import warnings; warnings.filterwarnings("ignore")

# Let us set up our pandas environment
import pandas as pd
import matplotlib.pyplot as plt

# Data setup: Load airline passengers dataset
url = "https://raw.githubusercontent.com/jbrownlee/Datasets/master/airline-passengers.csv"
df = pd.read_csv(url)

# Show the size and first few rows
print("Shape:", df.shape)
df.head()

# Plot passenger counts over time
plt.figure(figsize=(10,4))
plt.plot(df['Month'], df['Passengers'])
plt.title("Airline Passengers Over Time")
plt.xlabel("Month")
plt.ylabel("Number of Passengers")
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

# Make Month the index and parse it as a date
df['Month'] = pd.to_datetime(df['Month'])
df = df.set_index('Month')
df.head()

# Resample to yearly totals
yearly = df.resample('Y').sum()
yearly

# Plot the yearly totals
plt.figure(figsize=(8,3))
plt.bar(yearly.index.year, yearly['Passengers'])
plt.title("Total Airline Passengers per Year")
plt.xlabel("Year")
plt.ylabel("Yearly Total Passengers")
plt.show()

# Resample with mean for yearly averages
yearly_avg = df.resample('Y').mean()
yearly_avg

# Quarterly totals
quarterly = df.resample('Q').sum()
quarterly.head()

# Find the busiest month of all time
max_month = df['Passengers'].idxmax()
max_value = df['Passengers'].max()
print(f"Busiest month: {max_month.strftime('%Y-%m')} with {max_value} passengers")

# Aggregating with multiple functions
agg = df.resample('Y').agg(['sum', 'mean', 'max'])
agg

# Resample and fill missing data
monthly_all = df.resample('M').sum()
monthly_all = monthly_all.reindex(pd.date_range(monthly_all.index.min(), monthly_all.index.max(), freq='M'))
monthly_all = monthly_all.fillna(method='ffill')
monthly_all.head()

# Rolling mean: smooth out ups and downs
df['RollingMean_6'] = df['Passengers'].rolling(window=6).mean()
df[['Passengers','RollingMean_6']].plot(figsize=(10,5))
plt.title('Passenger Trend (6-month Moving Average)')
plt.show()

# Apply a custom aggregation: percent change each year
pct_change = yearly['Passengers'].pct_change() * 100
pct_change = pct_change.round(2)
print(pct_change)

# Mini-project part 1: Find the year with most new passengers
diffs = yearly['Passengers'].diff()
year = diffs.idxmax().year
print(f"Year with most new passengers: {year}")

# Mini-project part 2: Classify years as 'Growth' or 'Drop'
yearly['Change'] = yearly['Passengers'].diff()
yearly['Category'] = yearly['Change'].apply(lambda x: 'Growth' if x>0 else 'Drop')
print(yearly[['Passengers','Category']])

# Troubleshooting: Why did resample fail?
try:
    df_reset = df.reset_index()
    df_reset['FakeDate'] = df_reset['Passengers'].astype(str)
    pd.to_datetime(df_reset['FakeDate']).resample('Y')
except Exception as e:
    print("Error:", e)
    

# Extra: Resample and plot only the summer months
summer = df[df.index.month.isin([6,7,8])]
summer_sum = summer.resample('Y').sum()
summer_sum.plot(kind='bar', figsize=(7,3))
plt.title('Total Passengers each Summer')
plt.show()