tech.joshegan.com

mongoDB Cheatsheet

Posted on , updated on 

Resources

mongoDB

mongoose

About

What sets mongo apart?

History

The company was first established in 2007 as 10gen. Based in New York City, 10gen was founded by former DoubleClick founder and CTO Dwight Merriman and former DoubleClick CEO and Gilt Groupe founder Kevin P. Ryan with former Doubleclick engineer and ShopWiki founder and CTO Eliot Horowitz and received $81 million in venture capital funding from Flybridge Capital Partners, In-Q-Tel, Intel Capital, New Enterprise Associates (NEA), Red Hat, Sequoia Capital, and Union Square Ventures. 10gen originally aimed to build a platform as a service architecture based entirely on open source components; however, the company was unable to find an existing database platform that met their "principles" for a cloud architecture. As a result, the company began to develop a document-oriented database system it called MongoDB. After realizing the potential of the software on its own, 10gen's team decided to scrap its cloud platform and focus on maintaining MongoDB instead. - Wikipedia

Terminology

Term Definition
Database A physical container for collections. Each database gets its own set of files on the file system. A single MongoDB server typically has multiple databases.
Collection A group of MongoDB documents. It is the equivalent of an RDBMS table.
Document A set of key-value pairs. Documents have dynamic schema. Dynamic schema means that documents in the same collection do not need to have the same set of fields or structure, and common fields in a collection's documents may hold different types of data. A document is the equivalent of an RDBMS row.
Field A key-value pair of data within a document. A field is the equivalent of an RDBMS column.

Installation

https://docs.mongodb.com/manual/installation/

Installation on a mac:

brew update
brew install mongodb
brew services start mongodb

The default port on a mongo installation is 27017.

mongoDB CLI

https://docs.mongodb.com/manual/reference/mongo-shell/

The mongo shell is an interactive javascript shell. All of the commands are implemented in javascript. To see an implementation of a function, just leave off the parentheses and the function definition will be printed to the console.

Launch the mongoDB shell:

mongo
mongo --help
mongo -u username -p password --port 12345 --host localhost
mongo --shell my-script.js
mongo localhost:27017/myDatabase my-script.js

To exit the mongo shell type exit, type quit(), or use the keyboard shortcut Ctrl+C.

For a full list of options available when launching mongo see https://docs.mongodb.com/manual/reference/program/mongo/#bin.mongo.

Customize the Prompt

The prompt can be customized! That's cool. Full docs: https://docs.mongodb.com/manual/tutorial/configure-mongo-shell/#customize-the-prompt

  1. Open (create first if needed) your ~/.mongorc.js file.
  2. Add a function named prompt. This function must return a string.
  3. It appears that this function has access to all of the commands that are available in the mongo CLI, so you can really make this thing display anything you can dream up.
  4. If you want to use a special function like show, you'll need to use the JavaScript equivalent of the shell command. A list of equivalents can be found here: https://docs.mongodb.com/manual/tutorial/write-scripts-for-the-mongo-shell/#differences-between-interactive-and-scripted-mongo

If I end up doing much work from the command line hitting production boxes, I can imagine this being very useful. I could customize the output based on which host I'm on as a way to constantly remind me which database I'm connected to.

Here's the function I currently have in my .mongorc.js file:

function prompt () {
  return '\n\n' + db + ' > '
}

Root Commands

Command Notes
help Prints the help docs for the root commands.
exit Close the mongo db shell.
quit() Another way to close the shell.

Database Commands

Command Notes
db Displays the database that you're currently using.
show dbs Displays the databases available on the server.
use <db> Set the named database as the current database.
db.help() Print the help docs for all of the database commands.
db.stats() Print the stats for the current database.

Collection Commands

In this section, cn is used as an abbreviation for collectionName. Full docs for all collection methods: https://docs.mongodb.com/manual/reference/method/js-collection/

To create a collection, just insert a document into the collection. Mongo will create the collection if it does not exist.

