Lab: Pipeline de Dados com Google Colab

# Google Colab — análise de dados da API
!pip install sqlalchemy psycopg2-binary pandas matplotlib -q

import pandas as pd
import matplotlib.pyplot as plt
from sqlalchemy import create_engine
from google.colab import userdata # secrets do Colab

# Conectar ao banco (DATABASE_URL do Secret do Colab)
engine = create_engine(userdata.get('DATABASE_URL'))

# Extrair dados
df_orders = pd.read_sql("""
  SELECT o.id, o.total, o.status, o.created_at,
         u.name as user_name, u.email
  FROM orders o JOIN users u ON u.id = o.user_id
  WHERE o.created_at > NOW() - INTERVAL '30 days'
""", engine, parse_dates=['created_at'])

# Análise: receita por dia
df_daily = df_orders[df_orders.status == 'paid'].groupby(
    df_orders.created_at.dt.date
)['total'].sum()

# Visualização
plt.figure(figsize=(12, 4))
df_daily.plot(kind='bar', color='#00ff88')
plt.title('Receita diária — últimos 30 dias')
plt.ylabel('R$')
plt.tight_layout()
plt.savefig('receita.png', dpi=150)
plt.show()