Compare la répartition des genres minifigs vs personnages

README.md

@@ -273,8 +273,11 @@ Le script lit `data/intermediate/minifigs_by_set.csv` et `data/intermediate/sets
 
 1. `source .venv/bin/activate`
 2. `python -m scripts.plot_minifig_gender_share`
+3. `python -m scripts.plot_minifig_character_genders`
 
-Le script lit `data/intermediate/minifigs_by_set.csv`, agrège le nombre de minifigs distinctes par genre (basé sur `config/known_character_genders.csv`), écrit `data/intermediate/minifig_gender_counts.csv`, puis trace `figures/step25/minifig_gender_share.png` (donut indiquant la part des personnages féminins et masculins, les genres inconnus étant ignorés pour ce graphique).
+Le premier script lit `data/intermediate/minifigs_by_set.csv`, agrège le nombre de minifigs distinctes par genre (basé sur `config/known_character_genders.csv`), écrit `data/intermediate/minifig_gender_counts.csv`, puis trace `figures/step25/minifig_gender_share.png` (donut indiquant la part des personnages féminins et masculins, les genres inconnus étant ignorés pour ce graphique, étiquettes en valeurs absolues).
+
+Le second script lit `data/intermediate/minifigs_by_set.csv`, agrège le nombre de personnages distincts par genre (femmes/hommes uniquement), sérialise `data/intermediate/minifig_character_gender_counts.csv`, puis trace `figures/step25/minifig_character_gender_share.png` pour comparer la répartition des personnages identifiés.
 
 ### Étape 26 : corrélation pièces / minifigs
 

lib/plots/minifig_gender_share.py

@@ -25,12 +25,7 @@ def plot_minifig_gender_share(counts_path: Path, destination_path: Path) -> None
     genders = [row["gender"] for row in rows]
     counts = [int(row["minifig_count"]) for row in rows]
     colors = [GENDER_COLORS.get(gender.strip().lower(), GENDER_COLORS["unknown"]) for gender in genders]
-    total = sum(counts)
-    labels = []
-    for gender, count in zip(genders, counts):
-        percent = (count / total) * 100 if total else 0
-        label = f"{GENDER_LABELS.get(gender.strip().lower(), 'Inconnu')} ({percent:.1f} %)"
-        labels.append(label)
+    labels = [f"{GENDER_LABELS.get(g.strip().lower(), 'Inconnu')} ({count})" for g, count in zip(genders, counts)]
 
     fig, ax = plt.subplots(figsize=(6, 6))
     ax.pie(
@@ -48,3 +43,31 @@ def plot_minifig_gender_share(counts_path: Path, destination_path: Path) -> None
     fig.tight_layout()
     fig.savefig(destination_path, dpi=160)
     plt.close(fig)
+
+
+def plot_character_gender_share(counts_path: Path, destination_path: Path) -> None:
+    """Trace un diagramme circulaire de la répartition des personnages par genre."""
+    rows = [row for row in load_gender_counts(counts_path) if row["gender"].strip().lower() in ("male", "female")]
+    if not rows:
+        return
+    genders = [row["gender"] for row in rows]
+    counts = [int(row["character_count"]) for row in rows]
+    colors = [GENDER_COLORS.get(gender.strip().lower(), GENDER_COLORS["unknown"]) for gender in genders]
+    labels = [f"{GENDER_LABELS.get(g.strip().lower(), 'Inconnu')} ({count})" for g, count in zip(genders, counts)]
+
+    fig, ax = plt.subplots(figsize=(6, 6))
+    ax.pie(
+        counts,
+        labels=labels,
+        colors=colors,
+        startangle=90,
+        wedgeprops={"linewidth": 0.6, "edgecolor": "#0d0d0d"},
+    )
+    centre_circle = plt.Circle((0, 0), 0.5, fc="white")
+    ax.add_artist(centre_circle)
+    ax.set_title("Répartition des personnages par genre (hors inconnus)")
+
+    ensure_parent_dir(destination_path)
+    fig.tight_layout()
+    fig.savefig(destination_path, dpi=160)
+    plt.close(fig)

lib/rebrickable/minifig_characters.py

@@ -195,6 +195,28 @@ def aggregate_by_gender(rows: Iterable[dict]) -> List[dict]:
     return aggregates
 
 
+def aggregate_characters_by_gender(rows: Iterable[dict]) -> List[dict]:
+    """Compte les personnages distincts par genre (hors genres inconnus)."""
+    gender_by_character: Dict[str, str] = {}
+    counts: Dict[str, int] = defaultdict(int)
+    for row in rows:
+        character = row["known_character"].strip()
+        gender = row.get("gender", "").strip().lower()
+        if character == "":
+            continue
+        if gender not in ("male", "female"):
+            continue
+        if character in gender_by_character:
+            continue
+        gender_by_character[character] = gender
+        counts[gender] += 1
+    aggregates: List[dict] = []
+    for gender in ("female", "male"):
+        if gender in counts:
+            aggregates.append({"gender": gender, "character_count": str(counts[gender])})
+    return aggregates
+
+
 def write_character_counts(path: Path, rows: Sequence[dict]) -> None:
     """Écrit le CSV des comptes par personnage."""
     ensure_parent_dir(path)
@@ -206,6 +228,17 @@ def write_character_counts(path: Path, rows: Sequence[dict]) -> None:
             writer.writerow(row)
 
