feat: download files, show cloud video, solf deletion with no reprocessing (#920)

* fix: move upd ports out of MacOS internal Range

* feat: download files, show cloud video, solf deletion with no reprocessing

server/tests/test_cleanup.py

@@ -76,8 +76,10 @@ async def test_cleanup_old_public_data_deletes_old_anonymous_transcripts():
     assert result["transcripts_deleted"] == 1
     assert result["errors"] == []
 
-    # Verify old anonymous transcript was deleted
-    assert await transcripts_controller.get_by_id(old_transcript.id) is None
+    # Verify old anonymous transcript was soft-deleted
+    old = await transcripts_controller.get_by_id(old_transcript.id)
+    assert old is not None
+    assert old.deleted_at is not None
 
     # Verify new anonymous transcript still exists
     assert await transcripts_controller.get_by_id(new_transcript.id) is not None
@@ -150,15 +152,17 @@ async def test_cleanup_deletes_associated_meeting_and_recording():
     assert result["recordings_deleted"] == 1
     assert result["errors"] == []
 
-    # Verify transcript was deleted
-    assert await transcripts_controller.get_by_id(old_transcript.id) is None
+    # Verify transcript was soft-deleted
+    old = await transcripts_controller.get_by_id(old_transcript.id)
+    assert old is not None
+    assert old.deleted_at is not None
 
-    # Verify meeting was deleted
+    # Verify meeting was hard-deleted (cleanup deletes meetings directly)
     query = meetings.select().where(meetings.c.id == meeting_id)
     meeting_result = await get_database().fetch_one(query)
     assert meeting_result is None
 
-    # Verify recording was deleted
+    # Verify recording was hard-deleted (cleanup deletes recordings directly)
     assert await recordings_controller.get_by_id(recording.id) is None
 
 

server/tests/test_transcripts.py

@@ -1,7 +1,8 @@
 import pytest
 
+from reflector.db.recordings import Recording, recordings_controller
 from reflector.db.rooms import rooms_controller
-from reflector.db.transcripts import transcripts_controller
+from reflector.db.transcripts import SourceKind, transcripts_controller
 
 
 @pytest.mark.asyncio
@@ -192,9 +193,93 @@ async def test_transcript_delete(authenticated_client, client):
     assert response.status_code == 200
     assert response.json()["status"] == "ok"
 
+    # API returns 404 for soft-deleted transcripts
     response = await client.get(f"/transcripts/{tid}")
     assert response.status_code == 404
 
+    # But the transcript still exists in DB with deleted_at set
+    transcript = await transcripts_controller.get_by_id(tid)
+    assert transcript is not None
+    assert transcript.deleted_at is not None
+
+
+@pytest.mark.asyncio
+async def test_deleted_transcript_not_in_list(authenticated_client, client):
+    """Soft-deleted transcripts should not appear in the list endpoint."""
+    response = await client.post("/transcripts", json={"name": "testdel_list"})
+    assert response.status_code == 200
+    tid = response.json()["id"]
+
+    # Verify it appears in the list
+    response = await client.get("/transcripts")
+    assert response.status_code == 200
+    ids = [t["id"] for t in response.json()["items"]]
+    assert tid in ids
+
+    # Delete it
+    response = await client.delete(f"/transcripts/{tid}")
+    assert response.status_code == 200
+
+    # Verify it no longer appears in the list
+    response = await client.get("/transcripts")
+    assert response.status_code == 200
+    ids = [t["id"] for t in response.json()["items"]]
+    assert tid not in ids
+
+
+@pytest.mark.asyncio
+async def test_delete_already_deleted_is_idempotent(authenticated_client, client):
+    """Deleting an already-deleted transcript is idempotent (returns 200)."""
+    response = await client.post("/transcripts", json={"name": "testdel_idem"})
+    assert response.status_code == 200
+    tid = response.json()["id"]
+
+    # First delete
+    response = await client.delete(f"/transcripts/{tid}")
+    assert response.status_code == 200
+
+    # Second delete — idempotent, still returns ok
+    response = await client.delete(f"/transcripts/{tid}")
+    assert response.status_code == 200
+
+    # But deleted_at was only set once (not updated)
+    transcript = await transcripts_controller.get_by_id(tid)
+    assert transcript is not None
+    assert transcript.deleted_at is not None
+
+
+@pytest.mark.asyncio
+async def test_deleted_transcript_recording_soft_deleted(authenticated_client, client):
+    """Soft-deleting a transcript also soft-deletes its recording."""
+    from datetime import datetime, timezone
+
+    recording = await recordings_controller.create(
+        Recording(
+            bucket_name="test-bucket",
+            object_key="test.mp4",
+            recorded_at=datetime.now(timezone.utc),
+        )
+    )
+    transcript = await transcripts_controller.add(
+        name="with-recording",
+        source_kind=SourceKind.ROOM,
+        recording_id=recording.id,
+        user_id="randomuserid",
+    )
+
+    response = await client.delete(f"/transcripts/{transcript.id}")
+    assert response.status_code == 200
+
+    # Recording still in DB with deleted_at set
+    rec = await recordings_controller.get_by_id(recording.id)
+    assert rec is not None
+    assert rec.deleted_at is not None
+
+    # Transcript still in DB with deleted_at set
+    tr = await transcripts_controller.get_by_id(transcript.id)
+    assert tr is not None
+    assert tr.deleted_at is not None
+
 
