One place for hosting & domains

      SQL Server Security Best Practices, Part 2


      This guide is the second in a series of articles that covers SQL Server security best practices.
      Part 1 of this series discussed a SQL Server installation’s physical security, operating system security, and application maintenance. Additionally, the previous guide outlined how to disable unnecessary features, enable encryption, and implement data masking.

      The second part of this series describes how and why you should:

      SQL Server Authentication

      Protection of data stored with SQL Server depends upon the ability to authenticate access to specific sets of data. SQL Server provides two options for database authentication in a Windows or Linux environment:

      You are prompted to select one of these SQL Server authentication modes during SQL Server setup.

      Note

      You can change the SQL Server authentication mode even after the initial installation decision has been made.

      Windows or Linux Authentication Mode

      In this mode, an installer logs into SQL Server using their Windows or Linux account. SQL Server validates the account name and password via the Windows or Linux operating system. SQL Server does not prompt for a password and does not perform the validation.

      Windows or Linux authentication uses Active Directory (AD) accounts. As a result, you can have centralized policy control for authentication. Policies can govern password strength and complexity, password expiration, account lockout, and active directory groups in the active directory.

      Windows or Linux-based authentication is the default authentication mode and is much more secure than
      SQL Server Authentication (discussed in the next section). Windows or Linux Authentication uses the Kerberos security protocol to support the above-mentioned security features. A connection made using Windows or Linux Authentication is sometimes called a trusted connection because SQL Server trusts the credentials provided by the underlying Windows or Linux operating system.

      SQL Server and Windows/Linux Authentication Mode (Mixed-Mode)

      When using SQL Server Authentication, logins are created in SQL Server and are not based on Windows or Linux user accounts. Both the username and the password are created
      by SQL Server and are stored within SQL Server. Users connecting using SQL Server Authentication must provide their credentials (username and password) every time that they connect to SQL Server.

      This mode does not use the Windows or Linux Kerberos security protocol, and it is considered to be inferior to
      Windows or Linux Authentication mode.

      System Administrator (SA) Account

      If you are using
      SQL Server (mixed-mode) authentication, SQL Server automatically creates a System Administrator (SA) user login with sysadmin privileges and permissions. To increase the security of your SQL Server, you should perform the following:

      1. Rename the SA login account to a different, more obscure, name.
      2. Disable the account entirely, if you do not plan on using it.
      3. For the SA (or renamed) account, select a complex password, consisting of lower/upper case letters, numbers, and punctuation symbols.
      4. Do not allow applications to use the SA (or equivalently renamed) account in any of the application connection strings.

      Note

      Any other user-based (lower-privileged) SQL Server accounts should also use complex, unique passwords.

      High-Privileged Operating System Accounts

      SQL Server uses a Windows or Linux account to run its services. Typically one should not assign high-privileged, built-in accounts (or equivalents) such as Network Service or Local System to the various SQL Server services. This can increase the risk of nefarious database/server activity, should someone be able to log into these types of accounts.

      Only assign the appropriate level of security-required accounts to SQL Server services. If not needed, any high-privileged operating system accounts on the server housing the SQL Server should be disabled as appropriate.

      Restrict SQL Traffic

      Database servers typically have one or more servers connecting to them. Access to these servers must be allowed only to and from designated IP addresses. Doing this can potentially prevent a nefarious user from accessing the server. In certain cases, a user of SQL Server may need to connect directly to the database. Restricting those SQL connections to the specific IP addresses (or at least IP class block or segment) that require it should be implemented.

      These IP restrictions can be managed with different solutions on different platforms:

      SQL Server Patches (Service Packs)

      Microsoft regularly releases SQL Server service packs and/or cumulative packs for fixing known issues, bugs, and security issues. It is highly advisable to apply SQL Server patching on production instances of SQL Server. However, before applying a security patch to production systems, it is advisable to apply these patches in a test environment. This is done to validate the changes in the patch and ensure that your database operates as expected under the patch.

      Backups

      When dealing with production instances of SQL Server, it is important to regularly backup the server’s databases. A database backup creates a copy of the operational state, architecture, and stored data of a database. Backups help guard against potential database failures. These failures can happen because of corruption, disk array failure, power outages, disasters, and other scenarios.

      Backups can also assist with non-failure scenarios where a rollback of your database to a particular date may be necessary. Full database backups (on a regularly scheduled basis) and incremental backups (on a daily or running time basis) should be performed and maintained.

      Securing your backups is critical, and database professionals sometimes do not consider all of the requirements for securing database backups. This work includes:

      • Restriction of access to backup files. Do not provide all people in your organization the access rights (create, view, modify, and delete) to backup files.

      • Encrypting backup files properly.

      • Storing backups in an off-site facility. Depending on the organization and the critical nature of the database data, backups of a certain age should be preserved and archived.

      Auditing

      Auditing is another key component of SQL Server security. A designated database administrator or database security team should regularly review SQL Server auditing logs for failed logins.

      SQL Server provides a default login audit mechanism for reviewing all of the login accounts. These audit facilities record incoming requests by username and client IP address. Login failures can assist in discovering and eliminating suspicious database activity. The following types of activity can show up in the SQL Server audit logs:

      • Extended Events: Extended Events is a lightweight performance monitoring system that enables users to collect data needed to monitor and troubleshoot problems in SQL Server.

      • SQL Trace: SQL Trace is SQL Server’s built-in utility that monitors and records SQL Server database activity. This utility can display server activity, create filters that focus on the actions of users, applications, or workstations, and can filter at the SQL command level.

      • Change Data Capture: Change Data Capture (CDC) uses a SQL Server agent to record insert, update, and delete activity that applies to a specific table.

      • Triggers: Application-based SQL Server Triggers can be written specifically to populate a user-defined audit table to store changes to existing records in specific tables.

      • SQL Server-Level Audit Specifications: A Server Audit Specification defines which Audit Action Groups can be audited for the entire server (or instance). Some audit action groups consist of server-level actions such as the creation of a table or modification of a server role. These are only applicable to the server itself.

      Hardware and/or software firewall logs (that is, external to SQL Server) should be regularly examined to monitor and detect any nefarious attempts at server penetration.

      Conclusion

      In part two of this article series, you reviewed additional methods of enhancing the security of SQL Server databases. These included choosing an
      authentication mode, restricting the
      System Administrator account, assignment of
      security-friendly accounts to SQL Server,
      restricting SQL traffic, application of
      patch updates,
      backup strategies, and use of
      auditing. To review earlier security recommendations, revisit
      Part 1: SQL Server Security Best Practices.



      Source link

      Build a RESTful JSON API With Rails 5 – Part One


      Rails is popularly known for building web applications. Chances are if you’re reading this you’ve built a traditional server-rendered web application with Rails before. If not, I’d highly recommend going through the Getting Started with Rails page to familiarize yourself with the Rails framework before proceeding with this tutorial.

      As of version 5, Rails core now supports API only applications! In previous versions, we relied on an external gem: rails-api which has since been merged to core rails.

      API only applications are slimmed down compared to traditional Rails web applications. According to Rails 5 release notes, generating an API only application will:

      • Start the application with a limited set of middleware
      • Make the ApplicationController inherit from ActionController::API instead of ActionController::Base
      • Skip generation of view files

      This works to generate an API-centric framework excluding functionality that would otherwise be unused and unnecessary.

      In this three-part tutorial, we’ll build a todo list API where users can manage their to-do lists and todo items.

      Prerequisites

      Before we begin, make sure you have ruby version >=2.2.2 and rails version 5.

