"""Calcul des statistiques de base sur les sets LEGO filtrés."""

import csv
from pathlib import Path
from typing import Iterable, List, Sequence, Tuple

from lib.filesystem import ensure_parent_dir

def read_rows(path: Path) -> List[dict]:
    """Charge un fichier CSV en mémoire sous forme de dictionnaires."""
    with path.open() as csv_file:
        reader = csv.DictReader(csv_file)
        return list(reader)


def write_stats_csv(destination_path: Path, stats: Sequence[Tuple[str, str]]) -> None:
    """Écrit les statistiques dans un CSV à deux colonnes."""
    ensure_parent_dir(destination_path)
    with destination_path.open("w", newline="") as csv_file:
        writer = csv.writer(csv_file)
        writer.writerow(["libelle", "valeur"])
        for label, value in stats:
            writer.writerow([label, value])


def compute_median(values: List[int]) -> float:
    """Calcule la médiane d'une liste de valeurs entières."""
    sorted_values = sorted(values)
    middle = len(sorted_values) // 2
    if len(sorted_values) % 2 == 1:
        return float(sorted_values[middle])
    return (sorted_values[middle - 1] + sorted_values[middle]) / 2


def compute_basic_stats(
    themes: Iterable[dict],
    all_sets: Iterable[dict],
    filtered_sets: Iterable[dict],
    enriched_sets: Iterable[dict],
) -> List[Tuple[str, str]]:
    """Calcule les statistiques principales à partir des sets chargés."""
    themes_list = list(themes)
    all_sets_list = list(all_sets)
    filtered_sets_list = list(filtered_sets)
    enriched_sets_list = list(enriched_sets)

    theme_count_total = len(themes_list)
    total_sets = len(all_sets_list)
    filtered_sets_count = len(filtered_sets_list)
    avg_sets_per_theme = total_sets / theme_count_total
    percent_filtered = (filtered_sets_count / total_sets) * 100
    owned_sets_count = sum(1 for row in enriched_sets_list if row["in_collection"] == "true")
    missing_sets_count = sum(1 for row in enriched_sets_list if row["in_collection"] == "false")
    percent_owned = (owned_sets_count / filtered_sets_count) * 100
    parts_per_set = [int(row["num_parts"]) for row in filtered_sets_list]
    avg_parts_per_set = sum(parts_per_set) / filtered_sets_count
    median_parts_per_set = compute_median(parts_per_set)
    years = [int(row["year"]) for row in filtered_sets_list]
    avg_sets_per_year = filtered_sets_count / len(set(years))
    total_parts = sum(parts_per_set)
    theme_ids_filtered = {row["theme_id"] for row in filtered_sets_list}
    min_year = str(min(years))
    max_year = str(max(years))
    year_counts = {}
    for year in years:
        year_counts[year] = year_counts.get(year, 0) + 1
    prolific_year, prolific_count = max(year_counts.items(), key=lambda item: (item[1], -item[0]))
    richest_set = max(filtered_sets_list, key=lambda row: int(row["num_parts"]))
    lightest_set = min(filtered_sets_list, key=lambda row: int(row["num_parts"]))
    oldest_set = min(filtered_sets_list, key=lambda row: (int(row["year"]), row["set_num"]))
    latest_set = max(filtered_sets_list, key=lambda row: (int(row["year"]), row["set_num"]))
    owned_parts = [int(row["num_parts"]) for row in enriched_sets_list if row["in_collection"] == "true"]
    missing_parts = [int(row["num_parts"]) for row in enriched_sets_list if row["in_collection"] == "false"]
    avg_parts_owned = sum(owned_parts) / len(owned_parts)
    avg_parts_missing = sum(missing_parts) / len(missing_parts)
    total_parts_owned = sum(owned_parts)
    percent_parts_owned = (total_parts_owned / total_parts) * 100

    return [
        ("Nombre total de sets (catalogue complet)", str(total_sets)),
        ("Nombre total de thèmes (catalogue complet)", str(theme_count_total)),
        ("Nombre de sets après filtrage (thèmes ciblés)", str(filtered_sets_count)),
        ("Nombre moyen de sets par thème (catalogue complet)", f"{avg_sets_per_theme:.2f}"),
        ("Pourcentage des sets filtrés vs total", f"{percent_filtered:.2f}%"),
        ("Taux de possession (thèmes filtrés)", f"{percent_owned:.2f}%"),
        ("Sets dans la collection", str(owned_sets_count)),
        ("Sets manquants pour la collection", str(missing_sets_count)),
        ("Nombre moyen de pièces par set (thèmes filtrés)", f"{avg_parts_per_set:.2f}"),
        ("Médiane de pièces par set (thèmes filtrés)", f"{median_parts_per_set:.2f}"),
        ("Nombre moyen de sets commercialisés par an (thèmes filtrés)", f"{avg_sets_per_year:.2f}"),
        ("Total de pièces pour les thèmes filtrés", str(total_parts)),
        ("Total de pièces des sets possédés", str(total_parts_owned)),
        ("Pourcentage de pièces possédées (thèmes filtrés)", f"{percent_parts_owned:.2f}%"),
        ("Nombre de thèmes filtrés", str(len(theme_ids_filtered))),
        ("Première année de sortie (thèmes filtrés)", min_year),
        ("Dernière année de sortie (thèmes filtrés)", max_year),
        ("Année la plus prolifique (thèmes filtrés)", f"{prolific_year} ({prolific_count} sets)"),
        (
            "Set avec le plus de pièces (thèmes filtrés)",
            f"{richest_set['set_num']} - {richest_set['name']} ({richest_set['num_parts']} pièces)",
        ),
        (
            "Set avec le moins de pièces (thèmes filtrés)",
            f"{lightest_set['set_num']} - {lightest_set['name']} ({lightest_set['num_parts']} pièces)",
        ),
        (
            "Set le plus ancien (thèmes filtrés)",
            f"{oldest_set['set_num']} - {oldest_set['name']} ({oldest_set['year']})",
        ),
        (
            "Set le plus récent (thèmes filtrés)",
            f"{latest_set['set_num']} - {latest_set['name']} ({latest_set['year']})",
        ),
        (
            "Nombre moyen de pièces des sets possédés",
            f"{avg_parts_owned:.2f}",
        ),
        (
            "Nombre moyen de pièces des sets manquants",
            f"{avg_parts_missing:.2f}",
        ),
    ]