import matplotlib.pyplot as plt
import pandas as pd
import os

# Generowanie przykładowych danych
def generate_sample_data(num_records=100):
    import random
    from datetime import datetime, timedelta

    station_ids = [101, 102, 103, 104]
    pollutants = ["PM10", "WWA"]
    start_date = datetime(2023, 1, 1)
    end_date = datetime(2023, 12, 31)

    data = []
    for _ in range(num_records):
        record = {
            "station_id": random.choice(station_ids),
            "pollutant": random.choice(pollutants),
            "date": (start_date + timedelta(days=random.randint(0, (end_date - start_date).days))).strftime("%Y-%m-%d"),
            "value": round(random.uniform(5, 50), 2),  # Example values
        }
        data.append(record)

    return pd.DataFrame(data)

# Generowanie danych
data = generate_sample_data(100)

# Tworzenie katalogu na wykresy
output_dir = "figs"
os.makedirs(output_dir, exist_ok=True)

# Generowanie wykresów
# Histogramy stężeń dla każdego zanieczyszczenia
pollutants = data["pollutant"].unique()
for pollutant in pollutants:
    subset = data[data["pollutant"] == pollutant]
    plt.figure(figsize=(8, 6))
    plt.hist(subset["value"], bins=10, edgecolor="black", alpha=0.7)
    plt.title(f"Histogram of {pollutant} concentrations")
    plt.xlabel("Concentration")
    plt.ylabel("Frequency")
    plt.grid(True)
    plt.savefig(os.path.join(output_dir, f"histogram_{pollutant}.png"))
    plt.close()

# Średnie stężenia w czasie
data["date"] = pd.to_datetime(data["date"])
mean_over_time = data.groupby(["date", "pollutant"])["value"].mean().unstack()
mean_over_time.plot(figsize=(10, 6), marker="o")
plt.title("Mean Concentrations Over Time")
plt.xlabel("Date")
plt.ylabel("Mean Concentration")
plt.legend(title="Pollutant")
plt.grid(True)
plt.savefig(os.path.join(output_dir, "mean_concentration_over_time.png"))
plt.close()

print(f"Figures have been saved to the '{output_dir}' directory.")