Understanding the basic geospatial data & queries

[MBee]

Hi, just trying to understand the basic geospatial data & queries. In

Exploring the Data

Find Restaurants with Geospatial Queries — MongoDB Manual

I inspected an entry in the newly-created restaurants collection in
mongosh, with

db.restaurants.findOne()

which returns a document like the following:

{
   location: {
      type: "Point",
      coordinates: [-73.856077, 40.848447]
   },
   name: "Morris Park Bake Shop"
}

However, what I found in the MongoDB Atlas UI, the actual record looks like this:

{"_id":{"$oid":"5eb3d668b31de5d588f4292e"},"address":{"building":"1007","coord":[{"$numberDouble":"-73.856077"},{"$numberDouble":"40.848447"}],"street":"Morris Park Ave","zipcode":"10462"},"borough":"Bronx","cuisine":"Bakery","grades":[{"date":{"$date":{"$numberLong":"1393804800000"}},"grade":"A","score":{"$numberInt":"2"}},{"date":{"$date":{"$numberLong":"1378857600000"}},"grade":"A","score":{"$numberInt":"6"}},{"date":{"$date":{"$numberLong":"1358985600000"}},"grade":"A","score":{"$numberInt":"10"}},{"date":{"$date":{"$numberLong":"1322006400000"}},"grade":"A","score":{"$numberInt":"9"}},{"date":{"$date":{"$numberLong":"1299715200000"}},"grade":"B","score":{"$numberInt":"14"}}],"name":"Morris Park Bake Shop","restaurant_id":"30075445"}

So is the db.restaurants.findOne() a function that is doing its own data massaging?

As I want to understand what data format I can/should use for geospatial data in my DB, so that I can make use of the geospatial queries, yet, what the above query returned differs greatly with what was stored.

Any clarification/explanation please? thx.

[steevej (Steeve Juneau)]

Did you really

@MBee

I inspected an entry in the newly-created restaurants collection

and get simply the following as output

@MBee

{
   location: {
      type: "Point",
      coordinates: [-73.856077, 40.848447]
   },
   name: "Morris Park Bake Shop"
}

or you simply cut-n-paste what is in the documentation?

I think you simply cut-n-paste what is in the documentation because without a projection you will get an _id. I think that in the documentation they only shown the fields important to the discussion.

I think you are really looking at the same document. One is represented as EJSON (the one with $oid) and the other is only part of the whole document to make the documentation easier to read.

So no, findOne() is not doing its own data massaging.

[MBee]

Thanks,

Ah, indeed, there is an _id field. Here is the actual output from my side:

> db.restaurants.findOne()
{
  _id: ObjectId("55cba2476c522cafdb053add"),
  location: { coordinates: [ -73.856077, 40.848447 ], type: 'Point' },
  name: 'Morris Park Bake Shop'
}

What I’m asking is that, in the sample data, it was:

"address":{"building":"1007","coord":

Yet the db.restaurants.findOne() is showing me totally different thing.

So if findOne() is not doing its own data massaging, then where the differences are coming from?

Or I am looking at the wrong collection?

I was looking into the sample_restaurants collection in the MongoDB Atlas UI, where the mongoimport command in the tutorial didn’t specify a database…

[steevej (Steeve Juneau)]

@MBee

Or I am looking at the wrong collection?

Most likely you are.

@MBee

I was looking into the sample_restaurants collection in the MongoDB Atlas UI

From the screenshot you supplied (the part with show dbs), sample_restaurants is a database, not a collection.

What I suspect is that your mongoimport used the test database. In the same Atlas UI, do the following commands and share the output:

use test
show collections
db.restaurants.countDocuments()
use sample_restaurants
show collections
db.restaurants.countDocuments()

Also share the mongoimport command you used.

[MBee]

Ah, indeed:

Problem solved. Thanks.