MONGO get only the name of documents but not the whole documents

问题

query mongo to find only the names : in such a way that it should be fast, ie i don't want to get each document and then get the name from each of them.

I am new to mongo, db.company.find() --> this will give whole document Which is VERY BIG

NOTE ; i have used indexing to the company name & also the unique condition too. soo --> i think it should help me finding the company name fast and easily , BUT i don't know how

this is the collection

collection company :      --> list of companies in the collection    

 {

    {   
       "_id": "5b8ed214b460e7c17c5a33f9",
        "company_location": "USA",
        "company_name": "tesla",         -----> COMPANY NAME: TESLA
        "cars": 
            [
                 {---car 1---} ,
                 {---car 2---} ,
                 {---car n---} 
            ],
    },

    {   
       "_id": "5b8ed214b460e7c17c5a33f9",
        "company_location": "USA",
        "company_name": "gmc",         -----> COMPANY NAME :GMC
        "cars": 
            [
                 {---car 1---} ,
                 {---car 2---} ,
                 {---car n---} 
            ],
    },

    {   
       "_id": "5b8ed214b460e7c17c5a33f9",
        "company_location": "USA",
        "company_name": "bmw",         -----> COMPANY NAME:BMW
        "cars": 
            [
                 {---car 1---} ,
                 {---car 2---} ,
                 {---car n---} 
            ],
    },
    {   
       "_id": "5b8ed214b460e7c17c5a33f9",
        "company_location": "USA",
        "company_name": "audi",         -----> COMPANY NAME: AUDI
        "cars": 
            [
                 {---car 1---} ,
                 {---car 2---} ,
                 {---car n---} 
            ],
    },
}

So I just want the list of company names, and not the whole company collection using db.companies.find(). --> and then finding the names of each of them by traversing it

How can i do this : fast is necessary, data is huge

回答1:

You can use second parameter of .find() method to specify a projection:

db.companies.find({}, { _id: 0, company_name: 1 })

which returns:

{ "company_name" : "gmc" }
{ "company_name" : "tesla" }
...

Or you can use Aggregation Framework to get single document with an array of names:

db.companies.aggregate([{ $group: { _id: null, company_names: { $push: "$company_name" } } }])

which returns:

{ "_id" : null, "company_names" : [ "gmc", "tesla", ... ] }

First one should be the fastest way if you have an index on company_name. In that case your query don't need to scan collection and can use only index to get queried data (covered query).

来源：https://stackoverflow.com/questions/52212165/mongo-get-only-the-name-of-documents-but-not-the-whole-documents

标签

node.js

mongodb

collections