      $ ruby -v # ruby 2.3.0p0 (2015-12-25 revision 53290) [x86_64-darwin16]
      $ rails -v # Rails 5.0.1
      

      If your ruby version is not up to date, you can update it with a ruby version manager like rvm or rbenv.

      # when using rbenv
      $ rbenv install 2.3.1
      # set 2.3.1 as the global version
      $ rbenv global 2.3.1
      
      # when using rvm
      $ rvm install 2.3.1
      # set 2.3.1 as the global version
      $ rvm use 2.3.1
      

      If your rails version is not up to date, update to the latest version by running:

      $ gem update rails
      

      All good? Let’s get started!

      API Endpoints

      Our API will expose the following RESTful endpoints.

      EndpointFunctionality
      POST /signupSignup
      POST /auth/loginLogin
      GET /auth/logoutLogout
      GET /todosList all todos
      POST /todosCreate a new todo
      GET /todos/:idGet a todo
      PUT /todos/:idUpdate a todo
      DELETE /todos/:idDelete a todo and its items
      GET /todos/:id/itemsGet a todo item
      PUT /todos/:id/itemsUpdate a todo item
      DELETE /todos/:id/itemsDelete a todo item

      Part One will Cover:

      • Project setup
      • Todos API
      • TodoItems API

      Project Setup

      Generate a new project todos-api by running:

      $ rails new todos-api --api -T
      

      Note that we’re using the --api argument to tell Rails that we want an API application and -T to exclude Minitest the default
      testing framework. Don’t freak out, we’re going to write tests. We’ll be using RSpec instead to test our API. I find RSpec to be more expressive
      and easier to start with as compared to Minitest.

      Dependencies

      Let’s take a moment to review the gems that we’ll be using.

      • rspec-rails – Testing framework.
      • factorybotrails – A fixtures replacement with a more straightforward syntax. You’ll see.
      • shoulda_matchers – Provides RSpec with additional matchers.
      • database_cleaner – You guessed it! It literally cleans our test database to ensure
        a clean state in each test suite.
      • faker – A library for generating fake data. We’ll use this to generate test data.

      All good? Great! Let’s set them up. In your Gemfile:

      Add rspec-rails to both the :development and :test groups.

      # Gemfile
      group :development, :test do
        gem 'rspec-rails', '~> 3.5'
      end
      

      This is a handy shorthand to include a gem in multiple environments.

      Add factory_bot_rails, shoulda_matchers, faker and database_cleaner to the :test group.

      # Gemfile
      group :test do
        gem 'factory_bot_rails', '~> 4.0'
        gem 'shoulda-matchers', '~> 3.1'
        gem 'faker'
        gem 'database_cleaner'
      end
      

      Install the gems by running:

      $ bundle install
      

      Initialize the spec directory (where our tests will reside).

      $ rails generate rspec:install
      

      This adds the following files which are used for configuration:

      • .rspec
      • spec/spec_helper.rb
      • spec/rails_helper.rb

      Create a factories directory (factory bot uses this as the default directory). This is where we’ll define the model factories.

      $ mkdir spec/factories
      

      Configuration

      In spec/rails_helper.rb

      # require database cleaner at the top level
      require 'database_cleaner'
      
      # [...]
      # configure shoulda matchers to use rspec as the test framework and full matcher libraries for rails
      Shoulda::Matchers.configure do |config|
        config.integrate do |with|
          with.test_framework :rspec
          with.library :rails
        end
      end
      
      # [...]
      RSpec.configure do |config|
        # [...]
        # add `FactoryBot` methods
        config.include FactoryBot::Syntax::Methods
      
        # start by truncating all the tables but then use the faster transaction strategy the rest of the time.
        config.before(:suite) do
          DatabaseCleaner.clean_with(:truncation)
          DatabaseCleaner.strategy = :transaction
        end
      
        # start the transaction strategy as examples are run
        config.around(:each) do |example|
          DatabaseCleaner.cleaning do
            example.run
          end
        end
        # [...]
      end
      

      Phew! That was a rather long. Good thing is, it’s a smooth ride from here on out.


      Models

      Let’s start by generating the Todo model

      $ rails g model Todo title:string created_by:string
      

      Notice that we’ve included the model attributes in the model generation command. This way we don’t have to edit the migration file.
      The generator invokes active record and rspec to generate the migration, model, and spec respectively.

      # db/migrate/[timestamp]_create_todos.rb
      class CreateTodos < ActiveRecord::Migration[5.0]
        def change
          create_table :todos do |t|
            t.string :title
            t.string :created_by
      
            t.timestamps
          end
        end
      end
      

      And now the Item model

      $ rails g model Item name:string done:boolean todo:references
      

      By adding todo:references we’re telling the generator to set up an association with the Todo model.
      This will do the following:

      • Add a foreign key column todo_id to the items table
      • Setup a belongs_to association in the Item model
      # db/migrate/[timestamp]_create_items.rb
      class CreateItems < ActiveRecord::Migration[5.0]
        def change
          create_table :items do |t|
            t.string :name
            t.boolean :done
            t.references :todo, foreign_key: true
      
            t.timestamps
          end
        end
      end
      

      Looks good? Let’s run the migrations.

      $ rails db:migrate
      

      We’re Test Driven, let’s write the model specs first.

      # spec/models/todo_spec.rb
      require 'rails_helper'
      
      # Test suite for the Todo model
      RSpec.describe Todo, type: :model do
        # Association test
        # ensure Todo model has a 1:m relationship with the Item model
        it { should have_many(:items).dependent(:destroy) }
        # Validation tests
        # ensure columns title and created_by are present before saving
        it { should validate_presence_of(:title) }
        it { should validate_presence_of(:created_by) }
      end
      

      RSpec has a very expressive DSL (Domain Specific Language). You can almost read the tests like a paragraph.
      Remember our shoulda matchers gem? It provides RSpec with the nifty association and validation matchers above.

      # spec/models/item_spec.rb
      require 'rails_helper'
      
      # Test suite for the Item model
      RSpec.describe Item, type: :model do
        # Association test
        # ensure an item record belongs to a single todo record
        it { should belong_to(:todo) }
        # Validation test
        # ensure column name is present before saving
        it { should validate_presence_of(:name) }
      end
      

      Let’s execute the specs by running:

      $ bundle exec rspec
      

      And to no surprise, we have only one test passing and four failures. Let’s go ahead and fix the failures.

      # app/models/todo.rb
      class Todo < ApplicationRecord
        # model association
        has_many :items, dependent: :destroy
      
        # validations
        validates_presence_of :title, :created_by
      end
      
      # app/models/item.rb
      class Item < ApplicationRecord
        # model association
        belongs_to :todo
      
        # validation
        validates_presence_of :name
      end
      

      At this point run the tests again and…

      voila! All green.


      Controllers

      Now that our models are all setup, let’s generate the controllers.

      $ rails g controller Todos
      $ rails g controller Items
      

      You guessed it! Tests first… with a slight twist. Generating controllers by default generates controller specs.
      However, we won’t be writing any controller specs. We’re going to write request specs instead.

      Request specs are designed to drive behavior through the full stack, including routing. This means they can hit the applications’
      HTTP endpoints as opposed to controller specs which call methods directly. Since we’re building an API application, this is exactly the kind of behavior we want from our tests.

      According to RSpec, the official recommendation of the Rails team and the RSpec core team is to write request specs instead.

      Add a requests folder to the spec directory with the corresponding spec files.

