Mongo DB Join Collections - mongodb

Question

I am pretty new to Mongo db and coming from T-SQL background, I am finding little hard to understand how joins work in Mongo.
I have a very simple case where i have a "User Table.. err.. Collections" and "User Audit Collections"..
My User Collection looks something like this.
{
"_id": LUUID("d991e92a-766c-054e-9ad8-1c902acc6efc"),
"System": {
"VisitCount": 1
},
"UserData": {
"Uid": "46831",
"UserName": "abc.",
"FirstName": "abv",
"LastName": "test",
"EmailId": "abc#gmail.com",
"Region": "Georgia",
"Postal": "10000",
"Country": "United States",
"Phone": "800-000-1734",
}
}
and a User Audit Table :
{
"_id": LUUID("9561a583-0afe-e844-a090-43ffdab46ed2"),
"UserId": LUUID("914ed252-3fc7-d84c-9731-f382e7cf400b"),
"StartDateTime": ISODate("2016-05-12T04:07:37.299Z"),
"EndDateTime": ISODate("2016-05-12T04:07:42.715Z"),
"SaveDateTime": ISODate("2016-05-12T04:28:23.186Z"),
"Browser": {
"BrowserVersion": "50.0",
"BrowserMajorName": "Chrome",
"BrowserMinorName": "50.0"
},
"Pages": [
{
"DateTime": ISODate("2016-05-12T04:07:37.365Z"),
"Duration": 5416,
"Item": {
"_id": LUUID("f293157a-f22d-fe49-a7b0-f66f412408fe"),
"Language": "en",
"Version": 1
}"Url": {
"Path": "/"
},
"VisitPageIndex": 1
},
{
"DateTime": ISODate("2016-05-12T04:07:42.781Z"),
"Duration": 0,
"Item": {
"Version": 0
},
"SitecoreDevice": {
"_id": LUUID("df7f5dfe-c089-994d-9aa3-b5fbd009c9f3"),
"Name": "Default"
},
"MvTest": {
"ValueAtExposure": 0
},
"Url": {
"Path": "/Sample Page1"
},
"VisitPageIndex": 2
}
]
}
I need a Flat view where each row will hold all the user User information and the pages the user visited.
The Audit information can be grouped by user or repeated per user.. My main idea is to combine the User details with Page visited history.
I am looking for something like a Left outer join equivalent
something like
Select * from usertable, useraudittable
on usertable.id = userAuditTable.UserId
group by userID.

Answer 1

Mongo is a simple object storage database and does not offer a lot of relational operations like joins. Normally you have to do it programmatically doing multiple queries and processing the data using your application code and logic.
In Mongo 3.2 they introduced the lookup operation to the aggregation pipeline and fortunately it kinda does what you are looking for. You can use something like this (using mongo shell javascript syntax as example)
db.user.aggregate([{
$lookup: {
from: "audit",
localField: "_id",
foreignField: "UserId",
as: "VisitedPages"
}
}]);
If you are using the last version of mongo you can play with this approach otherwise you'll need to go with multiple queries on your application.
Take a look at the documentation