MongoDB Overview

it is a cross-platform
open source
document-oriented database that provides
- high performance
- high availability
- horizontal scaling
- rich query lanaguage
- multiple storage engines
  - MMAPv1, In-Memory, WiredTiger
Document is a set of key-value pairs similar to JSON objects
Collection is a group of documents
Databases hold collections of documents
Advantages of documents
- Native data types
- No expensive joins
- Dynamic schema

Getting started guide to setting up mongodb

go to root folder in app

mkdir data
mkdir data/db
run: mongod --dbpath /myapp/data/db

leave it running when working with database

for debugging

open a new terminal and open a mongo shell
run: mongo
or run : mongo --host localhost:27017
- 27017 is the default port mongodb uses

for graphical interface

open MongoDB Compass Community
ensure server is already run with mongod (refer to the above)
no settings is required, just click the connect button

Mongo shell basics

mongo uses the language called BSON (Binary JavaScript Object Notation)

help

display a list of basic features

show dbs

display all databases that was created

use databaseName

e.g.: use demo
all "db" called will be refered to demo database
use the database
database not created will be added into collections
but will not be display in "show dbs" until some data is added

CRUD

CREATE

Add 1 Collection

db.dataName.insertOne({key: "value"})

e.g.: db.dogs.insertOne({name: "Rusty", breed: "Mutt"})
dataName is a collection
add data into database
db is the database declared with use databaseName

Add multiple Collections

db.dataName.insertMany([{key1: "value1"}, {key2: "value2"}])

e.g.: db.dogs.insertMany([{name: "Rusty", breed: "Mutt"}, {name: "Bob", breed: "dog"}])
add data into database

Add 1 or multiple collections

db.dataName.insert({key: "value"})

e.g.: db.dogs.insert({name: "Rusty", breed: "Mutt"})
db.dataName.insert([{key1: "value1"}, {key2: "value2"}])
e.g.: db.dogs.insert([{name: "Rusty", breed: "Mutt"}, {name: "Bob", breed: "dog"}])

db.createCollection("dataName")

e.g.: db.createCollection("dogs")
similar to insert but does not require data

show collections

display data added

READ

db.dataName.find()
db.dataName.find({})

find all data keys and values under dataName

Dos and Don'ts

√ db.dataName.find({ _id: ObjectId("fu438h9784hgf847") })  // need to be stored as a hash
x db.dataName.find({ _id: "fu438h9784hgf847 })  // does not work
√ db.dataName.find({ lastLogin: ISODate("2018-10-10T00:00:00") })  // need to be written this way to work for dates

db.dataName.find({key: "value"})

e.g.: db.dogs.find({name: "Rusty"})
find a specific data key and value under dataName

db.dataName.find({key1: "value1", key2: "value2"})

retrieve data that has all of the searched values

db.dataName.find({$ord: [{key1, "value1"}, {key2: "value2"}]})

retrieve data that has either "value1" or "value2"

db.dataName.find({key: {$gt: n}})

display values in comparison to n
gt = greater than n
lt = less than n
eq = equal to n
gte = greate than or equal to n
lte = less than or equal to n

db.dataName.find({key: {$in: ["value1", "value2"]}})

display all data(s) that consist of the value in the array["value1", "value2"]
nin == not in (inverse of in)

db.dataName.find({key: {$exists: n}})

if n == true, show all data that has the key field
if n == false, show all data that does not have the key field

db.dataName.find({}, {_id: n})

if n == 0, display everything except id
if n == 1, display only id

db.dataName.find().limit(n)

e.g.: db.dogs.find().limit(2)
limit search results by n

db.dataName.find().sort({key: n})

if n == 1, sort by ascending order
if n == -1, sort by descending order

db.dataName.find().sort({key1: n1, key2: n2})

sort by multiple fields
sort by key1 then by key2

UPDATE

db.dataName.updateOne({key: "value"}, {changeKey: "changeValue"})
db.dataName.replaceOne({key: "value"}, {changeKey: "changeValue"})

e.g.: db.dogs.updateOne({name: "Rusty"}, {breed: "Labradoodle"})
key and value required to find the position to change
this will rewrite everything to just {breed: "Labradoodle"} without the name object

db.dataName.updateOne({key, "value"}, {$set: {changeKey: "changeValue", addKey: addValue}})
db.dataName.update({key, "value"}, {$set: {changeKey: "changeValue", addKey: addValue}})

e.g.: db.dogs.updateOne({name: "Rusty"}, {$set: {breed: "Labradoodle", isCute: true}})
use $set: {} to ensure data that aren't called will not be deleted
this can be used to update and add new key value pair
this will only update ONE data object in collection
- need to run multiple times to update all