      $ mkdir spec/requests && touch spec/requests/{todos_spec.rb,items_spec.rb} 
      

      Before we define the request specs, Let’s add the model factories which will provide the test data.

      Add the factory files:

      $ touch spec/factories/{todos.rb,items.rb}
      

      Define the factories.

      # spec/factories/todos.rb
      FactoryBot.define do
        factory :todo do
          title { Faker::Lorem.word }
          created_by { Faker::Number.number(10) }
        end
      end
      

      By wrapping faker methods in a block, we ensure that faker generates dynamic data every time the factory is invoked.
      This way, we always have unique data.

      # spec/factories/items.rb
      FactoryBot.define do
        factory :item do
          name { Faker::StarWars.character }
          done false
          todo_id nil
        end
      end
      

      Todo API

      # spec/requests/todos_spec.rb
      require 'rails_helper'
      
      RSpec.describe 'Todos API', type: :request do
        # initialize test data 
        let!(:todos) { create_list(:todo, 10) }
        let(:todo_id) { todos.first.id }
      
        # Test suite for GET /todos
        describe 'GET /todos' do
          # make HTTP get request before each example
          before { get '/todos' }
      
          it 'returns todos' do
            # Note `json` is a custom helper to parse JSON responses
            expect(json).not_to be_empty
            expect(json.size).to eq(10)
          end
      
          it 'returns status code 200' do
            expect(response).to have_http_status(200)
          end
        end
      
        # Test suite for GET /todos/:id
        describe 'GET /todos/:id' do
          before { get "/todos/#{todo_id}" }
      
          context 'when the record exists' do
            it 'returns the todo' do
              expect(json).not_to be_empty
              expect(json['id']).to eq(todo_id)
            end
      
            it 'returns status code 200' do
              expect(response).to have_http_status(200)
            end
          end
      
          context 'when the record does not exist' do
            let(:todo_id) { 100 }
      
            it 'returns status code 404' do
              expect(response).to have_http_status(404)
            end
      
            it 'returns a not found message' do
              expect(response.body).to match(/Couldn't find Todo/)
            end
          end
        end
      
        # Test suite for POST /todos
        describe 'POST /todos' do
          # valid payload
          let(:valid_attributes) { { title: 'Learn Elm', created_by: '1' } }
      
          context 'when the request is valid' do
            before { post '/todos', params: valid_attributes }
      
            it 'creates a todo' do
              expect(json['title']).to eq('Learn Elm')
            end
      
            it 'returns status code 201' do
              expect(response).to have_http_status(201)
            end
          end
      
          context 'when the request is invalid' do
            before { post '/todos', params: { title: 'Foobar' } }
      
            it 'returns status code 422' do
              expect(response).to have_http_status(422)
            end
      
            it 'returns a validation failure message' do
              expect(response.body)
                .to match(/Validation failed: Created by can't be blank/)
            end
          end
        end
      
        # Test suite for PUT /todos/:id
        describe 'PUT /todos/:id' do
          let(:valid_attributes) { { title: 'Shopping' } }
      
          context 'when the record exists' do
            before { put "/todos/#{todo_id}", params: valid_attributes }
      
            it 'updates the record' do
              expect(response.body).to be_empty
            end
      
            it 'returns status code 204' do
              expect(response).to have_http_status(204)
            end
          end
        end
      
        # Test suite for DELETE /todos/:id
        describe 'DELETE /todos/:id' do
          before { delete "/todos/#{todo_id}" }
      
          it 'returns status code 204' do
            expect(response).to have_http_status(204)
          end
        end
      end
      

      We start by populating the database with a list of 10 todo records (thanks to factory bot).
      We also have a custom helper method json which parses the JSON response to a Ruby Hash which is easier to work with in our tests.
      Let’s define it in spec/support/request_spec_helper.

      Add the directory and file:

      $ mkdir spec/support && touch spec/support/request_spec_helper.rb
      
      # spec/support/request_spec_helper
      module RequestSpecHelper
        # Parse JSON response to ruby hash
        def json
          JSON.parse(response.body)
        end
      end
      

      The support directory is not autoloaded by default. To enable this, open the rails helper and comment out the support directory auto-loading and then
      include it as shared module for all request specs in the RSpec configuration block.

      # spec/rails_helper.rb
      # [...]
      Dir[Rails.root.join('spec/support/**/*.rb')].each { |f| require f }
      # [...]
      RSpec.configuration do |config|
        # [...]
        config.include RequestSpecHelper, type: :request
        # [...]
      end
      

      Run the tests.

      We get failing routing errors. This is because we haven’t defined the routes yet. Go ahead and define them in config/routes.rb.

      # config/routes.rb
      Rails.application.routes.draw do
        resources :todos do
          resources :items
        end
      end
      

      In our route definition, we’re creating todo resource with a nested items resource. This enforces the 1:m (one to many) associations at the routing level.
      To view the routes, you can run:

      $ rails routes
      

      When we run the tests we see that the routing error is gone. As expected we have controller failures. Let’s go ahead and define the controller methods.

      # app/controllers/todos_controller.rb
      class TodosController < ApplicationController
        before_action :set_todo, only: [:show, :update, :destroy]
      
        # GET /todos
        def index
          @todos = Todo.all
          json_response(@todos)
        end
      
        # POST /todos
        def create
          @todo = Todo.create!(todo_params)
          json_response(@todo, :created)
        end
      
        # GET /todos/:id
        def show
          json_response(@todo)
        end
      
        # PUT /todos/:id
        def update
          @todo.update(todo_params)
          head :no_content
        end
      
        # DELETE /todos/:id
        def destroy
          @todo.destroy
          head :no_content
        end
      
        private
      
        def todo_params
          # whitelist params
          params.permit(:title, :created_by)
        end
      
        def set_todo
          @todo = Todo.find(params[:id])
        end
      end
      

      More helpers. Yay! This time we have:

      • json_response which does… yes, responds with JSON and an HTTP status code (200 by default).
        We can define this method in concerns folder.
      # app/controllers/concerns/response.rb
      module Response
        def json_response(object, status = :ok)
          render json: object, status: status
        end
      end
      
      • set_todo – callback method to find a todo by id. In the case where the record does not exist, ActiveRecord
        will throw an exception ActiveRecord::RecordNotFound. We’ll rescue from this exception and return a 404 message.
      # app/controllers/concerns/exception_handler.rb
      module ExceptionHandler
        # provides the more graceful `included` method
        extend ActiveSupport::Concern
      
        included do
          rescue_from ActiveRecord::RecordNotFound do |e|
            json_response({ message: e.message }, :not_found)
          end
      
          rescue_from ActiveRecord::RecordInvalid do |e|
            json_response({ message: e.message }, :unprocessable_entity)
          end
        end
      end
      

      In our create method in the TodosController, note that we’re using create! instead of create. This way, the model will raise
      an exception ActiveRecord::RecordInvalid. This way, we can avoid deep nested if statements in the controller. Thus, we rescue from this exception
      in the ExceptionHandler module.

      However, our controller classes don’t know about these helpers yet. Let’s fix that by including these modules in the
      application controller.

      # app/controllers/application_controller.rb
      class ApplicationController < ActionController::API
        include Response
        include ExceptionHandler
      end
      

      Run the tests and everything’s all green!

      Let’s fire up the server for some good old manual testing.

      $ rails s
      

      Now let’s go ahead and make requests to the API. I’ll be using httpie as my HTTP client.

