How to use $geoNear with aggregate in PyMongo

Question

Imagine that you have a collection with the following structure:

{
    "_id": ObjectId("543f1ec50204444c53ba39a0"), 
    "request_ip": "61.111.36.11", 
    "owner": ObjectId("543f227c0204444c53ba4b28"),
    "loc": [-116.199, 43.6186]
}

Where loc contains the location of a geographical point defined by its latitude and longitude respectively.

Well, what you need is to find the documents in the collection based on the distance calculated between your locations and another given point.

In this case, the aggregate () function is used:

cercanos = collection.aggregate(
[{ 
    "$geoNear": {
        "near": [ 52.15077 , 9.95112 ],
        "distanceField": "dist", 
        "spherical": True,
        "limit":2
    }
}])

When you follow the MongoDB documentation (https://docs.mongodb.com/manual/reference/operator/aggregation/geoNear) everything seems to be in order, however when you run several errors appear:

  Traceback (most recent call last):
  File "E:\Documents\EclipseProjects\Bonus8\src\pru.py", line 154, in <module>
    "spherical": True,
  File "D:\Python\lib\site-packages\pymongo\collection.py", line 1870, in aggregate
    collation=collation)
  File "D:\Python\lib\site-packages\pymongo\collection.py", line 232, in _command
    collation=collation)
  File "D:\Python\lib\site-packages\pymongo\pool.py", line 419, in command
    collation=collation)
  File "D:\Python\lib\site-packages\pymongo\network.py", line 116, in command
    parse_write_concern_error=parse_write_concern_error)
  File "D:\Python\lib\site-packages\pymongo\helpers.py", line 210, in _check_command_response
    raise OperationFailure(msg % errmsg, code, response)
pymongo.errors.OperationFailure: geoNear command failed: { ok: 0.0, errmsg: "no geo indices for geoNear" }

$geoNear Quoting from the manual page: "The collection must have a geospatial index." Which again also has a nice link to the other manual page describing the indexes themselves. Nothing "shell specific" about it. — Neil Lunn
@NeilLunn, you can't use the reference manual examples directly in PyMongo. This is just what I explain. — Mario Rodríguez
It's an aggregation pipeline stage. It has nothing PyMongo specific about it and actually works exactly the same in all languages. Maybe that is what you need to understand. — Neil Lunn

Mario Rodríguez Mario Rodríguez · Accepted Answer · 2017-06-16T07:39:24

What happen:

The problem is that the usage examples provided by the MongoDB reference manual is specific to the exclusive execution within your console (mongo shell). This causes significant changes to be made in the PyMongo environment, such as having to quote the variables.

Possible solution:

Create a Geospatial Index:

from pymongo import GEO2D
collection.create_index([("loc", GEO2D)])

This specifies a 2-dimensional geospatial index and solves the previously occurring error.

Output:

{
    "_id": ObjectId("543f1ec50204444c53ba39a0"), 
    "request_ip": "61.111.36.11", 
    "owner": ObjectId("543f227c0204444c53ba4b28"),
    "loc": [-116.199, 43.6186]
    "dist": 2.18848713
}

How to use $geoNear with aggregate in PyMongo

1 Answers