"""Tests Katalog-Kontext für Progressionsgraph-Matching."""
from planning_catalog_context import (
    ProgressionPlanningCatalogContext,
    PlanningCatalogContextItem,
    catalog_context_has_items,
    merge_catalog_context_into_target,
)
from planning_exercise_profiles import PlanningTargetProfile


def test_catalog_context_has_items():
    assert not catalog_context_has_items(None)
    assert not catalog_context_has_items(ProgressionPlanningCatalogContext())
    assert catalog_context_has_items(
        ProgressionPlanningCatalogContext(
            focus_areas=[PlanningCatalogContextItem(id=3, is_primary=True)],
        )
    )


def test_merge_catalog_context_into_target_sets_focus():
    base = PlanningTargetProfile(sources=["query_only"])
    merged = merge_catalog_context_into_target(
        base,
        ProgressionPlanningCatalogContext(
            focus_areas=[PlanningCatalogContextItem(id=7, is_primary=True)],
            training_types=[PlanningCatalogContextItem(id=2, is_primary=True)],
        ),
    )
    assert merged.focus_area_ids.get(7, 0) > 0.5
    assert merged.training_type_ids.get(2, 0) > 0.5
    assert "catalog_context" in merged.sources


def test_normalize_planning_roadmap_with_catalog_context():
    from progression_graph_planning_artifact import normalize_planning_roadmap_payload

    out = normalize_planning_roadmap_payload(
        {
            "goal_query": "Deeskalation Kinder",
            "planning_catalog_context": {
                "focus_areas": [{"id": 4, "is_primary": True}],
                "target_groups": [{"id": 9, "is_primary": True}],
            },
        }
    )
    assert out["planning_catalog_context"]["focus_areas"][0]["id"] == 4


def test_multistage_qa_splits_llm_highlights_from_fix_hints():
    from planning_path_qa_pipeline import run_multistage_path_qa

    result = run_multistage_path_qa(
        off_topic_steps=[],
        stripped_off_topic=[
            {
                "issue": "roadmap_unfilled",
                "step_index": 1,
                "reasons": ["Keine passende Übung"],
            }
        ],
        gaps=[],
        llm_qa={
            "overall_ok": True,
            "quality_score": 0.88,
            "recommendations": [
                "Gute didaktische Progression",
                "Optional: Vertiefung Koordination",
            ],
        },
        llm_applied=True,
    )
    hints = result["optimization_hints"]
    llm_hints = [h for h in hints if h.get("issue") == "llm_recommendation"]
    fix_hints = [h for h in hints if h.get("issue") != "llm_recommendation"]
    assert len(llm_hints) >= 2
    assert any(h.get("issue") == "roadmap_unfilled" for h in fix_hints)
    assert result["qa_tiers"][2]["recommendations"][0].startswith("Gute didaktische")