      # GET /todos
      $ http :3000/todos
      # POST /todos
      $ http POST :3000/todos title=Mozart created_by=1
      # PUT /todos/:id
      $ http PUT :3000/todos/1 title=Beethoven
      # DELETE /todos/:id
      $ http DELETE :3000/todos/1
      

      You should see similar output.


      TodoItems API

      # spec/requests/items_spec.rb
      require 'rails_helper'
      
      RSpec.describe 'Items API' do
        # Initialize the test data
        let!(:todo) { create(:todo) }
        let!(:items) { create_list(:item, 20, todo_id: todo.id) }
        let(:todo_id) { todo.id }
        let(:id) { items.first.id }
      
        # Test suite for GET /todos/:todo_id/items
        describe 'GET /todos/:todo_id/items' do
          before { get "/todos/#{todo_id}/items" }
      
          context 'when todo exists' do
            it 'returns status code 200' do
              expect(response).to have_http_status(200)
            end
      
            it 'returns all todo items' do
              expect(json.size).to eq(20)
            end
          end
      
          context 'when todo does not exist' do
            let(:todo_id) { 0 }
      
            it 'returns status code 404' do
              expect(response).to have_http_status(404)
            end
      
            it 'returns a not found message' do
              expect(response.body).to match(/Couldn't find Todo/)
            end
          end
        end
      
        # Test suite for GET /todos/:todo_id/items/:id
        describe 'GET /todos/:todo_id/items/:id' do
          before { get "/todos/#{todo_id}/items/#{id}" }
      
          context 'when todo item exists' do
            it 'returns status code 200' do
              expect(response).to have_http_status(200)
            end
      
            it 'returns the item' do
              expect(json['id']).to eq(id)
            end
          end
      
          context 'when todo item does not exist' do
            let(:id) { 0 }
      
            it 'returns status code 404' do
              expect(response).to have_http_status(404)
            end
      
            it 'returns a not found message' do
              expect(response.body).to match(/Couldn't find Item/)
            end
          end
        end
      
        # Test suite for PUT /todos/:todo_id/items
        describe 'POST /todos/:todo_id/items' do
          let(:valid_attributes) { { name: 'Visit Narnia', done: false } }
      
          context 'when request attributes are valid' do
            before { post "/todos/#{todo_id}/items", params: valid_attributes }
      
            it 'returns status code 201' do
              expect(response).to have_http_status(201)
            end
          end
      
          context 'when an invalid request' do
            before { post "/todos/#{todo_id}/items", params: {} }
      
            it 'returns status code 422' do
              expect(response).to have_http_status(422)
            end
      
            it 'returns a failure message' do
              expect(response.body).to match(/Validation failed: Name can't be blank/)
            end
          end
        end
      
        # Test suite for PUT /todos/:todo_id/items/:id
        describe 'PUT /todos/:todo_id/items/:id' do
          let(:valid_attributes) { { name: 'Mozart' } }
      
          before { put "/todos/#{todo_id}/items/#{id}", params: valid_attributes }
      
          context 'when item exists' do
            it 'returns status code 204' do
              expect(response).to have_http_status(204)
            end
      
            it 'updates the item' do
              updated_item = Item.find(id)
              expect(updated_item.name).to match(/Mozart/)
            end
          end
      
          context 'when the item does not exist' do
            let(:id) { 0 }
      
            it 'returns status code 404' do
              expect(response).to have_http_status(404)
            end
      
            it 'returns a not found message' do
              expect(response.body).to match(/Couldn't find Item/)
            end
          end
        end
      
        # Test suite for DELETE /todos/:id
        describe 'DELETE /todos/:id' do
          before { delete "/todos/#{todo_id}/items/#{id}" }
      
          it 'returns status code 204' do
            expect(response).to have_http_status(204)
          end
        end
      end
      

      As expected, running the tests at this point should output failing todo item tests. Let’s define the todo items controller.

      # app/controllers/items_controller.rb
      class ItemsController < ApplicationController
        before_action :set_todo
        before_action :set_todo_item, only: [:show, :update, :destroy]
      
        # GET /todos/:todo_id/items
        def index
          json_response(@todo.items)
        end
      
        # GET /todos/:todo_id/items/:id
        def show
          json_response(@item)
        end
      
        # POST /todos/:todo_id/items
        def create
          @todo.items.create!(item_params)
          json_response(@todo, :created)
        end
      
        # PUT /todos/:todo_id/items/:id
        def update
          @item.update(item_params)
          head :no_content
        end
      
        # DELETE /todos/:todo_id/items/:id
        def destroy
          @item.destroy
          head :no_content
        end
      
        private
      
        def item_params
          params.permit(:name, :done)
        end
      
        def set_todo
          @todo = Todo.find(params[:todo_id])
        end
      
        def set_todo_item
          @item = @todo.items.find_by!(id: params[:id]) if @todo
        end
      end
      

      Run the tests.

      Run some manual tests for the todo items API:

      # GET /todos/:todo_id/items
      $ http :3000/todos/2/items
      # POST /todos/:todo_id/items
      $ http POST :3000/todos/2/items name="Listen to 5th Symphony" done=false
      # PUT /todos/:todo_id/items/:id
      $ http PUT :3000/todos/2/items/1 done=true
      # DELETE /todos/:todo_id/items/1
      $ http DELETE :3000/todos/2/items/1
      


      Conclusion

      That’s it for part one! At this point you should have learned how to:

      • Generate an API application with Rails 5
      • Setup RSpec testing framework with Factory Bot, Database Cleaner, Shoulda Matchers and Faker.
      • Build models and controllers with TDD (Test Driven Development).
      • Make HTTP requests to an API with httpie.

      In the next part, we’ll cover authentication with JWT, pagination, and API versioning. Hope to see you there. Cheers!



      Source link

      Build a CRUD Web App With Python and Flask – Part One


      In this three-part tutorial, we’ll build a CRUD (Create, Read, Update, Delete) employee management web app using Flask, a microframework for Python. I’ve named the app Project Dream Team, and it will have the following features:

      1. Users will be able to register and login as employees
      2. The administrator will be able to create, update, and delete departments and roles
      3. The administrator will be able to assign employees to a department and assign them roles
      4. The administrator will be able to view all employees and their details

      Part One will cover:

      1. Database setup
      2. Models
      3. Migration
      4. Homepage
      5. Authentication

      Ready? Here we go!

      Prerequisites

      This tutorial builds on my introductory tutorial, Getting Started With Flask, picking up where it left off. It assumes you have, to begin with, the following dependencies installed:

      1. Python 2.7
      2. Flask
      3. virtualenv (and, optionally, virtualenvwrapper)

      You should have a virtual environment set up and activated. You should also have the following file and directory structure:

      ├── dream-team
             ├── app
             │   ├── __init__.py
             │   ├── templates
             │   ├── models.py
             │   └── views.py
             ├── config.py
             ├── requirements.txt
             └── run.py
      

      This project structure groups the similar components of the application together. The dream-team directory houses all the project files. The app directory is the application package, and houses different but interlinked modules of the application. All templates are stored in the templates directory, all models are in the models.py file, and all routes are in the views.py file. The run.py file is the application’s entry point, the config.py file contains the application configurations, and the requirements.txt file contains the software dependencies for the application.

      If you don’t have these set up, please visit the introductory tutorial and catch up!

      Database Setup

      Flask has support for several relational database management systems, including SQLite, MySQL, and PostgreSQL. For this tutorial, we will be using MySQL. It’s popular and therefore has a lot of support, in addition to being scalable, secure, and rich in features.

