Question

Suppose I create a document in a MongoDB collection with the following structure and command:

db.foo.insert(
    {
        "PatientCore" :
        { "Name" : "John Doe", "Age" : 30 },
        "OnArrivalData" :
        {
            "Readings" :
            { "Temperature" : 100.1, "HBP" : 140, "LBP" : 88, "TimeStamp" : new Date() },
            "Attending Doctor" : "Dr. Samuel John"
        },
        "TreatmentStage" :
        [
            {
                "StageName" : "Surgery",
                "Readings" :
                [
                    { "Temperature" : 100.1, "HBP" : 130, "LBP" : 70, "TimeStamp" : new Date() },
                    { "Temperature" : 99.1, "HBP" : 120, "LBP" : 70, "TimeStamp" : new Date() 
                ],
            },
            {
                "StageName" : "ICU",
                "Readings" :
                [
                    { "Temperature" : 99, "HBP" : 135, "LBP" : 72, "TimeStamp" : new Date() },
                    { "Temperature" : 98.6, "HBP" : 141, "LBP" : 80, "TimeStamp" : new Date() }
                ],
            }
        ]
    }
)

Firstly, please note that I have multiple occurrences of “Readings” in the document.

I want to query “Readings” across the document in such a way that I don’t have to qualify it with the reference to the embedded document. How can I achieve this? We can assume that the structure of “Readings” will be consistent.

I know that I can also work towards re-structuring the document, but here I don’t have the liberty to do so.

Answer 1

That is currently not possible as it would require MongoDB to understand field name wildcards. It would have to allow things like {'*.Readings.Temperature': 99} which it does not at time of writing.

Ther feature request is here https://jira.mongodb.org/browse/SERVER-267