Command Notes
show collections Prints the list of collections in the current database.
db.collection.help() Print the help docs for methods available on collection objects.
db.<cn>.<method>() To execute a method on a collection, the name of the collection is used on the db object. e.g. db.users.stats()
db.<cn>.insertOne({key: 'value'}) Create a single document. If the collection does not exist, it will be created. Use JavaScript object syntax.
db.<cn>.insertMany([]) Create many documents. Supply an array of objects.
db.<cn>.find() Retrieve documents. This method has a lot going on. See the section below dedicated specifically to this find method.
db.<cn>.count() Count the number of documents in a collection. A query can optionally be passed as a parameter.
db.<cn>.updateOne({name: 'foo'}, {$set: {age: 7}}, {upsert: true}) Update a document. The first param is a query filter and uses the same rules as the filter for the find function. The second param is the update to apply and should use update operators. The third param is an options object.
db.<cn>.updateMany(filter, update, options) Same signature as updateOne, but this function will update all documents that match the filter instead of only the first match.
db.<cn>.deleteOne({key: 'value'}) Delete a single document. This method accepts a query object similar to the query object accepted by the find method.
db.<cn>.deleteMany({key: 'value'}) Delete all documents that match the query object.
db.<cn>.explain().<method>() Show the query plan that would have been executed for the specified method.
db.<cn>.createIndex({orderDate: 1}) Creates an index on the specified keys. 1 == ascending and -1 == descending
db.<cn>.getIndexes() See all of the indexes on a collection.
db.<cn>.dropIndex(indexName) Removes the specified index.
db.<cn>.drop() Remove a collection from the database. Also removes any indexes on that collection.
db.<cn>.stats() Shows the stats for the collection.

db.<cn>.find()

The find method is one of the more complex collection methods, so it gets its own section here in the post. Complex methods have their own help docs. The find help docs can be accessed via: db.collection.find().help()

Here's some seed data that can be used to try out these commands on:

db.inventory.insertMany([
   { item: "journal", qty: 25, size: { h: 14, w: 21, uom: "cm" }, status: "A" },
   { item: "notebook", qty: 50, size: { h: 8.5, w: 11, uom: "in" }, status: "A" },
   { item: "paper", qty: 100, size: { h: 8.5, w: 11, uom: "in" }, status: "D" },
   { item: "planner", qty: 75, size: { h: 22.85, w: 30, uom: "cm" }, status: "D" },
   { item: "postcard", qty: 45, size: { h: 10, w: 15.25, uom: "cm" }, status: "A" }
]);

The anatomy of the find method:

var cursor = find(<predicate>, <projection>)

The predicate

The find predicate is used to define query criteria. The full query help docs.

mongo query SQL equivalent
db.inventory.find({}) SELECT * FROM inventory
db.inventory.find({ status: 'D' }) SELECT * FROM inventory WHERE status = "D"
db.inventory.find({ item: { $in: ['paper', 'planner'] } }) SELECT * FROM inventory WHERE item IN ("paper", "planner")
db.inventory.find({ status: 'A', qty: { $lt: 48 } }) SELECT * FROM inventory WHERE status = "A" AND qty < 48
db.inventory.find({ $or: [ { status: 'A' }, { qty: { $gt: 90 } } ] }) SELECT * FROM inventory WHERE status = "A" OR qty > 90
db.inventory.find({ item: /^p/, $or: [ { status: 'A' }, { qty: { $gt: 90 } } ] }) SELECT * FROM inventory WHERE item LIKE "p%" AND (status = "A" OR qty > 90)

Query operators: $eq, $ne, $gt, $gte, $lt, $lte, $in, $nin

The projection

Projection is used to transform the found object into a new object. Projection can be used to include or exclude fields, rename fields, and also to do simple transformations.

To include a field use a 1, to exclude a field use a 0. A projection cannot have a mixture of inclusions and exclusions, although you can exclude the _id field.

mongo query SQL equivalent
db.inventory.find({}) SELECT * FROM inventory
db.inventory.find({}, { item: 1, status: 1 }) SELECT _id, item, status FROM inventory
db.inventory.find({}, { _id: 0, item: 1 }) SELECT item FROM inventory
db.inventory.find({}, { _id: 0, size: 0 }) SELECT item, qty, status FROM inventory
db.inventory.find({}, { _id: 0, "size.uom": 1 })
db.inventory.find({}, { _id: 0, itemName: "$item" })
The cursor

The find() method returns a cursor. There are a whole bunch of cursor methods. Some of the cursor methods include: sort, limit, min, max, and forEach

db.inventory.find({}).limit(3).sort({ qty: 1})

Core Concepts

Documents

Every document contains a unique _id field. This field can be provided when inserting a document. If this field is not provided, then it will be automatically created. The _id field is a 12 byte hexadecimal number.

Transactions

mongoDB does not support true transactions because writing to a single document is atomic, but further atomicity cannot be guaranteed. The best transaction-like action that is available is the two-phase commit. A two phase commit is not atomic, which means that if the first write succeeds but the second write fails, a read from a separate process might read the first document before the rollback occurs and that first document is removed.

Aggregation

Projection Operator

Within aggregation, projection can do more than within a find() command. Aggregation projection docs are here.

Text Search

mongoDB supports searching for text within documents. First create a text index and then execute a text search.

author image
Josh Egan