.title), expected_match_result)
 
 class TasksTestCase(TestCase):
     def setUp(self):

tubesync/sync/views.py

@@ -294,7 +294,7 @@ def get_success_url(self):
 
 class EditSourceMixin:
     model = Source
-    fields = ('source_type', 'key', 'name', 'directory', 'media_format',
+    fields = ('source_type', 'key', 'name', 'directory', 'filter_text', 'media_format',
               'index_schedule', 'download_media', 'download_cap', 'delete_old_media',
               'delete_removed_media', 'days_to_keep', 'source_resolution', 'source_vcodec',
               'source_acodec', 'prefer_60fps', 'prefer_hdr', 'fallback', 'copy_thumbnails',