can aggregation give me a document with a count of sub-documents, based on criteria of sub-doc?

[Derick Bailey]

I have a complex data structure where a "schedule" contains "jobs". Right now, I'm doing a lot of "forEach" loops in my code to find all of the jobs that are in a specific state (have a "status" field of whatever value). This works fine, but is tremendously slow, with hundreds of jobs and dozens of schedules listed in a search result screen.

Is it possible to use the mongodb / mongoose aggregation pipeline to get the results that I want?

for example, I want to go from a structure like this:

{

  _id: "(object id)",

  name: "something",

  jobs: [

    {

      _id: "(object id)",

      name: "j1",

      status: 0,

      statusName: "unresolved"

    },

    {

      _id: "(object id)",

      name: "j2",

      status: 0,

      statusName: "unresolved"

    },

    {

      _id: "(object id)",

      name: "j3",

      status: 1,

      statusName: "resolved"

    }

  ]

}

and have this return a data structure that looks something like this (thought the actual resulting structure can vary at this point)

{

  _id: "(object id")

  name: "something",

  jobs: {

    unresolved: 2,

    resolved: 1

  }

}

is this possible? I'm not super picky about what the final data structure looks like... i just want to have mongodb do the work of aggregating and counting the jobs by status, instead of me doing it in code.

thanks,

  - Derick