I want to apply a mergeObjects and map of an Array of an Array:

I am able to complete the mapping and merging when I am one level of Array in but if the Array has another array, I am unsure how to apply another map/merge object.

{
_id: "123",
date: "1900-01-01T11:00:00.0000000",
name: "joe",
birthdayservice: [
 {
    "date": "1999-01-01",
    "team" : [ 
       {
          "requestedDate": "1999-01-01"
       },
       {
          "requestedDate": "1999-05-01"
       }
  },
  {
    "date": "1998-01-01",
    "team" : [ 
       {
          "requestedDate": "1999-01-01"
       },
       {
          "requestedDate": "1999-05-01"
       }
  }
]
}

I am able to map through the first array in birthdayservice to converte date to an ISO_Date/todate such as:

{
    "$addFields": {
    "birthdayservice": {
        $map: {
            input: "$birthdayservice", //this is an array
            in: {
            "$mergeObjects": [
                "$$this",
                    {
                        "date": {
                        "$toDate": "$$this.date"
                        }
                    }
                ]}
            }
        }
    }
 
}

but when I get into more of a nested value, I get

$mergeObjects requires object inputs, but input is an array

{
    "$addFields": {
    "birthdayservice": {
        $map: {
            input: "$birthdayservice", //this is an array
            in: {
            "$mergeObjects": [
                "$$this",
                    {
                        "date": {
                        "$toDate": "$$this.date"
                        }
                    }
                ]}
            }
        }
    }
 
},
{
    "$addFields": {
    "birthdayservice": {
        $map: {
            input: "$birthdayservice.team", //also an array
            in: {
            "$mergeObjects": [
                "$$this",
                    {
                        "requestedDate": {
                        "$toDate": "$$this.requestedDate"
                        }
                    }
                ]}
            }
        }
    }
}

My end goal is to relace the date string fields to a date convert

birthdayservice.date birthdayservice.team.requestedDate

Both to ISO_Dates as below:

{
_id: "123",
date: "1900-01-01T11:00:00.0000000",
name: "joe",
birthdayservice: [
 {
    "date": ISO_Date("1999-01-01T11:00:00.0000000Z"),
    "team" : [ 
       {
          "requestedDate": ISO_Date("1999-01-01T11:00:00.0000000Z")
       },
       {
          "requestedDate": ISO_Date("1999-05-01T11:00:00.0000000Z")
       }
  },
  {
    "date": "1998-01-01",
    "team" : [ 
       {
          "requestedDate": ISO_Date("1999-01-01T11:00:00.0000000Z")
       },
       {
          "requestedDate": ISO_Date("1999-05-01T11:00:00.0000000Z")
       }
  }
]
}

CodePudding user response：

Query

• this converts the all date strings to dates inside birthdayservice and teams array
• map birthdayservice
• convert the "date"
• nested map to convert the dates in the team
• the way the field is changed value is by mergeObjects (in mongodb5 we have also setField to do this)

Test code here

aggregate(
[{"$set":
  {"birthdayservice":
   {"$map":
    {"input":"$birthdayservice",
     "in":
     {"$mergeObjects":
      ["$$bs",
       {"date":{"$toDate":"$$bs.date"},
        "team":
        {"$map":
         {"input":"$$bs.team",
          "in":
          {"$mergeObjects":
           ["$$t", {"requestedDate":
                    {"$toDate":"$$t.requestedDate"}}]},
          "as":"t"}}}]},
     "as":"bs"}}}}])