      We will install the following (remember to activate your virtual environment):

      1. Flask-SQLAlchemy: This will allow us to use SQLAlchemy, a useful tool for SQL use with Python. SQLAlchemy is an Object Relational Mapper (ORM), which means that it connects the objects of an application to tables in a relational database management system. These objects can be stored in the database and accessed without the need to write raw SQL. This is convenient because it simplifies queries that may have been complex if written in raw SQL. Additionally, it reduces the risk of SQL injection attacks since we are not dealing with the input of raw SQL.

      2. MySQL-Python: This is a Python interface to MySQL. It will help us connect the MySQL database to the app.

      $ pip install flask-sqlalchemy mysql-python
      

      We’ll then create the MySQL database. Ensure you have MySQL installed and running, and then log in as the root user:

      $ mysql -u root
      
      mysql> CREATE USER 'dt_admin'@'localhost' IDENTIFIED BY 'dt2016';
      Query OK, 0 rows affected (0.00 sec)
      
      mysql> CREATE DATABASE dreamteam_db;
      Query OK, 1 row affected (0.00 sec)
      
      mysql> GRANT ALL PRIVILEGES ON dreamteam_db . * TO 'dt_admin'@'localhost';
      Query OK, 0 rows affected (0.00 sec)
      

      We have now created a new user dt_admin with the password dt2016, created a new database dreamteam_db, and granted the new user all database privileges.

      Next, let’s edit the config.py. Remove any exisiting code and add the following:

      # config.py
      
      class Config(object):
          """
          Common configurations
          """
      
          # Put any configurations here that are common across all environments
      
      
      class DevelopmentConfig(Config):
          """
          Development configurations
          """
      
          DEBUG = True
          SQLALCHEMY_ECHO = True
      
      
      class ProductionConfig(Config):
          """
          Production configurations
          """
      
          DEBUG = False
      
      app_config = {
          'development': DevelopmentConfig,
          'production': ProductionConfig
      }
      

      It is good practice to specify configurations for different environments. In the file above, we have specifed configurations for development, which we will use while building the app and running it locally, as well as production, which we will use when the app is deployed.

      Some useful configuration variables are:

      1. TESTING: setting this to True activates the testing mode of Flask extensions. This allows us to use testing properties that could for instance have an increased runtime cost, such as unittest helpers. It should be set to True in the configurations for testing. It defaults to False.
      2. DEBUG: setting this to True activates the debug mode on the app. This allows us to use the Flask debugger in case of an unhandled exception, and also automatically reloads the application when it is updated. It should however always be set to False in production. It defaults to False.
      3. SQLALCHEMY_ECHO: setting this to True helps us with debugging by allowing SQLAlchemy to log errors.

      You can find more Flask configuration variables here and SQLAlchemy configuration variables here.

      Next, create an instance directory in the dream-team directory, and then create a config.py file inside it. We will put configuration variables here that will not be pushed to version control due to their sensitive nature. In this case, we put the secret key as well as the database URI which contains the database user password.

      # instance/config.py
      
      SECRET_KEY = 'p9Bv<3Eid9%$i01'
      SQLALCHEMY_DATABASE_URI = 'mysql://dt_admin:dt2016@localhost/dreamteam_db'
      

      Now, let’s edit the app/__init__.py file. Remove any existing code and add the following:

      # app/__init__.py
      
      # third-party imports
      from flask import Flask
      from flask_sqlalchemy import SQLAlchemy
      
      # local imports
      from config import app_config
      
      # db variable initialization
      db = SQLAlchemy()
      
      
      def create_app(config_name):
          app = Flask(__name__, instance_relative_config=True)
          app.config.from_object(app_config[config_name])
          app.config.from_pyfile('config.py')
          db.init_app(app)
      
          return app
      

      We’ve created a function, create_app that, given a configuration name, loads the correct configuration from the config.py file, as well as the configurations from the instance/config.py file. We have also created a db object which we will use to interact with the database.

      Next, let’s edit the run.py file:

      # run.py
      
      import os
      
      from app import create_app
      
      config_name = os.getenv('FLASK_CONFIG')
      app = create_app(config_name)
      
      
      if __name__ == '__main__':
          app.run()
      

      We create the app by running the create_app function and passing in the configuration name. We get this from the OS environment variable FLASK_CONFIG. Because we are in development, we should set the environment variable to development.

      Let’s run the app to ensure everything is working as expected. First, delete the app/views.py file as well as the app/templates directory as we will not be needing them going forward. Next, add a temporary route to the app/__init__.py file as follows:

      # app/__init__.py
      
      # existing code remains
      
      
      def create_app(config_name):
          # existing code remains
      
          # temporary route
          @app.route('/')
          def hello_world():
              return 'Hello, World!'
      
          return app
      

      Make sure you set the FLASK_CONFIG and FLASK_APP environment variables before running the app:

      $ export FLASK_CONFIG=development
      $ export FLASK_APP=run.py
      $ flask run
       * Serving Flask app "run"
       * Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)
      

      We can see the “Hello, World” string we set in the route. The app is working well so far.

      Models

      Now to work on the models. Remember that a model is a representation of a database table in code. We’ll need three models: Employee, Department, and Role.

      But first, let’s install Flask-Login, which will help us with user management and handle logging in, logging out, and user sessions. The Employee model will inherit from Flask-Login’s UserMixin class which will make it easier for us to make use of its properties and methods.

      $ pip install flask-login
      

      To use Flask-Login, we need to create a LoginManager object and initialize it in the app/__init__.py file. First, remove the route we added earlier, and then add the following:

      # app/__init__.py
      
      # after existing third-party imports
      from flask_login import LoginManager
      
      # after the db variable initialization
      login_manager = LoginManager()
      
      
      def create_app(config_name):
          # existing code remains
      
          login_manager.init_app(app)
          login_manager.login_message = "You must be logged in to access this page."
          login_manager.login_view = "auth.login"
      
          return app
      

      In addition to initializing the LoginManager object, we’ve also added a login_view and login_message to it. This way, if a user tries to access a page that they are not authorized to, it will redirect to the specified view and display the specified message. We haven’t created the auth.login view yet, but we will when we get to authentication.

      Now add the following code to the app/models.py file:

      # app/models.py
      
      from flask_login import UserMixin
      from werkzeug.security import generate_password_hash, check_password_hash
      
      from app import db, login_manager
      
      
      class Employee(UserMixin, db.Model):
          """
          Create an Employee table
          """
      
          # Ensures table will be named in plural and not in singular
          # as is the name of the model
          __tablename__ = 'employees'
      
          id = db.Column(db.Integer, primary_key=True)
          email = db.Column(db.String(60), index=True, unique=True)
          username = db.Column(db.String(60), index=True, unique=True)
          first_name = db.Column(db.String(60), index=True)
          last_name = db.Column(db.String(60), index=True)
          password_hash = db.Column(db.String(128))
          department_id = db.Column(db.Integer, db.ForeignKey('departments.id'))
          role_id = db.Column(db.Integer, db.ForeignKey('roles.id'))
          is_admin = db.Column(db.Boolean, default=False)
      
          @property
          def password(self):
              """
              Prevent pasword from being accessed
              """
              raise AttributeError('password is not a readable attribute.')
      
