You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
116 lines
3.8 KiB
116 lines
3.8 KiB
"""Validate category decomposition data for Overton report."""
|
|
from __future__ import annotations
|
|
|
|
import duckdb
|
|
import pytest
|
|
|
|
DB_PATH = "data/motions.db"
|
|
|
|
|
|
@pytest.fixture(scope="module")
|
|
def con():
|
|
c = duckdb.connect(DB_PATH)
|
|
yield c
|
|
c.close()
|
|
|
|
|
|
def test_category_distribution(con):
|
|
"""There are exactly 10 categories and all 3,030 motions are classified."""
|
|
df = con.execute("""
|
|
SELECT category, COUNT(*) as cnt
|
|
FROM right_wing_motions
|
|
WHERE classified = TRUE
|
|
GROUP BY category
|
|
ORDER BY cnt DESC
|
|
""").fetchdf()
|
|
assert len(df) == 10
|
|
assert df["cnt"].sum() == 3030
|
|
assert df[df["category"] == "overig"]["cnt"].values[0] >= 100
|
|
|
|
|
|
def test_category_deltas(con):
|
|
"""Pre/post CS deltas (2024 split) — ALL categories gained, energie/klimaat leads."""
|
|
df = con.execute("""
|
|
SELECT
|
|
category,
|
|
AVG(CASE WHEN year < 2024 THEN centrist_support_strict END) as pre_cs,
|
|
AVG(CASE WHEN year >= 2024 THEN centrist_support_strict END) as post_cs,
|
|
COUNT(*) as n
|
|
FROM right_wing_motions
|
|
WHERE classified = TRUE AND year >= 2017
|
|
GROUP BY category
|
|
""").fetchdf()
|
|
|
|
assert len(df) == 10
|
|
df = df.copy()
|
|
df["delta"] = df["post_cs"] - df["pre_cs"]
|
|
top = df.sort_values("delta", ascending=False)
|
|
assert top.iloc[0]["category"] == "energie/klimaat"
|
|
assert 0.35 < abs(top.iloc[0]["delta"]) < 0.45
|
|
|
|
bottom = df.sort_values("delta", ascending=True)
|
|
assert bottom.iloc[0]["category"] in ("veiligheid/justitie", "onderwijs/wetenschap")
|
|
|
|
assert all(df["delta"] > 0)
|
|
|
|
|
|
def test_yearly_category_cs(con):
|
|
"""Yearly CS per category returns data for every year-category combo."""
|
|
df = con.execute("""
|
|
SELECT year, category, AVG(centrist_support_strict) as cs, COUNT(*) as n
|
|
FROM right_wing_motions
|
|
WHERE classified = TRUE AND year >= 2017
|
|
GROUP BY year, category
|
|
ORDER BY year, category
|
|
""").fetchdf()
|
|
assert len(df) >= 50
|
|
assert df["category"].nunique() == 10
|
|
assert df["year"].nunique() >= 8
|
|
|
|
|
|
def test_quarterly_category_data(con):
|
|
"""Quarterly CS per key categories returns expected shape."""
|
|
key_cats = ["asiel/vreemdelingen", "energie/klimaat", "buitenland/europa",
|
|
"landbouw/natuur", "economie"]
|
|
df = con.execute("""
|
|
SELECT
|
|
EXTRACT(YEAR FROM m.date) AS y,
|
|
CEIL(EXTRACT(MONTH FROM m.date) / 3.0) AS q,
|
|
r.category,
|
|
AVG(r.centrist_support_strict) AS cs,
|
|
COUNT(*) AS n
|
|
FROM right_wing_motions r
|
|
JOIN motions m ON r.motion_id = m.id
|
|
WHERE r.classified = TRUE AND m.date IS NOT NULL
|
|
AND r.category IN ({})
|
|
GROUP BY y, q, r.category
|
|
ORDER BY y, q
|
|
""".format(",".join(f"'{c}'" for c in key_cats))).fetchdf()
|
|
|
|
assert df["category"].nunique() <= 5
|
|
for cat in key_cats:
|
|
assert cat in df["category"].values
|
|
|
|
|
|
def test_qmd_has_domain_decomposition_section():
|
|
"""QMD should have a Domain Decomposition heading after implementation."""
|
|
qmd = open("reports/overton_window/overton_window.qmd").read()
|
|
assert "## Domain Decomposition" in qmd
|
|
|
|
|
|
def test_qmd_has_category_delta_chart():
|
|
"""QMD should have a category delta bar chart cell."""
|
|
qmd = open("reports/overton_window/overton_window.qmd").read()
|
|
assert "chart-7-category-delta" in qmd
|
|
|
|
|
|
def test_qmd_has_category_filter_dropdown():
|
|
"""Chart 1 should have updatemenu for category filtering."""
|
|
qmd = open("reports/overton_window/overton_window.qmd").read()
|
|
assert "updatemenus" in qmd
|
|
|
|
|
|
def test_qmd_has_domain_trajectories():
|
|
"""Quarterly chart should have category trajectories."""
|
|
qmd = open("reports/overton_window/overton_window.qmd").read()
|
|
assert "chart-8-domain-trajectories" in qmd
|
|
|