Result of Mongodb aggregation operation

Question

Result of Mongodb aggregation operation

I've created a Property controller :

//Retrieve Properties By Type
const getPropertiesByType = async (req: Request, res: Response) => {
    const { cities, type } = req.query;
    const citiesArr = typeof cities === 'string' ? cities.split(',') : [];
    try {
        const properties = await Promise.all(
            citiesArr?.map((item) => {
                return PropertyModel.aggregate([
                    {
                        $match: { city: item, propertyType: type },
                    },
                    {
                        $project: {
                            _id: 0,
                            city: 1,
                            country: 1,
                            cityPhoto: 1,
                        },
                    },
                    {
                        $group: {
                            _id: '$city',
                            country: { $first: '$country' },
                            totalProperties: { $sum: 1 },
                            cityPhoto: { $first: '$cityPhoto' },
                        },
                    },
                ]);
            })
        );
        res.status(201).json(properties.flat());
    } catch (error) {
        console.log(error);
        res.status(400).json(error);
    }
}

When using Postman, the response I'm receiving looks like this :

[
    {
        "_id": "Davenport",
        "country": "United States",
        "totalProperties": 1,
        "cityPhoto": "https://"
    },
    {
        "_id": "Destin",
        "country": "United States",
        "totalProperties": 1,
        "cityPhoto": "https://"
    }
]

The issue I'm facing is that all objects are nested within arrays, but I would prefer to have them in a single main array. Is there a way to achieve this without using the aggregation pipeline?

typescript mongodb express mongoose

Answer 1

Answer №1

To utilize the array.flatMap() method, simply apply it after receiving the aggregation result:

const property = [
    [
        {
            "_id": "Davenport",
            "country": "United States",
            "totalProperties": 1,
            "cityPhoto": "https://"
        }
    ],
    [
        {
            "_id": "Destin",
            "country": "United States",
            "totalProperties": 1,
            "cityPhoto": "https://"
        }
    ],
] ; // your aggregation output

const response = property.flatMap(x => x);
console.log(response);

The reason you can't directly perform this operation in the aggregation pipeline is due to using Promise.all, which returns an array. Additionally, each .aggregate() call also returns an array, resulting in a nested structure of arrays.

Answer 2

To utilize the array.flatMap() method, simply apply it after receiving the aggregation result:

const property = [
    [
        {
            "_id": "Davenport",
            "country": "United States",
            "totalProperties": 1,
            "cityPhoto": "https://"
        }
    ],
    [
        {
            "_id": "Destin",
            "country": "United States",
            "totalProperties": 1,
            "cityPhoto": "https://"
        }
    ],
] ; // your aggregation output

const response = property.flatMap(x => x);
console.log(response);

The reason you can't directly perform this operation in the aggregation pipeline is due to using Promise.all, which returns an array. Additionally, each .aggregate() call also returns an array, resulting in a nested structure of arrays.

Answer 3

Answer №2

It seems that @mickl has given a solution on how to structure the response using flatMap() after the database completes its processing, which appears to be a valid approach. However, there is an alternative method I would like to explore in response to a question raised in the comments.

From what I understand in the code, you are examining the cities query parameter and expecting a string with values separated by commas, such as "A,B,C". After splitting this string into an array named citiesArr, you proceed to send individual aggregate() requests to the database for each entry in the array (using the map function where each city becomes an item).

I believe it is possible (and advisable) to consolidate these requests into a single call to the database. In terms of the aggregation process itself, it would involve altering

                    {
                        $match: { city: item, propertyType: type },
                    }

To something like

                    {
                        $match: { city: { $in: citiesArr }, propertyType: type },
                    }

If you also modify the code structure surrounding this database query (eliminating the need for .all() and .map()), then the database's direct response should align more closely with your desired outcome.

Two other points to consider:

The $project stage in the middle may not be necessary. The database can determine the appropriate stages for the $group operation and retrieve them automatically. Removing $project could streamline the process.
Given this approach, I recommend verifying that citiesArr contains elements before initiating the aggregation. If the array is empty, there is no need for a database query since no results will be returned.

Answer 4

It seems that @mickl has given a solution on how to structure the response using flatMap() after the database completes its processing, which appears to be a valid approach. However, there is an alternative method I would like to explore in response to a question raised in the comments.

From what I understand in the code, you are examining the cities query parameter and expecting a string with values separated by commas, such as "A,B,C". After splitting this string into an array named citiesArr, you proceed to send individual aggregate() requests to the database for each entry in the array (using the map function where each city becomes an item).

I believe it is possible (and advisable) to consolidate these requests into a single call to the database. In terms of the aggregation process itself, it would involve altering

                    {
                        $match: { city: item, propertyType: type },
                    }

To something like

                    {
                        $match: { city: { $in: citiesArr }, propertyType: type },
                    }

If you also modify the code structure surrounding this database query (eliminating the need for .all() and .map()), then the database's direct response should align more closely with your desired outcome.

Two other points to consider:

The $project stage in the middle may not be necessary. The database can determine the appropriate stages for the $group operation and retrieve them automatically. Removing $project could streamline the process.
Given this approach, I recommend verifying that citiesArr contains elements before initiating the aggregation. If the array is empty, there is no need for a database query since no results will be returned.

Result of Mongodb aggregation operation

Answer №1

Answer №2

Similar questions

Create a dynamic Prisma data call by using the syntax: this.prisma['dataType'].count()

Increasing several array elements within a MongoDB object

Restrict the number of items in each list to just one

Error Detected: An unhandled error occurred, triggering a promise rejection: TypeError: The super expression must be either null or a function in Angular version 6

Learn how to retrieve specific parts of a subdocument in MongoDB using Spring Data

Observe the task while returning - Firebase Functions

Using React Query's useMutation hook may result in an error message of "No overload matches this call"

What is the best way to convert this into a distinct function using typescript?

Understanding the infer keyword in Typescript when working with generic type constraints

Storing the Outcome of a mongodb Search in a Variable

Is there a way to add a new Date to Spring Data MongoDB easily?

Having trouble with Axios PUT request not sending complete data to the server using JavaScript

Issue with Mongoose not triggering callback

Mongo models failing to react

When attempting to utilize the Next.js Head component in a location other than _document, the page experiences

Set the timezone of a Javascript Date to be zero

Avoiding the pitfalls of hierarchical dependency injection in Angular 6

Tips on querying a 3-level nested object in MongoDB

Angular 6 - Ensuring all child components are instances of the same component

What is the process for transferring a JWT token between servers to authenticate APIs operating on separate ports in a Node.js and Express environment?