Synopsis

This project converts JSON files into MongoDB queries.

It's fairly simple to use.

1. Describe the query in JSON format
2. Parse the JSON file into a JS object
3. Feed it to the mongoQueryBuilder
4. Get back the query
5. Use Mongoose on Mongo Driver: Model.find(query)

Installation

npm install mongoquerybuilder

https://www.npmjs.com/package/mongoquerybuilder

API

Mongo queries: The simplest way to describe MongoQueries, would be

       LEFT           RIGHT
{ documentProperty: condition }

Our JSON file consists of conditions and expressions. Conditions define simple Mongo Queries as described above, and expressions either use these conditions or combine them using OR - AND operators.

The syntax described below should be used at all times (builder includes a JOI validator)

It's important to understand that the condition labels can be used either to target a condition in 'use' expression, or to target a condition in the left/right parts of the expressions.

Conditions

{
    prop: 'country',  // the document's property that we wish to evaluate
    op: 'equals',     // the operation that will construct the query
    value: 'UK'       // the desired value (may be an array in case of the "inside" operatio
    isDate: false     // a flag that shows whether the value contains a date string
}

The allowed operations are:

equals is the simplest Mongo query to check equality

{ prop: value }

different uses the $ne operand to check inequality

{ prop: { $ne: value } }

is_one_of uses the $in operand to check inside an array

{ prop: { $in: value } }

is_not_one_of uses the $nin operand to check inside an array

{ prop: { $in: value } }

less_than uses the $lt operand to compare values

{ prop: { $lt: value } }

greater_than uses the $gt operand to compare values

{ prop: { $gt: value } }

greater_equal uses the $gt operand to compare values

{ prop: { $gte: value } }

less_equal uses the $gt operand to compare values

{ prop: { $lte: value } }

exists uses the $exists operand to check if that property exists

{ prop: { $exists: value } }

pattern uses the $regex operand to search via a pattern

{ prop: { $regex: value } }

Expressions

{
    op: 'and',                      // indicates that will be adding an AND query
    left: 'COUNTRY_IS_UK'           // targets the left part of the expression (in this case a simple condition)
    right: {                        // targets the right part of the expression (in this case, another expression)
        op: 'or',                   // indicates that will be adding an OR query
        left: 'USER_IS_OLD',   // simple condition
        right: 'USER_IS_RICH'  // simple condition
    }
}

OR

{
    op: 'use',                     // indicates that will be adding a simple condition in the expression
    condition: 'COUNTRY_IS_UK'     // targets the condition that we need to evaluate
}

OR some combination of the above.

The allowed operations are:

or uses the $or operand to combine left and right expressions

{ $or: [left, right]

and uses the $and operand to combine left and right expressions

{ $and: [left, right]

not uses the $not operand to negate a condition or expression

{ $not: condition / expression

nor uses the $nor operand to combine left and right expressions

{ $nor: [left, right]

match_element uses the $elemMatch operand to match an array element

{ $elemMatch: condition

use is the operation that includes a simple condition to the query

Code Example

Start of by requiring the index file in the root folder:

const queryBuilder = require('mongoquerybuilder')

Then define your expressions and conditions:

// A series of expressions that will build the query //
var expressions = {
    FIND_SMARTPHONE_USERS_IN_USA: {        // LABEL
        op: 'and',                         // OPERATION
        left: 'COUNTRY_IS_USA',            // LEFT PART OF THE EXPRESSION
        right: {                           // RIGHT PART OF THE EXPRESSION
            op: 'or',
            left: 'MODEL_IS_APPLE',
            right: 'MODEL_IS_SAMSUNG'
        }
    }
}
// A series of facts that you would like to evaluate //
var conditions = {
    COUNTRY_IS_USA: {                     // LABEL
        prop: 'build.country',            // DOCUMENT PROPERTY
        op: 'equals',                     // OPERATION
        value: 'USA'                      // VALUE
    },
    MODEL_IS_APPLE: {
        prop: 'build.model',
        op: 'equals',
        value: 'APPLE'
    },
    MODEL_IS_SAMSUNG: {
        prop: 'build.model',
        op: 'equals',
        value: 'SAMSUNG'
    }
}

Follow up by calling the builder (async method)

return queryBuilder(expressions, conditions).then((query) => {
    // Call your database by using 'query'
})

The query produces from the builder, has the following format:

{
    "$and": [
        {
            "build.country": "USA"
        },
        {
            "$or": [
                {
                    "build.model": "APPLE"
                },
                {
                    "build.model": "SAMSUNG"
                }
            ]
        }
    ]
}

Motivation

This project was build in order to enable users not familiar with MongoDB, to construct complex queries, by using a simple/readable JSON format.

Tests

There's a mocha unit test in unit/mongoQueryBuilder_spec.js Works also as an example!