Add CONTENT_ITEM relation

src/tribler/core/components/knowledge/community/tests/test_knowledge_validator.py

@@ -5,6 +5,7 @@
 from tribler.core.components.knowledge.db.knowledge_db import Operation, ResourceType
 
 VALID_TAGS = [
+    'nl',
     'tag',
     'Tag',
     'Тэг',
@@ -13,7 +14,7 @@
 
 INVALID_TAGS = [
     '',
-    'ta',  # less than 3
+    't',
     't' * 51,  # more than 50
 ]
 

src/tribler/core/components/knowledge/db/knowledge_db.py

@@ -51,6 +51,7 @@ class ResourceType(IntEnum):
     # this is a section for extra types
     TAG = 101
     TORRENT = 102
+    CONTENT_ITEM = 103
 
 
 @dataclass

src/tribler/core/components/knowledge/db/tests/test_knowledge_db.py

@@ -539,24 +539,26 @@ def test_get_subjects(self):
             self.db,
             {
                 'infohash1': [
-                    Resource(predicate=ResourceType.TITLE, name='ubuntu', auto_generated=True),
+                    Resource(predicate=ResourceType.CONTENT_ITEM, name='ubuntu', auto_generated=True),
                     Resource(predicate=ResourceType.TAG, name='linux', auto_generated=True),
                 ],
                 'infohash2': [
-                    Resource(predicate=ResourceType.TITLE, name='ubuntu', auto_generated=True),
+                    Resource(predicate=ResourceType.CONTENT_ITEM, name='ubuntu', auto_generated=True),
                     Resource(predicate=ResourceType.TAG, name='linux', auto_generated=True),
                 ],
                 'infohash3': [
-                    Resource(predicate=ResourceType.TITLE, name='debian', auto_generated=True),
+                    Resource(predicate=ResourceType.CONTENT_ITEM, name='debian', auto_generated=True),
                     Resource(predicate=ResourceType.TAG, name='linux', auto_generated=True),
                 ],
             }
         )
 
-        actual = self.db.get_subjects(subject_type=ResourceType.TORRENT, predicate=ResourceType.TITLE, obj='missed')
+        actual = self.db.get_subjects(subject_type=ResourceType.TORRENT, predicate=ResourceType.CONTENT_ITEM,
+                                      obj='missed')
         assert actual == []
 
-        actual = self.db.get_subjects(subject_type=ResourceType.TORRENT, predicate=ResourceType.TITLE, obj='ubuntu')
+        actual = self.db.get_subjects(subject_type=ResourceType.TORRENT, predicate=ResourceType.CONTENT_ITEM,
+                                      obj='ubuntu')
         assert actual == ['infohash1', 'infohash2']
 
         actual = self.db.get_subjects(subject_type=ResourceType.TORRENT, predicate=ResourceType.TAG, obj='linux')
@@ -568,11 +570,11 @@ def test_get_statements(self):
             self.db,
             {
                 'infohash1': [
-                    Resource(predicate=ResourceType.TITLE, name='ubuntu', auto_generated=True),
+                    Resource(predicate=ResourceType.CONTENT_ITEM, name='ubuntu', auto_generated=True),
                     Resource(predicate=ResourceType.TYPE, name='linux', auto_generated=True),
                 ],
                 'infohash2': [
-                    Resource(predicate=ResourceType.TITLE, name='debian', auto_generated=True),
+                    Resource(predicate=ResourceType.CONTENT_ITEM, name='debian', auto_generated=True),
                     Resource(predicate=ResourceType.TYPE, name='linux', auto_generated=True),
                 ],
                 'INFOHASH1': [
@@ -582,7 +584,7 @@ def test_get_statements(self):
         )
 
