Convert nested document collection to model tree structure with parent references

Question

I need to convert a collection with nested documents to a model tree structure with parent references. So this is how my structure looks like:

{
    "_id" : "sdaGREsfRdfGdFdwG",
    "docTitle" : "Document 1",
    "group" : [
        {
            "id" : "cdPhkTpMXi8z6TqMT"
            "title" : "title 1",
            "data" : [
                {
                    "id" : "Nkspf5kKfPo3axeJA",
                    "some" : "data",
                    "other" : "things",
                    "and" : "so on",    
                },
                {
                    "id" : "Ca8ncFgq83RoeD8he",
                    "some" : "data",
                    "other" : "things",
                    "and" : "so on",    
                },
                {
                    "id" : "vyjgkuNXRN9KkCd5o",
                    "some" : "data",
                    "other" : "things",
                    "and" : "so on",    
                }
            ],

        },
        {
            "id" : "TuibXPe5qMvqdMW6q"
            "title" : "title 2",
            "data" : [
                {
                    "id" : "f5L5zsSNRSQKWoAXL",
                    "some" : "data",
                    "other" : "things",
                    "and" : "so on",    
                }
            ],

        }
    ]
}

As you can see, there is a group array, which has also a data array. Now I need a main document like this:

{
    "_id" : "sdaGREsfRdfGdFdwG",
    "docTitle" : "Document 1"
}

And the next child elements, which would be:

{
    "id" : "cdPhkTpMXi8z6TqMT"
    "title" : "title 1",
    "type" : "group",
    "parent" : "sdaGREsfRdfGdFdwG"
},
{
    "id" : "TuibXPe5qMvqdMW6q"
    "title" : "title 2",
    "type" : "group",
    "parent" : "sdaGREsfRdfGdFdwG"
}

At least the data elements, which are children of the group-elements:

{
    "id" : "Nkspf5kKfPo3axeJA",
    "some" : "data",
    "other" : "things",
    "and" : "so on",
    "type" : "element",
    "parent" : "cdPhkTpMXi8z6TqMT"
},
{
    "id" : "Ca8ncFgq83RoeD8he",
    "some" : "data",
    "other" : "things",
    "and" : "so on",
    "type" : "element",
    "parent" : "cdPhkTpMXi8z6TqMT"  
},
{
    "id" : "vyjgkuNXRN9KkCd5o",
    "some" : "data",
    "other" : "things",
    "and" : "so on",
    "type" : "element",
    "parent" : "cdPhkTpMXi8z6TqMT"  
},
{
    "id" : "f5L5zsSNRSQKWoAXL",
    "some" : "data",
    "other" : "things",
    "and" : "so on",
    "type" : "element",
    "parent" : "TuibXPe5qMvqdMW6q"      
}

So my attempt would be something like this:

Collection.find({}).forEach(function(doc) {     // get all documents
    var docID = doc._id;

    doc.group.forEach(function(group) {         // get all groups
        var parent = group.id;
        group.type = "group";
        group.parent = docID;

        group.data.forEach(function(element) {  // get all elements
            element.type ="element";
            element.parent = parent;
        });
    });
});

1) I don't know how to do that in a better way

2) And I don't know how to save the result back to the collection

Answer 1

Fot the groups:

db.myCol.aggregate([{$unwind:"$group"}, 
    {$project:{_id:"$group.id", some:"$group.some", 
               parent:"$_id", type:{$literal:"group"}}}])

For the Elements:

db.myCol.aggregate([{$unwind:"$group"}, 
                    {$unwind:"$group.data"}, 
                    {$project:{_id:"$group.data.id", some:"$group.data.some", 
                               parent:"$group.id", type:{$literal:"element"}}}])

And to your second question: You can add an $out step to the aggregation pipeline at the end. For example: {$out:"outputCol"}