          @password.setter
          def password(self, password):
              """
              Set password to a hashed password
              """
              self.password_hash = generate_password_hash(password)
      
          def verify_password(self, password):
              """
              Check if hashed password matches actual password
              """
              return check_password_hash(self.password_hash, password)
      
          def __repr__(self):
              return '<Employee: {}>'.format(self.username)
      
      
      # Set up user_loader
      @login_manager.user_loader
      def load_user(user_id):
          return Employee.query.get(int(user_id))
      
      
      class Department(db.Model):
          """
          Create a Department table
          """
      
          __tablename__ = 'departments'
      
          id = db.Column(db.Integer, primary_key=True)
          name = db.Column(db.String(60), unique=True)
          description = db.Column(db.String(200))
          employees = db.relationship('Employee', backref="department",
                                      lazy='dynamic')
      
          def __repr__(self):
              return '<Department: {}>'.format(self.name)
      
      
      class Role(db.Model):
          """
          Create a Role table
          """
      
          __tablename__ = 'roles'
      
          id = db.Column(db.Integer, primary_key=True)
          name = db.Column(db.String(60), unique=True)
          description = db.Column(db.String(200))
          employees = db.relationship('Employee', backref="role",
                                      lazy='dynamic')
      
          def __repr__(self):
              return '<Role: {}>'.format(self.name)
      

      In the Employee model, we make use of some of Werkzeug’s handy security helper methods, generate_password_hash, which allows us to hash passwords, and check_password_hash, which allows us ensure the hashed password matches the password. To enhance security, we have a password method which ensures that the password can never be accessed; instead an error will be raised. We also have two foreign key fields, department_id and role_id, which refer to the ID’s of the department and role assigned to the employee.

      Note that we have an is_admin field which is set to False by default. We will override this when creating the admin user. Just after the Employee model, we have a user_loader callback, which Flask-Login uses to reload the user object from the user ID stored in the session.

      The Department and Role models are quite similar. Both have name and description fields. Additionally, both have a one-to-many relationship with the Employee model (one department or role can have many employees). We define this in both models using the employees field. backref allows us to create a new property on the Employee model such that we can use employee.department or employee.role to get the department or role assigned to that employee. lazy defines how the data will be loaded from the database; in this case it will be loaded dynamically, which is ideal for managing large collections.

      Migration

      Migrations allow us to manage changes we make to the models, and propagate these changes in the database. For example, if later on we make a change to a field in one of the models, all we will need to do is create and apply a migration, and the database will reflect the change.

      We’ll begin by installing Flask-Migrate, which will handle the database migrations using Alembic, a lightweight database migration tool. Alembic emits ALTER statements to a database thus implememting changes made to the models. It also auto-generates minimalistic migration scripts, which may be complex to write.

      $ pip install flask-migrate
      

      We’ll need to edit the app/__init__.py file:

      # app/__init__.py
      
      # after existing third-party imports
      from flask_migrate import Migrate
      
      # existing code remains
      
      
      def create_app(config_name):
          # existing code remains
      
          migrate = Migrate(app, db)
      
          from app import models
      
          return app
      

      We have created a migrate object which will allow us to run migrations using Flask-Migrate. We have also imported the models from the app package. Next, we’ll run the following command to create a migration repository:

      $ flask db init
      

      This creates a migrations directory in the dream-team directory:

      └── migrations
          ├── README
          ├── alembic.ini
          ├── env.py
          ├── script.py.mako
          └── versions
      

      Next, we will create the first migration:

      $ flask db migrate
      

      Finally, we’ll apply the migration:

      $ flask db upgrade
      

      We’ve sucessfully created tables based on the models we wrote! Let’s check the MySQL database to confirm this:

      $ mysql -u root
      
      mysql> use dreamteam_db;
      
      mysql> show tables;
      +------------------------+
      | Tables_in_dreamteam_db |
      +------------------------+
      | alembic_version        |
      | departments            |
      | employees              |
      | roles                  |
      +------------------------+
      4 rows in set (0.00 sec)
      

      Blueprints

      Blueprints are great for organising a flask app into components, each with its own views and forms. I find that blueprints make for a cleaner and more organised project structure because each blueprint is a distinct component that addresses a specific functionality of the app. Each blueprint can even have its own cutsom URL prefix or subdomain. Blueprints are particularly convenient for large applications.

      We’re going to have three blueprints in this app:

      1. Home – this will have the homepage and dashboard views
      2. Admin – this will have all administrator (department and role) forms and views
      3. Auth – this will have all authentication (registration and login) forms and views

      Create the relevant files and directories so that your directory structure resembles this:

      └── dream-team
          ├── app
          │   ├── __init__.py
          │   ├── admin
          │   │   ├── __init__.py
          │   │   ├── forms.py
          │   │   └── views.py
          │   ├── auth
          │   │   ├── __init__.py
          │   │   ├── forms.py
          │   │   └── views.py
          │   ├── home
          │   │   ├── __init__.py
          │   │   └── views.py
          │   ├── models.py
          │   ├── static
          │   └── templates
          ├── config.py
          ├── instance
          │   └── config.py
          ├── migrations
          │   ├── README
          │   ├── alembic.ini
          │   ├── env.py
          │   ├── script.py.mako
          │   └── versions
          │       └── a1a1d8b30202_.py
          ├── requirements.txt
          └── run.py
      

      I chose not to have static and templates directories for each blueprint, because all the application templates will inherit from the same base template and use the same CSS file. Instead, the templates directory will have sub-directories for each blueprint so that blueprint templates can be grouped together.

      In each blueprint’s __init__.py file, we need to create a Blueprint object and initialize it with a name. We also need to import the views.

      # app/admin/__init__.py
      
      from flask import Blueprint
      
      admin = Blueprint('admin', __name__)
      
      from . import views
      
      # app/auth/__init__.py
      
      from flask import Blueprint
      
      auth = Blueprint('auth', __name__)
      
      from . import views
      
      # app/home/__init__.py
      
      from flask import Blueprint
      
      home = Blueprint('home', __name__)
      
      from . import views
      

      Then, we can register the blueprints on the app in the app/__init__.py file, like so:

      # app/__init__.py
      
      # existing code remains
      
      
      def create_app(config_name):
          # existing code remains
      
          from app import models
      
          from .admin import admin as admin_blueprint
          app.register_blueprint(admin_blueprint, url_prefix='/admin')
      
          from .auth import auth as auth_blueprint
          app.register_blueprint(auth_blueprint)
      
          from .home import home as home_blueprint
          app.register_blueprint(home_blueprint)
      
          return app
      

      We have imported each blueprint object and registered it. For the admin blueprint, we have added a url prefix, /admin. This means that all the views for this blueprint will be accessed in the browser with the url prefix admin.

      Home Blueprint

      Time to work on fleshing out the blueprints! We’ll start with the home blueprint, which will have the homepage as well as the dashboard.

      # app/home/views.py
      
      from flask import render_template
      from flask_login import login_required
      
      from . import home
      
      
      @home.route('/')
      def homepage():
          """
          Render the homepage template on the / route
          """
          return render_template('home/index.html', title="Welcome")
      
      
      @home.route('/dashboard')
      @login_required
      def dashboard():
          """
          Render the dashboard template on the /dashboard route
          """
          return render_template('home/dashboard.html', title="Dashboard")
      

      Each view function has a decorator, home.route, which has a URL route as a parameter (remember that home is the name of the blueprint as specified in the app/home/__init__.py file). Each view handles requests to the specified URL.

      The homepage view renders the home template, while the dashboard view renders the dashboard template. Note that the dashboard view has a login_required decorator, meaning that users must be logged in to access it.

