# Always suppress warnings for a cleaner notebook
import warnings; warnings.filterwarnings("ignore")
import numpy as np
np.random.seed(42)

# Data setup: Load Uber trip data from New York City (April 2014)
import pandas as pd
url = 'https://raw.githubusercontent.com/fivethirtyeight/uber-tlc-foil-response/master/uber-trip-data/uber-raw-data-apr14.csv'
df = pd.read_csv(url)
print(df.shape)
print(df.head(3))

# What do our Uber data columns mean?
print(df.columns.tolist())

# Clean up column names if needed
df.columns = [col.strip() for col in df.columns]
print(df.columns.tolist())

# Check for missing values in the table
print(df.isnull().sum())

# Change date column to a datetime type
df['Date/Time'] = pd.to_datetime(df['Date/Time'])
print(df['Date/Time'].dtype)
print(df['Date/Time'].head(2))

# Add new columns for hour, day of week, and day of month
df['hour'] = df['Date/Time'].dt.hour
df['weekday'] = df['Date/Time'].dt.dayofweek
df['day'] = df['Date/Time'].dt.day
print(df[['hour', 'weekday', 'day']].head(3))

# Import plotting libraries
import matplotlib.pyplot as plt
import seaborn as sns
sns.set(style="whitegrid")

# Plot the number of pickups by hour
plt.figure(figsize=(10, 5))
sns.countplot(x='hour', data=df, color='skyblue')
plt.title('Number of Uber Pickups by Hour')
plt.xlabel('Hour of Day (0 to 23)')
plt.ylabel('Number of Trips')
plt.show()

# Compare pickups on different weekdays
plt.figure(figsize=(10, 5))
sns.countplot(x='weekday', data=df, color='lightgreen')
plt.title('Uber Pickups by Weekday (0=Monday, 6=Sunday)')
plt.xlabel('Day of Week')
plt.ylabel('Number of Trips')
plt.show()

# Make a simple time series plot: Total Uber trips per day
trips_by_day = df.groupby('day').size()
plt.figure(figsize=(12, 5))
trips_by_day.plot(kind='line', marker='o', color='coral')
plt.title('Number of Uber Trips per Day (April 2014)')
plt.xlabel('Day of Month')
plt.ylabel('Number of Trips')
plt.xticks(range(1, 31, 2))
plt.grid(True)
plt.show()

# Where do pickups happen the most? Plot pickup locations on a map
plt.figure(figsize=(8,8))
plt.scatter(df['Lon'], df['Lat'], s=1, alpha=0.03)
plt.title('Uber Pickup Spots in New York City (April 2014)')
plt.xlabel('Longitude')
plt.ylabel('Latitude')
plt.xlim(-74.1, -73.7)
plt.ylim(40.6, 40.9)
plt.show()

# Quick stats: mean, minimum, and maximum hour of pickups
print("Earliest pickup hour:", df['hour'].min())
print("Latest pickup hour:", df['hour'].max())
print("Mean pickup hour:", df['hour'].mean())

# Let us try a bar plot: Trips per Uber base location
plt.figure(figsize=(8,5))
sns.countplot(x='Base', data=df, palette='Set2')
plt.title('Number of Pickups by Uber Base Code')
plt.xlabel('Base Code')
plt.ylabel('Number of Trips')
plt.show()

# Try filtering: Show only trips from a specific base (input)
base = input("Enter an Uber base code (e.g., B02512): ")
filtered = df[df['Base'] == base]
print(filtered.shape)
print(filtered.head())

# Save your filtered results to a new file
filtered.to_csv('uber_base_filtered.csv', index=False)
print("Filtered data saved to 'uber_base_filtered.csv'")