 
+def write_character_gender_counts(path: Path, rows: Sequence[dict]) -> None:
+    """Écrit le CSV des comptes de personnages par genre."""
+    ensure_parent_dir(path)
+    fieldnames = ["gender", "character_count"]
+    with path.open("w", newline="") as csv_file:
+        writer = csv.DictWriter(csv_file, fieldnames=fieldnames)
+        writer.writeheader()
+        for row in rows:
+            writer.writerow(row)
+
+
 def write_new_characters_by_year(path: Path, rows: Sequence[dict]) -> None:
     """Écrit le CSV des personnages introduits chaque année."""
     ensure_parent_dir(path)

scripts/plot_minifig_character_genders.py

@@ -0,0 +1,27 @@
+"""Trace la répartition des personnages identifiés par genre."""
+
+from pathlib import Path
+
+from lib.plots.minifig_gender_share import plot_character_gender_share
+from lib.rebrickable.minifig_characters import (
+    aggregate_characters_by_gender,
+    load_minifigs_by_set,
+    write_character_gender_counts,
+)
+
+
+MINIFIGS_BY_SET_PATH = Path("data/intermediate/minifigs_by_set.csv")
+COUNTS_PATH = Path("data/intermediate/minifig_character_gender_counts.csv")
+DESTINATION_PATH = Path("figures/step25/minifig_character_gender_share.png")
+
+
+def main() -> None:
+    """Construit le CSV de répartition des personnages par genre et trace le graphique."""
+    rows = load_minifigs_by_set(MINIFIGS_BY_SET_PATH)
+    aggregates = aggregate_characters_by_gender(rows)
+    write_character_gender_counts(COUNTS_PATH, aggregates)
+    plot_character_gender_share(COUNTS_PATH, DESTINATION_PATH)
+
+
+if __name__ == "__main__":
+    main()

tests/test_minifig_characters.py

@@ -5,6 +5,7 @@ from pathlib import Path
 from lib.rebrickable.minifig_characters import (
     aggregate_by_character,
     aggregate_by_gender,
+    aggregate_characters_by_gender,
     aggregate_new_character_sets,
     aggregate_new_characters_by_year,
     aggregate_variations_and_totals,
@@ -12,6 +13,7 @@ from lib.rebrickable.minifig_characters import (
     aggregate_presence_by_year,
     load_sets_enriched,
     write_character_counts,
+    write_character_gender_counts,
     write_new_character_sets_csv,
     write_new_character_sets_markdown,
     write_new_characters_by_year,
@@ -143,6 +145,23 @@ def test_aggregate_by_gender_counts_unique_figs() -> None:
     ]
 
 
+def test_aggregate_characters_by_gender_unique_characters() -> None:
+    """Compter les personnages distincts par genre (ignorer unknown)."""
+    aggregates = aggregate_characters_by_gender(
+        [
+            {"known_character": "A", "gender": "male"},
+            {"known_character": "A", "gender": "male"},
+            {"known_character": "B", "gender": "female"},
+            {"known_character": "C", "gender": "unknown"},
+        ]
+    )
+
+    assert aggregates == [
+        {"gender": "female", "character_count": "1"},
+        {"gender": "male", "character_count": "1"},
+    ]
+
+
 def test_aggregate_new_characters_by_year_limits_range(tmp_path: Path) -> None:
     """Compter les nouveaux personnages par année en respectant la plage."""
     sets_path = tmp_path / "sets_enriched.csv"
@@ -268,6 +287,19 @@ def test_write_character_variations_totals_outputs_csv(tmp_path: Path) -> None:
     assert destination.read_text() == "known_character,gender,variation_count,total_minifigs\nA,male,2,3\nB,female,1,1\n"
 
 
+def test_write_character_gender_counts_outputs_csv(tmp_path: Path) -> None:
+    """Écrit le CSV des comptes de personnages par genre."""
+    destination = tmp_path / "character_gender.csv"
+    rows = [
+        {"gender": "female", "character_count": "2"},
+        {"gender": "male", "character_count": "3"},
+    ]
+
+    write_character_gender_counts(destination, rows)
+
+    assert destination.read_text() == "gender,character_count\nfemale,2\nmale,3\n"
+
+
 def test_write_new_characters_by_year_outputs_csv(tmp_path: Path) -> None:
     """Écrit le CSV des nouveaux personnages par année."""
     destination = tmp_path / "new_characters.csv"

tests/test_minifig_gender_share_plot.py

@@ -4,6 +4,7 @@ import matplotlib
 from pathlib import Path
 
 from lib.plots.minifig_gender_share import plot_minifig_gender_share
+from lib.plots.minifig_gender_share import plot_character_gender_share
 
 
 matplotlib.use("Agg")
@@ -24,3 +25,19 @@ def test_plot_minifig_gender_share(tmp_path: Path) -> None:
 
     assert destination.exists()
     assert destination.stat().st_size > 0
+
+
+def test_plot_character_gender_share(tmp_path: Path) -> None:
+    """Génère le graphique de répartition par genre au niveau personnages."""
+    counts_path = tmp_path / "character_gender.csv"
+    destination = tmp_path / "figures" / "step25" / "minifig_character_gender_share.png"
+    counts_path.write_text(
+        "gender,character_count\n"
+        "male,3\n"
+        "female,2\n"
+    )
+
+    plot_character_gender_share(counts_path, destination)
+
+    assert destination.exists()
+    assert destination.stat().st_size > 0