         expected = [
-            SimpleStatement(subject_type=ResourceType.TORRENT, subject='infohash1', predicate=ResourceType.TITLE,
+            SimpleStatement(subject_type=ResourceType.TORRENT, subject='infohash1', predicate=ResourceType.CONTENT_ITEM,
                             object='ubuntu'),
             SimpleStatement(subject_type=ResourceType.TORRENT, subject='infohash1', predicate=ResourceType.TYPE,
                             object='linux')
@@ -600,11 +602,11 @@ def test_various_queries(self):
             self.db,
             {
                 'infohash1': [
-                    Resource(predicate=ResourceType.TITLE, name='ubuntu'),
+                    Resource(predicate=ResourceType.CONTENT_ITEM, name='ubuntu'),
                     Resource(predicate=ResourceType.TYPE, name='linux'),
                 ],
                 'infohash2': [
-                    Resource(predicate=ResourceType.TITLE, name='debian'),
+                    Resource(predicate=ResourceType.CONTENT_ITEM, name='debian'),
                     Resource(predicate=ResourceType.TYPE, name='linux'),
                 ],
                 'infohash3': [
@@ -632,7 +634,7 @@ def _subjects(subject_type=None, obj='', predicate=None):
 
         assert _subjects(obj='linux') == {'infohash1', 'infohash2', 'infohash3'}
         assert _subjects(predicate=ResourceType.TAG, obj='linux') == {'infohash3'}
-        assert _subjects(predicate=ResourceType.TITLE) == {'infohash1', 'infohash2'}
+        assert _subjects(predicate=ResourceType.CONTENT_ITEM) == {'infohash1', 'infohash2'}
 
     @db_session
     def test_non_existent_misc(self):

src/tribler/core/components/knowledge/knowledge_constants.py

@@ -1,2 +1,2 @@
-MIN_RESOURCE_LENGTH = 3
+MIN_RESOURCE_LENGTH = 2
 MAX_RESOURCE_LENGTH = 50

src/tribler/core/components/knowledge/restapi/knowledge_endpoint.py

@@ -9,8 +9,8 @@
 
 from tribler.core.components.knowledge.community.knowledge_community import KnowledgeCommunity
 from tribler.core.components.knowledge.community.knowledge_payload import StatementOperation
+from tribler.core.components.knowledge.community.knowledge_validator import is_valid_resource
 from tribler.core.components.knowledge.db.knowledge_db import KnowledgeDatabase, Operation, ResourceType
-from tribler.core.components.knowledge.knowledge_constants import MAX_RESOURCE_LENGTH, MIN_RESOURCE_LENGTH
 from tribler.core.components.restapi.rest.rest_endpoint import HTTP_BAD_REQUEST, RESTEndpoint, RESTResponse
 from tribler.core.components.restapi.rest.schema import HandledErrorSchema
 from tribler.core.utilities.utilities import froze_it
@@ -65,24 +65,21 @@ async def update_knowledge_entries(self, request):
             return error_response
 
         # Validate whether the size of the tag is within the allowed range and filter out duplicate tags.
-        tags = set()
         statements = []
+        self._logger.info(f'Statements about {infohash}: {params["statements"]}')
         for statement in params["statements"]:
             obj = statement["object"]
-            if statement["predicate"] == ResourceType.TAG and \
-                    (len(obj) < MIN_RESOURCE_LENGTH or len(obj) > MAX_RESOURCE_LENGTH):
+            if not is_valid_resource(obj):
                 return RESTResponse({"error": "Invalid tag length"}, status=HTTP_BAD_REQUEST)
 
-            if obj not in tags:
-                tags.add(obj)
-                statements.append(statement)
+            statements.append(statement)
 
         self.modify_statements(infohash, statements)
 
         return RESTResponse({"success": True})
 
     @db_session
-    def modify_statements(self, infohash: str, statements):
+    def modify_statements(self, infohash: str, statements: list):
         """
         Modify the statements of a particular content item.
         """
@@ -92,7 +89,9 @@ def modify_statements(self, infohash: str, statements):
         # First, get the current statements and compute the diff between the old and new statements
         old_statements = self.db.get_statements(subject_type=ResourceType.TORRENT, subject=infohash)
         old_statements = {(stmt.predicate, stmt.object) for stmt in old_statements}
+        self._logger.info(f'Old statements: {old_statements}')
         new_statements = {(stmt["predicate"], stmt["object"]) for stmt in statements}
+        self._logger.info(f'New statements: {new_statements}')
         added_statements = new_statements - old_statements
         removed_statements = old_statements - new_statements
 
