import pandas as pd

sales_df = pd.read_csv("LA_retail_sales.csv")
sales_df.info()  # 750 rows, 8 columns

import glob
import pandas as pd

file_paths = glob.glob("sales_by_day/*.csv")           # grabs all CSVs in folder
sales_df = pd.concat([pd.read_csv(p) for p in file_paths],
                     ignore_index=True)

import pandas as pd

sales_df = pd.DataFrame()
for day in range(1, 6):  # 1..5
    daily = pd.read_csv(f"sales_by_day/sales_2024-09-{day:02d}.csv")
    sales_df = pd.concat([sales_df, daily], ignore_index=True)

# JSON (newline-delimited/“lines” JSON)
sales_json = pd.read_json("LA_retail_sales.json", lines=True)

# Parquet (columnar, cloud-friendly)
sales_parquet = pd.read_parquet("LA_retail_sales.parquet")

import sqlite3, pandas as pd

con = sqlite3.connect("local.db")
stores = pd.read_sql("""
    SELECT store_id, SUM(dollar_sales) AS total_sales
    FROM sales
    GROUP BY store_id
    ORDER BY total_sales DESC
""", con)

import requests, time
import pandas as pd

# unique dates from sales
dates = (sales_df["date"].sort_values().dropna().dt.normalize().unique())

LA_LAT, LA_LON = 34.0522, -118.2437
API_KEY = "YOUR_API_KEY"

weather_rows = []
for dt in dates:
    unix_ts = int(pd.Timestamp(dt).timestamp())
    url = (f"https://api.openweathermap.org/data/3.0/onecall/timemachine"
           f"?lat={LA_LAT}&lon={LA_LON}&dt={unix_ts}&appid={API_KEY}&units=imperial")
    resp = requests.get(url, timeout=15)
    if resp.status_code == 200:
        data = resp.json()["data"][0]   # first record for the day
        weather_rows.append({
            "date": pd.to_datetime(dt),
            "temp_f": data["temp"],
            "humidity": data["humidity"],
            "description": data.get("weather", [{}])[0].get("description")
        })
        time.sleep(1)  # be polite to the API
weather_df = pd.DataFrame(weather_rows)