      Now to work on the base template, which all other templates will inherit from. Create a base.html file in the app/templates directory and add the following code:

      <!-- app/templates/base.html -->
      
      <!DOCTYPE html>
      <html lang="en">
      <head>
          <title>{{ title }} | Project Dream Team</title>
          <link href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/css/bootstrap.min.css" rel="stylesheet">
          <link href="https://www.digitalocean.com/community/tutorials/{{ url_for("static', filename="css/style.css") }}" rel="stylesheet">
          <link rel="shortcut icon" href="https://www.digitalocean.com/community/tutorials/{{ url_for("static', filename="img/favicon.ico") }}">
      </head>
      <body>
          <nav class="navbar navbar-default navbar-fixed-top topnav" role="navigation">
              <div class="container topnav">
                <div class="navbar-header">
                    <button type="button" class="navbar-toggle" data-toggle="collapse" data-target="#bs-example-navbar-collapse-1">
                        <span class="sr-only">Toggle navigation</span>
                        <span class="icon-bar"></span>
                        <span class="icon-bar"></span>
                        <span class="icon-bar"></span>
                    </button>
                    <a class="navbar-brand topnav" href="https://www.digitalocean.com/community/tutorials/{{ url_for("home.homepage') }}">Project Dream Team</a>
                </div>
                <div class="collapse navbar-collapse" id="bs-example-navbar-collapse-1">
                    <ul class="nav navbar-nav navbar-right">
                        <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("home.homepage') }}">Home</a></li>
                        <li><a href="#">Register</a></li>
                        <li><a href="#">Login</a></li>
                    </ul>
                </div>
              </div>
          </nav>
          <div class="wrapper">
            {% block body %}
            {% endblock %}
            <div class="push"></div>
          </div>
          <footer>
              <div class="container">
                  <div class="row">
                      <div class="col-lg-12">
                          <ul class="list-inline">
                              <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("home.homepage') }}">Home</a></li>
                              <li class="footer-menu-divider">⋅</li>
                              <li><a href="#">Register</a></li>
                              <li class="footer-menu-divider">⋅</li>
                              <li><a href="#">Login</a></li>
                          </ul>
                          <p class="copyright text-muted small">Copyright © 2016. All Rights Reserved</p>
                      </div>
                  </div>
              </div>
          </footer>
      </body>
      </html>
      

      Note that we use # for the Register and Login links. We will update this when we are working on the auth blueprint.

      Next, create a home directory inside the app/templates directory. The homepage template, index.html, will go inside it:

      <!-- app/templates/home/index.html -->
      
      {% extends "base.html" %}
      {% block title %}Home{% endblock %}
      {% block body %}
      <div class="intro-header">
          <div class="container">
              <div class="row">
                  <div class="col-lg-12">
                      <div class="intro-message">
                          <h1>Project Dream Team</h1>
                          <h3>The best company in the world!</h3>
                          <hr class="intro-divider">
                          </ul>
                      </div>
                  </div>
              </div>
          </div>
      </div>
      {% endblock %}
      

      Inside the static directory, add css and img directories. Add the following CSS file, style.css, to your static/css directory (note that you will need a background image, intro-bg.jpg, as well as a favicon in your static/img directory):

      /* app/static/css/style.css */
      
      body, html {
          width: 100%;
          height: 100%;
      }
      
      body, h1, h2, h3 {
          font-family: "Lato", "Helvetica Neue", Helvetica, Arial, sans-serif;
          font-weight: 700;
      }
      
      a, .navbar-default .navbar-brand, .navbar-default .navbar-nav>li>a {
        color: #aec251;
      }
      
      a:hover, .navbar-default .navbar-brand:hover, .navbar-default .navbar-nav>li>a:hover {
        color: #687430;
      }
      
      footer {
          padding: 50px 0;
          background-color: #f8f8f8;
      }
      
      p.copyright {
          margin: 15px 0 0;
      }
      
      .alert-info {
          width: 50%;
          margin: auto;
          color: #687430;
          background-color: #e6ecca;
          border-color: #aec251;
      }
      
      .btn-default {
          border-color: #aec251;
          color: #aec251;
      }
      
      .btn-default:hover {
          background-color: #aec251;
      }
      
      .center {
          margin: auto;
          width: 50%;
          padding: 10px;
      }
      
      .content-section {
          padding: 50px 0;
          border-top: 1px solid #e7e7e7;
      }
      
      .footer, .push {
        clear: both;
        height: 4em;
      }
      
      .intro-divider {
          width: 400px;
          border-top: 1px solid #f8f8f8;
          border-bottom: 1px solid rgba(0,0,0,0.2);
      }
      
      .intro-header {
          padding-top: 50px;
          padding-bottom: 50px;
          text-align: center;
          color: #f8f8f8;
          background: url(../img/intro-bg.jpg) no-repeat center center;
          background-size: cover;
          height: 100%;
      }
      
      .intro-message {
          position: relative;
          padding-top: 20%;
          padding-bottom: 20%;
      }
      
      .intro-message > h1 {
          margin: 0;
          text-shadow: 2px 2px 3px rgba(0,0,0,0.6);
          font-size: 5em;
      }
      
      .intro-message > h3 {
          text-shadow: 2px 2px 3px rgba(0,0,0,0.6);
      }
      
      .lead {
          font-size: 18px;
          font-weight: 400;
      }
      
      .topnav {
          font-size: 14px;
      }
      
      .wrapper {
        min-height: 100%;
        height: auto !important;
        height: 100%;
        margin: 0 auto -4em;
      }
      

      Run the app; you should be able to see the homepage now.

      Auth Blueprint

      For the auth blueprint, we’ll begin by creating the registration and login forms. We’ll use Flask-WTF, which will allow us to create forms that are secure (thanks to CSRF protection and reCAPTCHA support).

      pip install Flask-WTF
      

      Now to write the code for the forms:

      # app/auth/forms.py
      
      from flask_wtf import FlaskForm
      from wtforms import PasswordField, StringField, SubmitField, ValidationError
      from wtforms.validators import DataRequired, Email, EqualTo
      
      from ..models import Employee
      
      
      class RegistrationForm(FlaskForm):
          """
          Form for users to create new account
          """
          email = StringField('Email', validators=[DataRequired(), Email()])
          username = StringField('Username', validators=[DataRequired()])
          first_name = StringField('First Name', validators=[DataRequired()])
          last_name = StringField('Last Name', validators=[DataRequired()])
          password = PasswordField('Password', validators=[
                                              DataRequired(),
                                              EqualTo('confirm_password')
                                              ])
          confirm_password = PasswordField('Confirm Password')
          submit = SubmitField('Register')
      
          def validate_email(self, field):
              if Employee.query.filter_by(email=field.data).first():
                  raise ValidationError('Email is already in use.')
      
          def validate_username(self, field):
              if Employee.query.filter_by(username=field.data).first():
                  raise ValidationError('Username is already in use.')
      
      
      class LoginForm(FlaskForm):
          """
          Form for users to login
          """
          email = StringField('Email', validators=[DataRequired(), Email()])
          password = PasswordField('Password', validators=[DataRequired()])
          submit = SubmitField('Login')
      

      Flask-WTF has a number of validators that make writing forms much easier. All the fields in the models have the DataRequired validator, which means that users will be required to fill all of them in order to register or login.

      For the registration form, we require users to fill in their email address, username, first name, last name, and their password twice. We use the Email validator to ensure valid email formats are used (e.g [email protected].) We use the EqualTo validator to confirm that the password and confirm_password fields in the RegistrationForm match. We also create methods (validate_email and validate_username) to ensure that the email and username entered have not been used before.