 @pytest.mark.asyncio
 async def test_transcript_mark_reviewed(authenticated_client, client):

server/tests/test_transcripts_download.py

@@ -0,0 +1,36 @@
+import io
+import zipfile
+
+import pytest
+
+
+@pytest.mark.asyncio
+async def test_download_zip_returns_valid_zip(
+    authenticated_client, client, fake_transcript_with_topics
+):
+    """Test that the zip download endpoint returns a valid zip file."""
+    transcript = fake_transcript_with_topics
+    response = await client.get(f"/transcripts/{transcript.id}/download/zip")
+    assert response.status_code == 200
+    assert response.headers["content-type"] == "application/zip"
+
+    # Verify it's a valid zip
+    zip_buffer = io.BytesIO(response.content)
+    with zipfile.ZipFile(zip_buffer) as zf:
+        names = zf.namelist()
+        assert "metadata.json" in names
+        assert "audio.mp3" in names
+
+
+@pytest.mark.asyncio
+async def test_download_zip_requires_auth(client):
+    """Test that zip download requires authentication."""
+    response = await client.get("/transcripts/nonexistent/download/zip")
+    assert response.status_code in (401, 403, 422)
+
+
+@pytest.mark.asyncio
+async def test_download_zip_not_found(authenticated_client, client):
+    """Test 404 for non-existent transcript."""
+    response = await client.get("/transcripts/nonexistent-id/download/zip")
+    assert response.status_code == 404

server/tests/test_transcripts_recording_deletion.py

@@ -1,5 +1,4 @@
 from datetime import datetime, timezone
-from unittest.mock import AsyncMock, patch
 
 import pytest
 
@@ -9,6 +8,7 @@ from reflector.db.transcripts import SourceKind, transcripts_controller
 
 @pytest.mark.asyncio
 async def test_recording_deleted_with_transcript():
+    """Soft-delete: recording and transcript remain in DB with deleted_at set, no files deleted."""
     recording = await recordings_controller.create(
         Recording(
             bucket_name="test-bucket",
@@ -22,16 +22,13 @@ async def test_recording_deleted_with_transcript():
         recording_id=recording.id,
     )
 
-    with patch("reflector.db.transcripts.get_transcripts_storage") as mock_get_storage:
-        storage_instance = mock_get_storage.return_value
-        storage_instance.delete_file = AsyncMock()
+    await transcripts_controller.remove_by_id(transcript.id)
 
-        await transcripts_controller.remove_by_id(transcript.id)
+    # Both should still exist in DB but with deleted_at set
+    rec = await recordings_controller.get_by_id(recording.id)
+    assert rec is not None
+    assert rec.deleted_at is not None
 
