Reputation: 43
I have the following data:
{
"_id" : ObjectId("55a8c1ba3996c909184d7a22"),
"uid" : "1db82e8a-2038-4818-b805-76a46ba62639",
"createdate" : ISODate("2015-07-17T08:50:02.892Z"),
"palce" : "aa",
"sex" : 1,
"longdis" : 1,
"location" : [ 106.607312, 29.575281 ]
}
{
"_id" : ObjectId("55a8c1ba3996c909184d7a24"),
"uid" : "1db82e8a-2038-4818-b805-76a46ba62639",
"createdate" : ISODate("2015-07-17T08:50:02.920Z"),
"palce" : "bbb",
"sex" : 1,
"longdis" : 1,
"location" : [ 106.589896, 29.545098 ]
}
{
"_id" : ObjectId("55a8c1ba3996c909184d7a25"),
"uid" : "1db82e8a-2038-4818-b805-76a46ba62639",
"createdate" : ISODate("2015-07-17T08:50:02.922Z"),
"palce" : "ccc",
"sex" : 1,
"longdis" : 1,
"location" : [ 106.590758, 29.566713 ]
}
{
"_id" : ObjectId("55a8c1ba3996c909184d7a26"),
"uid" : "1db82e8a-2038-4818-b805-76a46ba62639",
"createdate" : ISODate("2015-07-17T08:50:02.923Z"),
"palce" : "ddd",
"sex" : 1,
"longdis" : 1,
"location" : [ 106.637039, 29.561436 ]
}
{
"_id" : ObjectId("55a8c1bc3996c909184d7a27"),
"uid" : "1db82e8a-2038-4818-b805-76a46ba62639",
"createdate" : ISODate("2015-07-17T08:50:04.499Z"),
"palce" : "eee",
"sex" : 1,
"longdis" : 1,
"location" : [ 106.539522, 29.57929 ]
}
{
"_id" : ObjectId("55a8d12e78292fa3837ebae4"),
"uid" : "1db82e8a-2038-4818-b805-76a46ba62639",
"createdate" : ISODate("2015-07-17T09:55:58.947Z"),
"palce" : "fff",
"sex" : 1,
"longdis" : 1,
"location" : [ 106.637039, 29.561436 ]
}
I want to first of all, sort by the distance, if the distance is the same, sort by the time.
my command :
db.runCommand( {
geoNear: "paging",
near: [106.606033,29.575897 ],
spherical : true,
maxDistance : 1/6371,
minDistance:0/6371,
distanceMultiplier: 6371,
num:2,
query: {'_id': {'$nin': []}}
})
or
db.paging.find({
'location':{
$nearSphere: [106.606033,29.575897],
$maxDistance:1
}
}).limit(5).skip((2 - 1) * 2).sort({createdate:-1})
How can I sort on both "nearest" and "createddate"?
Upvotes: 4
Views: 6139
Reputation: 50406
The correct query to use here uses the aggregation framework which has the $geoNear
pipeline stage to assist with this. It's also the only place you get to "sort" by multiple keys, as unforntunately the "geospatial" $nearSphere
does not have a "meta" projection for "distance" like $text
has a "score".
Also the geoNear
database command you are using can also not be used with "cursor" .sort()
in that way either.
db.paging.aggregate([
{ "$geoNear": {
"near": [106.606033,29.575897 ],
"spherical": true,
"distanceField": "distance",
"distanceMuliplier": 6371,
"maxDistance": 1/6371
}},
{ "$sort": { "distance": 1, "createdate": -1 } },
{ "$skip": ( 2-1 ) * 2 },
{ "$limit": 5 }
])
That is the equivalent of what you are trying to do.
With the aggregation framework you use the "pipeline operators" instead of "cursor modifiers" to do things like $sort
, $skip
and $limit
. Also these must be in a Logical order, whereas the cursor modifiers generally work it out.
It's a "pipeline", just like "Unix pipe". |
Also, be careful with "maxDistance" and "distanceMuliplier". Since your co-ordinates are in "legacy co-ordinate pairs" and not GeoJSON
format, then the distances are measured in "radians". If you have GeoJSON stored location data then the result is returned in "meters".
Upvotes: 10