db.dataName.updateMany({key, "value"}, {$set: {changeKey: "changeValue", addKey: addValue}})

this will update ALL data objects in collection

db.dataName.update({key, "value"}, {$set: {changeKey: "changeValue", addKey: addValue}, {multi: true}})

need to add {multi: true} to enable multiple updates

REMOVE (Delete value(s))

db.dataName.remove({key: "value"}, true)

justOne parameter need to set as true to delete only 1 value
db.dataName.deleteOne({key: "value"})
delete 1 value
db.dataName.remove({key: "value"}) db.dataName.deleteMany({key: "value"})
e.g.: db.dogs.remove({breed: "Labradoodle"})
this will remove all dogs that has the same breed value

db.dataName.remove({key: "value"}).limit(n)

e.g.: db.dogs.remove(breed: "Labradoodle").limit(1)
limit n number of data to remove

Delete collection

db.dataName.drop()

e.g.: db.dogs.drop()

Delete database

db.dropDatabase()

thi will delete the currently used database

BULKWRITE

db.dataName.bulkWrite(
  [
    {insertOne:
      {"document": {key1: "value1", key2: "value2"}}
    },
    {updateOne:
      {
        filter: {key: "value"},
        update: {$set: {changekey: "changeValue"}}
      }
    },
    {deleteOne:
      {filter: {key: "value"}}
    }
  ])

enables multiple requests in ONE request

SEARCH

db.dataName.createIndex({key1: "text", key2: "text"})

provide text indexes to enable text search queries on string content

db.dataName.find({$text: {$search: "searchText1 searchText2"}})

$text query operator is required
$search operator required to start searching
key in the search contents
add - in front of searchText to exclude from search e.g.: -searchText1

db.dataName.find({$text: {$search: "\"search phrase\""}})

use \" to enable search by phrase

Get search scores

db.dataName.find({$text: {$search: "searchText"}, {score: {$meta: "textScore"}}})

need to add 1 score object to enable search rankings
search rankings enable sorting if required
$meta projection operator returns for each matching document the metadata

AGGREGATION

Count

db.dataName.count({key: "value"})

this will count the total number of values that is in the data collection

List all values

db.dataName.distinct("key")

display a list of values under key

Find the sum of values under key for each id key

db.dataName.aggregate([
  {$match: {key: "value"}},
  {$group: {_id: "$idKey1", total: {$sum: "$idKey2"}}}
])

$match operator is used to filter data in collection
- use {} to search all
$group is used to group similar idKey1 together
- $ is added in front of idKey1 to group them as a single unit
- idKey1 refers to the key in the data to group similar values together
$sum gives the total sum of all the values given from idKey2
- idKey2 refers to the key in the data to get all the different values

db.dataName.aggregate([
  {$match: {key: "value"}},
  {$group: {_id: "$idKey1", total: {$sum: "$idKey2"}}},
  {$sort: {total: n}}
])

return value can be sorted by adding {$sort: {total: n}}
- if n == 1, sort in ascending order
- if n == -1, sort in descending order

DATA TYPES
{
    string: "string",
    int: 123,
    double: 1.23,
    boolean: true,
    array: [1, 2, 3],
    object: {key1: "value1", key2: "value2"},
    date: new Date("<YYYY-mm-dd>"),
    object_id: <ObjectId>,
    no_value: null
}

ADDITIONAL DATA TYPES

Timestamp, Binary data, Regular expressions, javascript code

Getting started on nodejs

const { MongoClient } = require("mongodb");
const assert = require("assert");

// Connection URL
const url = "mongodb://localhost:27017/db_name";

// Use connect method to connect to the server
MongoClient.connect(url, (err, connection) => {
  assert.equal(null, err)l
  console.log("Connected correctly to server");
  connection.close();
});

MongoDB Overview

Getting started guide to setting up mongodb

for debugging

for graphical interface

Mongo shell basics

help​

show dbs​

use databaseName​

CRUD​

CREATE​

Add multiple Collections​

Add 1 or multiple collections​

READ​

Dos and Don'ts​

UPDATE​

REMOVE (Delete value(s))​

Delete collection​

Delete database​

BULKWRITE​

SEARCH​

Get search scores​

AGGREGATION​

Count​

List all values​

Find the sum of values under key for each id key​

ADDITIONAL DATA TYPES​

Getting started on nodejs

help

show dbs

use databaseName

CRUD

CREATE

Add multiple Collections

Add 1 or multiple collections

READ

Dos and Don'ts

UPDATE

REMOVE (Delete value(s))

Delete collection

Delete database

BULKWRITE

SEARCH

Get search scores

AGGREGATION

Count

List all values

Find the sum of values under key for each id key

ADDITIONAL DATA TYPES