-        # Should be called with bucket override
-        storage_instance.delete_file.assert_awaited_once_with(
-            recording.object_key, bucket=recording.bucket_name
-        )
-
-    assert await recordings_controller.get_by_id(recording.id) is None
-    assert await transcripts_controller.get_by_id(transcript.id) is None
+    tr = await transcripts_controller.get_by_id(transcript.id)
+    assert tr is not None
+    assert tr.deleted_at is not None

server/tests/test_transcripts_video.py

@@ -0,0 +1,105 @@
+from datetime import datetime, timedelta, timezone
+from unittest.mock import AsyncMock, patch
+
+import pytest
+
+from reflector.db.transcripts import SourceKind, transcripts_controller
+
+
+@pytest.mark.asyncio
+async def test_video_url_returns_404_when_no_meeting(authenticated_client, client):
+    """Test that video URL returns 404 when transcript has no meeting."""
+    response = await client.post("/transcripts", json={"name": "no-meeting"})
+    assert response.status_code == 200
+    tid = response.json()["id"]
+
+    response = await client.get(f"/transcripts/{tid}/video/url")
+    assert response.status_code == 404
+
+
+@pytest.mark.asyncio
+async def test_video_url_returns_404_when_no_cloud_video(authenticated_client, client):
+    """Test that video URL returns 404 when meeting has no cloud video."""
+    from reflector.db import get_database
+    from reflector.db.meetings import meetings
+
+    meeting_id = "test-meeting-no-video"
+    await get_database().execute(
+        meetings.insert().values(
+            id=meeting_id,
+            room_name="No Video Meeting",
+            room_url="https://example.com",
+            host_room_url="https://example.com/host",
+            start_date=datetime.now(timezone.utc),
+            end_date=datetime.now(timezone.utc) + timedelta(hours=1),
+            room_id=None,
+        )
+    )
+
+    transcript = await transcripts_controller.add(
+        name="with-meeting",
+        source_kind=SourceKind.ROOM,
+        meeting_id=meeting_id,
+        user_id="randomuserid",
+    )
+
+    response = await client.get(f"/transcripts/{transcript.id}/video/url")
+    assert response.status_code == 404
+
+
+@pytest.mark.asyncio
+async def test_video_url_returns_presigned_url(authenticated_client, client):
+    """Test that video URL returns a presigned URL when cloud video exists."""
+    from reflector.db import get_database
+    from reflector.db.meetings import meetings
+
+    meeting_id = "test-meeting-with-video"
+    await get_database().execute(
+        meetings.insert().values(
+            id=meeting_id,
+            room_name="Video Meeting",
+            room_url="https://example.com",
+            host_room_url="https://example.com/host",
+            start_date=datetime.now(timezone.utc),
+            end_date=datetime.now(timezone.utc) + timedelta(hours=1),
+            room_id=None,
+            daily_composed_video_s3_key="recordings/video.mp4",
+            daily_composed_video_duration=120,
+        )
+    )
+
+    transcript = await transcripts_controller.add(
+        name="with-video",
+        source_kind=SourceKind.ROOM,
+        meeting_id=meeting_id,
+        user_id="randomuserid",
+    )
+
+    with patch("reflector.views.transcripts_video.get_source_storage") as mock_storage:
+        mock_instance = AsyncMock()
+        mock_instance.get_file_url = AsyncMock(
+            return_value="https://s3.example.com/presigned-url"
+        )
+        mock_storage.return_value = mock_instance
+
+        response = await client.get(f"/transcripts/{transcript.id}/video/url")
+
+    assert response.status_code == 200
+    data = response.json()
+    assert data["url"] == "https://s3.example.com/presigned-url"
+    assert data["duration"] == 120
+    assert data["content_type"] == "video/mp4"
+
+
+@pytest.mark.asyncio
+async def test_transcript_get_includes_video_fields(authenticated_client, client):
+    """Test that transcript GET response includes has_cloud_video field."""
+    response = await client.post("/transcripts", json={"name": "video-fields"})
+    assert response.status_code == 200
+    tid = response.json()["id"]
+
+    response = await client.get(f"/transcripts/{tid}")
+    assert response.status_code == 200
+    data = response.json()
+    assert data["has_cloud_video"] is False
+    assert data["cloud_video_duration"] is None