Access deeply nested sub-documents in MongoDB at various levels

Question

Access deeply nested sub-documents in MongoDB at various levels

I have a complex document structure stored in my mongoDB database and I need to extract specific sub-objects from it.

For example:

{
   "organization": "Tech Inc",
   "CEO": "Alice Johnson",
   "departments": [
      {
         "name": "Engineering"
         "head": "Bob Smith"
         "employees": [
            {
               "name": "Sarah"
               "position": "Software Developer"
            },
            // ... many more employees
         ]
      },
      // ... many more departments
   ]
}

MongoDB recently introduced the ability to retrieve 1-level-deep sub-objects using $elemMatch projection:

var projection = { _id: 0, departments: { $elemMatch: { name: "Engineering" } } };
db.organizations.find({"organization": "Tech Inc"}, projection);
// returns { "departments": [ /* array containing only the matching department */ ]  }

However, attempting to fetch a sub-object at a deeper level (e.g., retrieving a specific employee) using the same method results in an error:

var projection = { _id: 0, "departments.employees": { $elemMatch: { name: "Sarah" } } };
db.organizations.find({"organization": "Tech Inc"}, projection);
// "$err": "Cannot use $elemMatch projection on a nested field (currently unsupported).", "code": 16344

Is there a workaround for retrieving arbitrarily deep sub-objects within a MongoDB document?

I am currently using MongoDB version 2.2.1

javascript mongodb

Answer 1

Answer №1

In a recent query similar to this one, I was able to provide a general answer that may be helpful (refer to Using MongoDB's positional operator $ in a deeply nested document query)

This approach is specifically designed for Mongo 2.6 and above, utilizing the aggregation framework's $redact function.

Below is an example query tailored to retrieve just the student named Bort from a school called Cool School:

db.users.aggregate({

    $match: { schoolName: 'Cool School' }

}, {

    $project: {

        _id: 0,
        'schoolName': 1,
        'rooms.number': 1,
        'rooms.students': 1

    }

}, {

    $redact: {
        $cond: {
            "if": {
                $or: [{
                    $gte: ['$schoolName', '']
                }, {
                    $eq: ['$number', 100]
                }]
            },
            "then": "$$DESCEND",
            "else": {
                $cond: {
                    "if": {
                        $eq: ['$name', 'Bort']
                    },
                    "then": "$$KEEP",
                    "else": "$$PRUNE"
                }
            }
        }
    }

});

The $redact feature allows for sub-queries by selectively matching or pruning sub-documents within the resolved documents.

To gain a deeper understanding of how $redact works, consider reading more about it here. The design pattern outlined here has specific requirements:

The redact condition functions at each level of sub-documents, necessitating unique fields at every level (e.g. avoiding duplicate keys like 'number' for both rooms and students)
$redact operates on data fields rather than array indices, meaning if you require the position of a nested document for operations such as updates (for instance), it must be included and maintained in your documents
Each segment of the $or statement in $redact should align with desired documents at a given level
Thus, each segment of the $or statement must include a match to the unique field at that level. For instance, $eq: ['$number', 100] matches the room numbered 100
If no explicit query is specified at a level, the unique field still needs to be encompassed. An empty string can serve as a placeholder with $gte: ['$uniqueField': '']
The final document level belongs in the second section of the if expression to ensure retention of the entire document

Answer 2

In a recent query similar to this one, I was able to provide a general answer that may be helpful (refer to Using MongoDB's positional operator $ in a deeply nested document query)

This approach is specifically designed for Mongo 2.6 and above, utilizing the aggregation framework's $redact function.

Below is an example query tailored to retrieve just the student named Bort from a school called Cool School:

db.users.aggregate({

    $match: { schoolName: 'Cool School' }

}, {

    $project: {

        _id: 0,
        'schoolName': 1,
        'rooms.number': 1,
        'rooms.students': 1

    }

}, {

    $redact: {
        $cond: {
            "if": {
                $or: [{
                    $gte: ['$schoolName', '']
                }, {
                    $eq: ['$number', 100]
                }]
            },
            "then": "$$DESCEND",
            "else": {
                $cond: {
                    "if": {
                        $eq: ['$name', 'Bort']
                    },
                    "then": "$$KEEP",
                    "else": "$$PRUNE"
                }
            }
        }
    }

});

The $redact feature allows for sub-queries by selectively matching or pruning sub-documents within the resolved documents.

To gain a deeper understanding of how $redact works, consider reading more about it here. The design pattern outlined here has specific requirements:

The redact condition functions at each level of sub-documents, necessitating unique fields at every level (e.g. avoiding duplicate keys like 'number' for both rooms and students)
$redact operates on data fields rather than array indices, meaning if you require the position of a nested document for operations such as updates (for instance), it must be included and maintained in your documents
Each segment of the $or statement in $redact should align with desired documents at a given level
Thus, each segment of the $or statement must include a match to the unique field at that level. For instance, $eq: ['$number', 100] matches the room numbered 100
If no explicit query is specified at a level, the unique field still needs to be encompassed. An empty string can serve as a placeholder with $gte: ['$uniqueField': '']
The final document level belongs in the second section of the if expression to ensure retention of the entire document

Answer 3

Answer №2

Unfortunately, I currently don't have easy access to mongodb 2.2 for testing purposes. However, have you given this a try?

var projection = { _id: 0, rooms: { $elemMatch: { "students.name": "Bort" } } };
db.schools.find({"schoolName": "Cool School"}, projection);

Answer 4

Unfortunately, I currently don't have easy access to mongodb 2.2 for testing purposes. However, have you given this a try?

var projection = { _id: 0, rooms: { $elemMatch: { "students.name": "Bort" } } };
db.schools.find({"schoolName": "Cool School"}, projection);

Access deeply nested sub-documents in MongoDB at various levels

Answer №1

Answer №2

Similar questions

How can we maintain line breaks in Textfields when using React-Admin and Material UI?

Converting JSON data into an array using JavaScript

Tips for creating a blur effect on all options except for the selected 2 out of 4 using jQuery

Can someone explain why I am consistently receiving the value of undefined when accessing the file Object?

JustGage error: Unable to locate element with ID 0 in AngularJS

Ways to verify DOM changes in Vue.js using Mocha testing framework

Executing JavaScript POST Requests Based on User Input Changes

The HTML modal window is experiencing loading issues

Can anyone explain the functionality of passport.initialize() in the context of Node.js and Express

Matching Tables with JavaScript and JSON

What is the best method for updating the name of a specific field in all of my documents using mongojs?

Manipulating and updating intricate data stored in a MongoDB object nested within an array, which is nested within an object, which is then

The SyntaxError message indicates that there was an unexpected non-whitespace character found after the JSON data when parsing it

Locating records containing a subset of sub-documents within a specified search term

Adjust the height of a div element and enable scrolling to fit the

Next.js encountered an issue with the element type, as it expected either a string for built-in components or a class/function for composite components, but received undefined instead

What is the best way to silence console.log outputs during testing?

Tips for retrieving error messages using the Scala/Java MongoDB api

Linking a Checkbox to a Field's Value in AngularJS

Unusual Behavior of Next.js Layouts