how to query embedded document using mongodb

Question

Need help constructing this mongo query. So far I can query on the first level, but unable to do so at the next embedded level ("labels" > 2")

For example, the document structure looks like this:

> db.versions_20170420.findOne();
{
    "_id" : ObjectId("54bf146b77ac503bbf0f0130"),
    "account" : "foo",
    "labels" : {
        "1" : {
            "name" : "one",
            "color" : "color1"
        },
        "2" : {
            "name" : "two",
            "color" : "color2"
        },
        "3" : {
            "name" : "three",
            "color" : "color3"
        }
    },
    "profile" : "bar",
    "version" : NumberLong("201412192106")

This query I can filter at the first level (account, profile).

db.profile_versions_20170420.find({"account":"foo", "profile": "bar"}).pretty()

However, given this structure, I'm looking for documents where the "label" > "2". It doesn't look like "2" is a number, but a string. Is there a way to construct the mongo query to do that? Do I need to do some conversion?

cn0047 · Accepted Answer · 2017-04-25 05:25:08Z

If I correctly understand you and your data structure, "label" > "2" means that object labels must contain property labels.3, and it is easy to check with next code:

db.profile_versions_20170420.find(
    {"account": "foo", "profile": "bar", "labels.3": {$exists: true}}
).pretty();

But it doesn't mean that your object contains at least 3 properties, because it is not $size function which calculates count of elements in array, and we cannot use $size because labels is object not array. Hence in our case, we only know that labels have property 3 even it is the only one property which labels contains.

You can improve find criteria:

db.profile_versions_20170420.find({
    "account": "foo",
    "profile": "bar",
    "labels.1": {$exists: true},
    "labels.2": {$exists: true},
    "labels.3": {$exists: true}
}).pretty();

and ensure that labes contains elements 1, 2, 3, but in this case, you have to care about object structure on application level during insert/update/delete data in document.

As another option, you can update your db and add extra field labelsConut and after that you will be able to run query like this:

db.profile_versions_20170420.find(
    {"account": "foo", "profile": "bar", "labelsConut": {$gt: 2}}
).pretty();

btw, it will work faster...

Collectives™ on Stack Overflow

how to query embedded document using mongodb

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related