feat: background tasks manager and book sentence mining for CEFR tests

tests/test_api.py

@@ -1393,11 +1393,10 @@ def test_cefr_testing_and_gap_ingestion():
     assert "clue" in q1
     assert len(q1["options"]) == 4
 
-    # 5. Answer INCORRECTLY -> Automated Dual Gap Ingestion (target + selected option)
     # Pick a distractor from the question's actual options
-    # Find option that is not correct (we test with the first distractor)
-    # Let's inspect options and find one that is wrong
-    wrong_opt = [opt for opt in q1["options"] if opt.lower() != q1.get("term", "").lower()][0]
+    with get_db() as conn:
+        correct_term = conn.execute("SELECT term FROM cefr_questions WHERE id = ?", (q1["id"],)).fetchone()["term"]
+    wrong_opt = [opt for opt in q1["options"] if opt.lower() != correct_term.lower()][0]
 
     ans_wrong = client.post("/api/cefr/answer", json={
         "question_id": q1["id"],
@@ -1576,6 +1575,77 @@ def test_grammar_engine_and_handwritten_journaling():
     assert u1_after["handwritten_logged"] is True
 
 
+def test_background_tasks_and_book_mining():
+    # 1. Unauthenticated check
+    client.cookies.clear()
+    unauth = client.post("/api/tasks/mine-books", json={"level": "A1", "count": 2})
+    assert unauth.status_code == 401
+
+    # 2. Login
+    login_res = client.post("/api/login", json={"username": "testadmin", "password": "testpass12345"})
+    assert login_res.status_code == 200
+
+    # 3. Create a book with known vocabulary
+    book_resp = client.post("/api/books", json={
+        "title": "Adventures in London",
+        "author": "Arthur Conan Doyle",
+        "content": "Chapter 1: The Arrival\nThey decided to invite their neighbor for breakfast. The journey was long and dangerous.\n\nChapter 2: The Mystery\nHe could not remember where he left his expensive ticket.",
+        "source_type": "text"
+    })
+    assert book_resp.status_code == 200
+    book_id = book_resp.json()["id"]
+
+    # 4. Start mining task
+    mine_resp = client.post("/api/tasks/mine-books", json={
+        "book_id": book_id,
+        "level": "A1",
+        "count": 2
+    })
+    assert mine_resp.status_code == 200
+    data = mine_resp.json()
+    assert data["status"] == "ok"
+    assert "task_id" in data
+    task_id = data["task_id"]
+
+    # 5. Check task status via active and ID endpoint
+    active_resp = client.get("/api/tasks/active")
+    assert active_resp.status_code == 200
+
+    # Wait briefly or poll task status
+    import time
+    t_data = None
+    for _ in range(30):
+        t_res = client.get(f"/api/tasks/{task_id}")
+        assert t_res.status_code == 200
+        t_data = t_res.json()["task"]
+        if t_data["status"] in ("completed", "failed", "cancelled"):
+            break
+        time.sleep(0.1)
+
+    assert t_data is not None
+    assert t_data["status"] == "completed"
+    assert t_data["progress_current"] > 0
+
+    # 6. Verify quiz retrieval includes source book metadata
+    quiz_res = client.get("/api/cefr/quiz?level=A1&count=20")
+    assert quiz_res.status_code == 200
+    items = quiz_res.json()["quiz"]
+    book_items = [q for q in items if q.get("source_book_title") == "Adventures in London"]
+    assert len(book_items) > 0
+    assert book_items[0]["source_book_title"] == "Adventures in London"
+    assert "source_chapter_title" in book_items[0]
+
+    # 7. Test task cancellation
+    from app.tasks import create_task
+    with get_db() as conn:
+        task_id_pending = create_task(conn, 1, "mine_books", total=10)
+    cancel_resp = client.post(f"/api/tasks/{task_id_pending}/cancel")
+    assert cancel_resp.status_code == 200
+    assert cancel_resp.json()["status"] == "ok"
+    assert cancel_resp.json()["cancelled"] is True
+
+
+