Tair (Redis® OSS-Compatible):Implement a local shopping service using TairGIS

Why TairGIS over MySQL or PostGIS

How it works

Each store on an e-commerce platform defines a service area — the geographic zone within which it accepts orders. A service area can be a specific district, an irregularly shaped area, or a sphere with a given radius around the store.

When a consumer places an order, the system checks whether the consumer's location falls inside any store's service area. TairGIS handles this with two operations:

1. Add the available scopes of stores of a brand to TairGIS.

2. Query available stores that sit near a consumer based on the location of the consumer.

Prerequisites

Before you begin, ensure that you have:

• A Tair (Enterprise Edition) instance with TairGIS enabled

• Python 3.8

• The Tair-py SDK installed: run pip3 install tair

Connect to a Tair instance

Create a connection to your Tair instance. Replace the placeholder values with your actual endpoint and password.

# -*- coding: utf-8 -*-
from typing import Dict
from tair import Tair
from tair import ResponseError


def get_tair() -> Tair:
    """
    Connect to a Tair instance.
    - host: the endpoint of your Tair instance
    - port: the port number (default: 6379)
    - password: the password of the default database account.
      To connect with a custom database account, use the username:password format.
    """
    tair: Tair = Tair(
        host="r-8vb************.redis.zhangbei.rds.aliyuncs.com",
        port=6379,
        db=0,
        password="D******23",
        decode_responses=True
    )
    return tair

Add store service areas

Use GIS.ADD to store the service area of each store as a Well-Known Text (WKT) POLYGON geometry. Group stores by brand ID so you can query all stores for a given brand in one call.

def add_brand_store(brandID: str, mapping: Dict[str, str]) -> bool:
    """
    Add service areas for all stores of a brand using GIS.ADD.
    - brandID: the brand identifier (used as the TairGIS key)
    - mapping: a dict mapping store IDs to their WKT POLYGON geometries
    """
    try:
        tair = get_tair()
        ret = tair.gis_add(brandID, mapping)
        return ret == 1
    except ResponseError as e:
        print(e)
        return False

Query available stores by consumer location

Use gis_contains to find which stores have the consumer's location inside their service area. Pass the consumer's coordinates as a WKT POINT geometry.

def get_service_store(brandID: str, user_location: str):
    """
    Return stores whose service area contains the consumer's location.
    - brandID: the brand identifier
    - user_location: the consumer's location as a WKT POINT, e.g. 'POINT(lon lat)'
    """
    try:
        tair = get_tair()
        return tair.gis_contains(brandID, user_location)
    except:
        return None

(Optional) Query all service areas for a brand

Use GIS.GETALL to retrieve the service areas of all stores for a given brand — useful for debugging or building a store-map view.

def get_brand_all(brandID):
    """
    Return all stores and their service areas for a brand using GIS.GETALL.
    """
    try:
        tair = get_tair()
        return tair.gis_getall(brandID)
    except:
        return None

Example

The following example adds two stores for brand1, retrieves all service areas, and then finds which store serves a consumer at coordinates (120.19707, 30.19539).

if __name__ == "__main__":
    tair = get_tair()

    # Add service areas for two stores of brand1
    add_brand_store(
        "brand1",
        {
            "store_1": "POLYGON ((120.14772 30.19513, 120.15370 30.17838, 120.19385 30.18011, 120.18853 30.20817))",
            "store_2": "POLYGON ((120.18986 30.20852, 120.19651 30.17988, 120.22512 30.17978, 120.21667 30.22292))"
        },
    )

    print("Query all the stores of brand1 and their available scopes:")
    print(get_brand_all('brand1'))

    print("Query the available stores for a consumer who is located at coordinates (120.19707, 30.19539):")
    print(get_service_store('brand1', 'POINT(120.19707 30.19539)'))

Expected output:

Query all the stores of brand1 and their available scopes.
['store_1', 'POLYGON((120.14772 30.19513,120.1537 30.17838,120.19385 30.18011,120.18853 30.20817))', 'store_2', 'POLYGON((120.18986 30.20852,120.19651 30.17988,120.22512 30.17978,120.21667 30.22292))']
Query the available stores for a consumer who is located at coordinates (120.19707, 30.19539).
[1, ['store_2', 'POLYGON((120.18986 30.20852,120.19651 30.17988,120.22512 30.17978,120.21667 30.22292))']]

The available store for the consumer who is located at coordinates (120.19707, 30.19539) is store_2.

What's next

• Implement digital fences by using TairGIS — use TairGIS to monitor whether a tracked entity enters or exits a defined geographic boundary

• TairGIS command reference — full reference for GIS.ADD, GIS.GETALL, GIS.CONTAINS, and other TairGIS commands