zipcode-coordinates-tz

Welcome to the zipcode-coordinates-tz documentation! Here you will find links to the core modules and examples of how to use each.

A Python package that enables converting a US Zip Code into a timezone.

This is done through the querying of the USPS API, then joining it with the GeoLocation data from the US Census, and finally taking the coordinates and using timezonefinder to determine the timezone.

Modules

If there is functionality that is missing or an error in the docs, please open a new issue here.

Install

To install zipcode-coordinates-tz from PyPI, use the following command:

$ pip install zipcode-coordinates-tz

You can also clone the repo and run the following command in the project root to install the source code as editable:

$ pip install -e .

Indices and tables

Examples

The following example will query for all the US Locales, filter by the NJ state. Enrich the Pandas DataFrame with the Longitude and Latitude Coordinates, then add a TZ column for the timezones.

from zipcode_coordinates_tz import census, postal, timezone

df_postal_locales = await postal.get_locales()
df_postal_locales = df_postal_locales.loc[df_postal_locales.State == "NJ"]
df_postal_locales = await census.get_coordinates(df_postal_locales)
df_postal_locales = timezone.fill_timezones(df_postal_locales, fill_missing=True)
print(df_postal_locales)

This example makes the same query as above, but saves it to a JSON file.

python -m zipcode_coordinates_tz save NJ.json --state NJ --timezones --fill