CODE: Weather Station Data Filtering

import os
os.getcwd()

import tarfile
from pathlib import Path
import pandas as pd

# ===== 1. PATH TO YOUR GSOD YEAR FILE =====
# change this to your actual file, e.g. 2024.tar.gz or 2025.tar.gz
tar_path = Path("data/2025.tar.gz")

# ===== 2. BUILD A STATION TABLE FROM ALL CSVs =====
stations = []

with tarfile.open(tar_path, "r:gz") as tar:
    for member in tar.getmembers():
        if not member.name.endswith(".csv"):
            continue  # skip non-csv files

        f = tar.extractfile(member)
        if f is None:
            continue

        df_head = pd.read_csv(f, nrows=1)

        stations.append({
            "file": member.name,
            "STATION": df_head.get("STATION", [None])[0],
            "NAME": df_head.get("NAME", [None])[0],
            "LATITUDE": df_head.get("LATITUDE", [None])[0],
            "LONGITUDE": df_head.get("LONGITUDE", [None])[0],
        })

stations_df = pd.DataFrame(stations)
print("Total stations in this year:", len(stations_df))

# ===== 3. FIND CANDIDATE NYC STATIONS =====
# (1) by rough lat/lon box around NYC
nyc_box = stations_df[
    (stations_df["LATITUDE"].between(40.5, 41.1)) &
    (stations_df["LONGITUDE"].between(-74.3, -73.6))
]

print("\nStations in NYC bounding box:")
print(nyc_box[["file", "STATION", "NAME", "LATITUDE", "LONGITUDE"]])

# (2) by name search (NEW YORK / CENTRAL PARK)
nyc_name = stations_df[
    stations_df["NAME"].str.contains("NEW YORK|CENTRAL PARK", case=False, na=False)
]

print("\nStations with NEW YORK or CENTRAL PARK in name:")
print(nyc_name[["file", "STATION", "NAME", "LATITUDE", "LONGITUDE"]])

import tarfile
import pandas as pd
from pathlib import Path

# ========== You already have these ==========
# stations_df  → a DataFrame containing columns: file, STATION, NAME, LATITUDE, LONGITUDE
# tar_path     → path to 2024.tar.gz (or 2025.tar.gz)

# Directory to save selected station csv
output_dir = Path("data/nyc_two_stations")
output_dir.mkdir(exist_ok=True)

# Central Park
cp_row = stations_df[
    stations_df["NAME"].str.contains("NY CITY CENTRAL PARK", case=False, na=False)
].iloc[0]

# JFK
jfk_row = stations_df[
    stations_df["NAME"].str.contains("JFK INTERNATIONAL AIRPORT", case=False, na=False)
].iloc[0]

cp_file = cp_row["file"]
jfk_file = jfk_row["file"]

print("Central Park file:", cp_file)
print("JFK file:", jfk_file)

def read_csv_from_tar(tar_path, csv_file_name):
    with tarfile.open(tar_path, "r:gz") as tar:
        f = tar.extractfile(csv_file_name)
        return pd.read_csv(f)

cp_data = read_csv_from_tar(tar_path, cp_file)
jfk_data = read_csv_from_tar(tar_path, jfk_file)

cp_out = output_dir / "NYC_Central_Park.csv"
jfk_out = output_dir / "NYC_JFK_Airport.csv"

cp_data.to_csv(cp_out, index=False)
jfk_data.to_csv(jfk_out, index=False)

print("Saved:", cp_out)
print("Saved:", jfk_out)