Mongodb – Project Array Items as document fields

Modify key dynamically according to the size of the array not seems to be the best efficient query… I suggests to do it in your code, running this simple query you can loop and get Emails[index] and Phones[index].

By the way, using only MongoDB I think you can try this one (maybe it is possible to improve):

The trick here is to do a "loop" using $map to iterate over the Emails or Phones size.

• So the map wil be from 0 to emails size, which is input: {$range: [0, { $size: "$Emails" }]}.
• Every step will be the "index", setting as as: "index".
• And for every step we want to create and object with structure {"k":"EmailX", "v":"[email protected]}. And that’s why we set k value as $concat the "Email" text and index + 1 using $add (i.e. Email1, Email2, … EmailN).
• Then v value is simply the elements at that index. So at this point you have [{"k":"Email1", "v": "[email protected]"}, {"k": "Email2", "v": "[email protected]"}, { ... }]
• And this is done for Emails and Phones and also is into $concatArrays stage to get in a single array all k and v pairs for emails and phones.
• Next stage is to use $arrayToObject to transform the pairs k and v to and object like {"Email1": "[email protected]"}.
• And last part is using $replaceRoot to get calues at top level, merging the Name value and the result value which is all emails and phones.

db.collection.aggregate([
  {
    $project: {
      Name: 1,
      result: {
        "$concatArrays": [
          {
            $map: {
              input: {$range: [0, { $size: "$Emails" }]},
              as: "index",
              in: {
                k: {$concat: ["Email", { $toString: { $add: ["$$index",1]}}]},
                v: {$arrayElemAt: ["$Emails.address","$$index"]}
              }
            }
          },
          {
            $map: {
              input: {$range: [0, {$size: "$Phones"}]},
              as: "index",
              in: {
                k: { $concat: ["Phone", {$toString: { $add: [ "$$index", 1]}}]},
                v: { $arrayElemAt: [ "$Phones.number", "$$index" ] }
              }
            }
          }
        ]
      }
    }
  },
  {
    $project: {
      "_id": 0,
      "Name": 1,
      "result": {
        $map: {
          input: "$result",
          as: "item",
          in: {
            $arrayToObject: [
              [
                {
                  k: "$$item.k",
                  v: "$$item.v"
                }
              ]
            ]
          }
        }
      }
    }
  },
  {
    $replaceRoot: {
      newRoot: {
        $mergeObjects: [
          {
            "Name": "$Name"
          },
          {
            $mergeObjects: "$result"
          }
        ]
      }
    }
  }
])

Example here

P.S. I’ve tried to be as clear as possible explaining the query. Also I think it is a better option do the loop into the code not getting the computed result from DB. But that’s a dev decision so maybe it’s ok for you doing in this way.