作業ログを月次集計表とグラフにまとめる

python --version
python -m pip install -r requirements.txt
python example.py

"""Summarize fictional work-log records by calendar month."""
import csv
import math
from collections import defaultdict
from datetime import date
from decimal import Decimal, InvalidOperation
from pathlib import Path

import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt

BASE = Path(__file__).resolve().parent
HEADERS = ["record_id", "date", "task", "hours"]


def main():
    destination = BASE / "outputs"
    if destination.exists():
        raise ValueError("outputs already exists; rename it before running again.")
    totals = defaultdict(lambda: {"records": 0, "hours": Decimal("0")})
    seen = set()
    with (BASE / "work_log.csv").open(encoding="utf-8-sig", newline="") as file:
        reader = csv.DictReader(file)
        if reader.fieldnames != HEADERS:
            raise ValueError(f"Expected headers: {HEADERS}")
        for line, row in enumerate(reader, start=2):
            if None in row or any(value is None or not value.strip() for value in row.values()):
                raise ValueError(f"Row {line}: missing or extra field.")
            if row["record_id"] in seen:
                raise ValueError(f"Row {line}: duplicate record_id.")
            seen.add(row["record_id"])
            parsed_date = date.fromisoformat(row["date"])
            if parsed_date.isoformat() != row["date"]:
                raise ValueError(f"Row {line}: use YYYY-MM-DD dates.")
            try:
                hours = Decimal(row["hours"])
            except InvalidOperation as error:
                raise ValueError(f"Row {line}: hours must be a number.") from error
            if not hours.is_finite() or hours < 0 or hours.as_tuple().exponent < -2:
                raise ValueError(f"Row {line}: hours must be finite, nonnegative, with at most 2 decimals.")
            month = parsed_date.strftime("%Y-%m")
            totals[month]["records"] += 1
            totals[month]["hours"] += hours
    if not totals:
        raise ValueError("No work records were found.")

    months = sorted(totals)
    figure, axis = plt.subplots(figsize=(8, 4.6), layout="constrained")
    values = [float(totals[month]["hours"]) for month in months]
    if not all(math.isfinite(value) and math.isfinite(value * 1.2) for value in values):
        plt.close(figure)
        raise ValueError("Monthly total is too large to plot as a finite number.")
    bars = axis.bar(months, values, color="#2B6578", width=0.55)
    axis.bar_label(bars, labels=[f"{value:.2f}" for value in values], padding=4)
    axis.set(title="Monthly work hours | Fictional data", xlabel="Calendar month", ylabel="Hours")
    axis.set_ylim(0, max(values) * 1.2 if max(values) > 0 else 1)
    axis.spines[["top", "right"]].set_visible(False)
    axis.set_axisbelow(True)
    axis.grid(axis="y", alpha=0.2)
    destination.mkdir(exist_ok=False)
    with (destination / "monthly_summary.csv").open("x", encoding="utf-8-sig", newline="") as file:
        writer = csv.writer(file)
        writer.writerow(["month", "record_count", "total_hours"])
        for month in months:
            writer.writerow([month, totals[month]["records"], f"{totals[month]['hours']:.2f}"])
    figure.savefig(destination / "monthly_hours.png", dpi=160)
    plt.close(figure)
    total_hours = sum((values["hours"] for values in totals.values()), Decimal("0"))
    print(f"Records: {len(seen)}; months: {len(months)}; hours: {total_hours:.2f}")
    print("Created outputs/monthly_summary.csv and outputs/monthly_hours.png.")


if __name__ == "__main__":
    try:
        main()
    except (OSError, ValueError, csv.Error) as error:
        raise SystemExit(f"Stopped: {error}") from error