@@ -109,6 +108,9 @@ def modify_statements(self, infohash: str, statements):
             signature = self.community.sign(operation)
             self.db.add_operation(operation, signature, is_local_peer=True)
 
+        self._logger.info(f'Added statements: {added_statements}')
+        self._logger.info(f'Removed statements: {removed_statements}')
+
     @docs(
         tags=["General"],
         summary="Get tag suggestions for a torrent with a particular infohash.",

src/tribler/core/components/knowledge/rules/knowledge_rules_processor.py

@@ -1,7 +1,6 @@
 import logging
 import queue
 import time
-from asyncio import Event
 from dataclasses import dataclass
 from typing import Optional, Set
 
@@ -187,8 +186,8 @@ async def process_torrent_title(self, infohash: Optional[bytes] = None, title: O
                                  objects=tags)
 
         if content_items := set(extract_only_valid_tags(title, rules=content_items_rules)):
-            self.save_statements(subject_type=ResourceType.TORRENT, subject=infohash_str, predicate=ResourceType.TITLE,
-                                 objects=content_items)
+            self.save_statements(subject_type=ResourceType.TORRENT, subject=infohash_str,
+                                 predicate=ResourceType.CONTENT_ITEM, objects=content_items)
 
         return len(tags) + len(content_items)
 

src/tribler/core/components/knowledge/rules/tests/test_tag_rules_base.py

@@ -4,4 +4,4 @@
 
 def test_extract_only_valid_tags():
     # test that extract_only_valid_tags extracts only valid tags
-    assert set(extract_only_valid_tags('[valid-tag, in va li d]', rules=general_rules)) == {'valid-tag'}
+    assert set(extract_only_valid_tags('[valid-tag, i n v a l i d]', rules=general_rules)) == {'valid-tag'}

src/tribler/core/components/metadata_store/restapi/search_endpoint.py

@@ -50,7 +50,7 @@ def build_snippets(self, search_results: List[Dict]) -> List[Dict]:
             with db_session:
                 content_items: List[str] = self.knowledge_db.get_objects(subject_type=ResourceType.TORRENT,
                                                                          subject=search_result["infohash"],
-                                                                         predicate=ResourceType.TITLE)
+                                                                         predicate=ResourceType.CONTENT_ITEM)
             if content_items:
                 for content_id in content_items:
                     content_to_torrents[content_id].append(search_result)

src/tribler/gui/dialogs/editmetadatadialog.py

@@ -14,7 +14,12 @@
 from tribler.gui.utilities import connect, get_objects_with_predicate, get_ui_file_path, tr
 from tribler.gui.widgets.tagbutton import TagButton
 
-METADATA_TABLE_PREDICATES = [ResourceType.TITLE, ResourceType.DESCRIPTION, ResourceType.DATE, ResourceType.LANGUAGE]
+METADATA_TABLE_PREDICATES = [
+    ResourceType.CONTENT_ITEM,
+    ResourceType.DESCRIPTION,
+    ResourceType.DATE,
+    ResourceType.LANGUAGE
+]
 
 
 class EditMetadataDialog(DialogContainer):

src/tribler/gui/qt_resources/edit_metadata_dialog.ui

@@ -276,7 +276,7 @@ margin: 0px;
         </column>
         <item>
          <property name="text">
-          <string>Title</string>
+          <string>Content Item</string>
          </property>
         </item>
         <item>

src/tribler/gui/tests/test_gui.py

@@ -696,7 +696,7 @@ def test_tags_dialog(window):
     assert widget.content_table.add_tags_dialog.dialog_widget.error_text_label.isVisible()
 
     QTest.keyClick(tags_input, "c")
-    assert tags_input.tags[-1].text == "aac"
+    assert tags_input.tags[-1].text == "ac"
 
     # Test creating a new tag by clicking to the right of the right-most tag
     QTest.mouseClick(tags_input, Qt.LeftButton, pos=tags_input.tags[-1].rect.topRight().toPoint() + QPoint(10, 0))