Search in Firebase based on distance

问题

I'm migrating an application that was using Mysql to firebase, and I managed to migrate it almost all, just missing the part of search, I did not quite understand how it works. I'm trying to do this query below in firebase.

SELECT *, (6371 * acos( cos(radians(?)) * cos(radians(latitude)) * cos(radians(?) - radians(longitude)) + sin(radians(?)) * sin(radians(latitude)) )) AS distance FROM usuario ORDER BY distance

What I'm trying to do is pass the latitude and longitude of my User to firebase and he returns me all users ordering the nearest to the farthest.

Note: This is not the same query that I use in my application, that I just find on google and do not know if this work, but you can get an idea of what I'm trying to do. Note ²: English is not my native language :P

回答1:

The example SQL query you provided uses a capability of SQL that does not exist in Firebase: the ability to calculate an expression using the column values for each row in the table and then use the value of that expression to filter and sort the query result.

I don't see a way to use Firebase to perform the type of query you want.

You might want to take look at the Geofire library. I have not used it and its capabilities seem to be related to proximity filtering, not the sort-by-distance feature you need, but maybe you can adjust your requirements to make use of its features.

回答2:

I concur with Qbix and would normally post this as a comment, but I want to make sure the Geohash comment doesn't get lost in the small-text shuffle.

GeoFire uses Geohash codes to create its keys and allow range-matching. For basic apps this may be fine, but the second you go beyond the US this creates a lot of trouble because it doesn't work well around the equator and Prime Meridian (UK). See the Wikipedia Page on Geohash for details, specifically: Edge case locations close to each other but on opposite sides of the 180 degree meridian will result in Geohash codes with no common prefix (different longitudes for near physical locations)..

Firebase is an amazing product, but not a one-size-fits-all tool. If you need good Geo-based search/matching, use a tool like ElasticSearch, MySQL, Algolia, etc. that support it directly. Reducing the number of components in cases like this doesn't decrease complexity, it increases it.

回答3:

You can use Geofirestore library https://github.com/geofirestore/geofirestore-js

here is the example used:

import * as firebase from 'firebase/app';
import 'firebase/firestore';
import { GeoCollectionReference, GeoFirestore, GeoQuery, GeoQuerySnapshot } from 'geofirestore';

// Initialize the Firebase SDK
firebase.initializeApp({
  // ...
});

// Create a Firestore reference
const firestore = firebase.firestore();

// Create a GeoFirestore reference
const geofirestore: GeoFirestore = new GeoFirestore(firestore);

// Create a GeoCollection reference
const geocollection: GeoCollectionReference = geofirestore.collection('restaurants');

// Create a GeoQuery based on a location
const query: GeoQuery = geocollection.near({ center: new firebase.firestore.GeoPoint(40.7589, -73.9851), radius: 1000 });

// Get query (as Promise)
query.get().then((value: GeoQuerySnapshot) => {
  console.log(value.docs); // All docs returned by GeoQuery
});

where {1000} in the query is the radius of location search

来源：https://stackoverflow.com/questions/37759727/search-in-firebase-based-on-distance