978 add youtube title (#1321)

buzz/db/dao/transcription_dao.py

@@ -250,6 +250,22 @@ class TranscriptionDAO(DAO[Transcription]):
         if not query.exec():
             raise Exception(query.lastError().text())
 
+    def update_transcription_file_and_name(self, id: UUID, file_path: str, name: str | None = None):
+        query = self._create_query()
+        query.prepare(
+            """
+            UPDATE transcription
+            SET file = :file, name = COALESCE(:name, name)
+            WHERE id = :id
+        """
+        )
+
+        query.bindValue(":id", str(id))
+        query.bindValue(":file", file_path)
+        query.bindValue(":name", name)
+        if not query.exec():
+            raise Exception(query.lastError().text())
+
     def update_transcription_name(self, id: UUID, name: str):
         query = self._create_query()
         query.prepare(

buzz/db/service/transcription_service.py

@@ -47,6 +47,9 @@ class TranscriptionService:
                 )
             )
 
+    def update_transcription_file_and_name(self, id: UUID, file_path: str, name: str | None = None):
+        self.transcription_dao.update_transcription_file_and_name(id, file_path, name)
+
     def update_transcription_name(self, id: UUID, name: str):
         self.transcription_dao.update_transcription_name(id, name)
 

buzz/transcriber/file_transcriber.py

@@ -38,12 +38,35 @@ class FileTranscriber(QObject):
     @pyqtSlot()
     def run(self):
         if self.transcription_task.source == FileTranscriptionTask.Source.URL_IMPORT:
-            temp_output_path = tempfile.mktemp()
+            cookiefile = os.getenv("BUZZ_DOWNLOAD_COOKIEFILE")
+
+            # First extract info to get the video title
+            extract_options = {
+                "logger": logging.getLogger(),
+            }
+            if cookiefile:
+                extract_options["cookiefile"] = cookiefile
+
+            try:
+                with YoutubeDL(extract_options) as ydl_info:
+                    info = ydl_info.extract_info(self.transcription_task.url, download=False)
+                    video_title = info.get("title", "audio")
+            except Exception as exc:
+                logging.debug(f"Error extracting video info: {exc}")
+                video_title = "audio"
+
+            # Sanitize title for use as filename
+            video_title = YoutubeDL.sanitize_info({"title": video_title})["title"]
+            # Remove characters that are problematic in filenames
+            for char in ['/', '\\', ':', '*', '?', '"', '<', '>', '|']:
+                video_title = video_title.replace(char, '_')
+
+            # Create temp directory and use video title as filename
+            temp_dir = tempfile.mkdtemp()
+            temp_output_path = os.path.join(temp_dir, video_title)
             wav_file = temp_output_path + ".wav"
             wav_file = str(Path(wav_file).resolve())
 
-            cookiefile = os.getenv("BUZZ_DOWNLOAD_COOKIEFILE")
-
             options = {
                 "format": "bestaudio/best",
                 "progress_hooks": [self.on_download_progress],

buzz/widgets/main_window.py

@@ -385,6 +385,15 @@ class MainWindow(QMainWindow):
         pass
 
     def on_task_completed(self, task: FileTranscriptionTask, segments: List[Segment]):
+        # Update file path in database (important for URL imports where file is downloaded)
+        if task.file_path:
+            logging.debug(f"Updating transcription file path: {task.file_path}")
+            # For URL imports, use the file basename (video title) as the display name
+            name = None
+            if task.source == FileTranscriptionTask.Source.URL_IMPORT:
+                basename = os.path.basename(task.file_path)
+                name = os.path.splitext(basename)[0]  # Remove .wav extension
+            self.transcription_service.update_transcription_file_and_name(task.uid, task.file_path, name)
         self.transcription_service.update_transcription_as_completed(task.uid, segments)
         self.table_widget.refresh_row(task.uid)
 

buzz/widgets/transcription_tasks_table_widget.py

@@ -95,8 +95,8 @@ column_definitions = [
         width=400,
         delegate=RecordDelegate(
             text_getter=lambda record: record.value("name") or (
-                record.value("url") if record.value("url") != ""
-                else os.path.basename(record.value("file"))
+                os.path.basename(record.value("file")) if record.value("file")
+                else record.value("url") or ""
             )
         ),
         hidden_toggleable=False,

tests/widgets/transcription_tasks_table_widget_test.py

@@ -286,16 +286,16 @@ class TestTranscriptionTasksTableWidget:
         text = file_column_def.delegate.callback(record_with_name)
         assert text == "Custom Name"
 
-        # Test fallback to URL when no name
-        record_url_fallback = mock_record({"name": None, "url": "http://example.com/audio.mp3", "file": "/path/file.mp3"})
+        # Test fallback to file basename when no name (file takes priority over URL)
+        record_file_fallback = mock_record({"name": None, "url": "http://example.com/audio.mp3", "file": "/path/file.mp3"})
+        text = file_column_def.delegate.callback(record_file_fallback)
+        assert text == "file.mp3"
+
+        # Test fallback to URL when no name and no file
+        record_url_fallback = mock_record({"name": None, "url": "http://example.com/audio.mp3", "file": ""})
         text = file_column_def.delegate.callback(record_url_fallback)
         assert text == "http://example.com/audio.mp3"
 
-        # Test fallback to filename when no name or URL
-        record_file_fallback = mock_record({"name": None, "url": "", "file": "/path/to/audio.mp3"})
-        text = file_column_def.delegate.callback(record_file_fallback)
-        assert text == "audio.mp3"
-
     def test_notes_column_text_getter(self, widget):
         """Test that notes column displays notes or empty string"""
         notes_column_def = next((col for col in column_definitions if col.column == Column.NOTES), None)