      The submit field in both forms will be represented as a button that users will be able to click to register and login respectively.

      With the forms in place, we can write the views:

      # app/auth/views.py
      
      from flask import flash, redirect, render_template, url_for
      from flask_login import login_required, login_user, logout_user
      
      from . import auth
      from forms import LoginForm, RegistrationForm
      from .. import db
      from ..models import Employee
      
      
      @auth.route('/register', methods=['GET', 'POST'])
      def register():
          """
          Handle requests to the /register route
          Add an employee to the database through the registration form
          """
          form = RegistrationForm()
          if form.validate_on_submit():
              employee = Employee(email=form.email.data,
                                  username=form.username.data,
                                  first_name=form.first_name.data,
                                  last_name=form.last_name.data,
                                  password=form.password.data)
      
              # add employee to the database
              db.session.add(employee)
              db.session.commit()
              flash('You have successfully registered! You may now login.')
      
              # redirect to the login page
              return redirect(url_for('auth.login'))
      
          # load registration template
          return render_template('auth/register.html', form=form, title="Register")
      
      
      @auth.route('/login', methods=['GET', 'POST'])
      def login():
          """
          Handle requests to the /login route
          Log an employee in through the login form
          """
          form = LoginForm()
          if form.validate_on_submit():
      
              # check whether employee exists in the database and whether
              # the password entered matches the password in the database
              employee = Employee.query.filter_by(email=form.email.data).first()
              if employee is not None and employee.verify_password(
                      form.password.data):
                  # log employee in
                  login_user(employee)
      
                  # redirect to the dashboard page after login
                  return redirect(url_for('home.dashboard'))
      
              # when login details are incorrect
              else:
                  flash('Invalid email or password.')
      
          # load login template
          return render_template('auth/login.html', form=form, title="Login")
      
      
      @auth.route('/logout')
      @login_required
      def logout():
          """
          Handle requests to the /logout route
          Log an employee out through the logout link
          """
          logout_user()
          flash('You have successfully been logged out.')
      
          # redirect to the login page
          return redirect(url_for('auth.login'))
      

      Just like in the home blueprint, each view here handles requests to the specified URL. The register view creates an instance of the Employee model class using the registration form data to populate the fields, and then adds it to the database. This esentially registers a new employee.

      The login view queries the database to check whether an employee exists with an email address that matches the email provided in the login form data. It then uses the verify_password method to check that the password in the database for the employee matches the password provided in the login form data. If both of these are true, it proceeds to log the user in using the login_user method provided by Flask-Login.

      The logout view has the login_required decorator, which means that a user must be logged in to access it. It calles the logout_user method provided by Flask-Login to log the user out.

      Note the use of flash method, which allows us to use Flask’s message flashing feature. This allows us to communicate feedback to the user, such as informing them of successful registration or unsuccessful login.

      Finally, let’s work on the templates. First, we’ll install Flask-Bootstrap so we can use its wtf and utils libraries. The wtf library will allow us to quickly generate forms in the templates based on the forms in the forms.py file. The utils library will allow us to display the flash messages we set earlier to give feedback to the user.

      pip install flask-bootstrap
      

      We need to edit the app/__init__.py file to use Flask-Bootstrap:

      # app/__init__.py
      
      # after existing third-party imports
      from flask_bootstrap import Bootstrap
      
      # existing code remains
      
      
      def create_app(config_name):
          # existing code remains
      
          Bootstrap(app)
      
          from app import models
      
          # blueprint registration remains here
      
          return app
      

      We’ve made quite a number of edits to the app/__init__.py file. This is the final version of the file and how it should look at this point (note that I have re-arranged the imports and variables in alphabetical order):

      # app/__init__.py
      
      # third-party imports
      from flask import Flask
      from flask_bootstrap import Bootstrap
      from flask_login import LoginManager
      from flask_migrate import Migrate
      from flask_sqlalchemy import SQLAlchemy
      
      # local imports
      from config import app_config
      
      db = SQLAlchemy()
      login_manager = LoginManager()
      
      
      def create_app(config_name):
          app = Flask(__name__, instance_relative_config=True)
          app.config.from_object(app_config[config_name])
          app.config.from_pyfile('config.py')
      
          Bootstrap(app)
          db.init_app(app)
          login_manager.init_app(app)
          login_manager.login_message = "You must be logged in to access this page."
          login_manager.login_view = "auth.login"
          migrate = Migrate(app, db)
      
          from app import models
      
          from .admin import admin as admin_blueprint
          app.register_blueprint(admin_blueprint, url_prefix='/admin')
      
          from .auth import auth as auth_blueprint
          app.register_blueprint(auth_blueprint)
      
          from .home import home as home_blueprint
          app.register_blueprint(home_blueprint)
      
          return app
      

      We need two templates for the auth blueprint: register.html and login.html, which we’ll create in an auth directory inside the templates directory.

      <!-- app/templates/auth/register.html -->
      
      {% import "bootstrap/wtf.html" as wtf %}
      {% extends "base.html" %}
      {% block title %}Register{% endblock %}
      {% block body %}
      <div class="content-section">
        <div class="center">
          <h1>Register for an account</h1>
          <br/>
          {{ wtf.quick_form(form) }}
        </div>
      </div>
      {% endblock %}
      
      <!-- app/templates/auth/login.html -->
      
      {% import "bootstrap/utils.html" as utils %}
      {% import "bootstrap/wtf.html" as wtf %}
      {% extends "base.html" %}
      {% block title %}Login{% endblock %}
      {% block body %}
      <div class="content-section">
        <br/>
        {{ utils.flashed_messages() }}
        <br/>
        <div class="center">
          <h1>Login to your account</h1>
          <br/>
          {{ wtf.quick_form(form) }}
        </div>
      </div>
      {% endblock %}
      

      The forms are loaded from the app/auth/views.py file, where we specified which template files to display for each view. Remember the Register and Login links in the base template? Let’s update them now so we can access the pages from the menus:

      <!-- app/templates/base.html -->
      
      <!-- Modify nav bar menu -->
      <ul class="nav navbar-nav navbar-right">
          <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("home.homepage') }}">Home</a></li>
          <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.register') }}">Register</a></li>
          <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.login') }}">Login</a></li>
      </ul>
      
      <!-- Modify footer menu -->
      <ul class="list-inline">
          <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("home.homepage') }}">Home</a></li>
          <li class="footer-menu-divider">⋅</li>
          <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.register') }}">Register</a></li>
          <li class="footer-menu-divider">⋅</li>
          <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.login') }}">Login</a></li>
      </ul>
      

      Run the app again and click on the Register and Login menu links. You should see the templates loaded with the appropriate form.

      Try to fill out the registration form; you should be able to register a new employee. After registration, you should be redirected to the login page, where you will see the flash message we configured in the app/auth/views.py file, inviting you to login.

      Logging in should be successful; however you should get a Template Not Found error after logging in, because the dashboard.html template has not been created yet. Let’s do that now:

      <!-- app/templates/home/dashboard.html -->
      
      {% extends "base.html" %}
      {% block title %}Dashboard{% endblock %}
      {% block body %}
      <div class="intro-header">
          <div class="container">
              <div class="row">
                  <div class="col-lg-12">
                      <div class="intro-message">
                          <h1>The Dashboard</h1>
                          <h3>We made it here!</h3>
                          <hr class="intro-divider">
                          </ul>
                      </div>
                  </div>
              </div>
          </div>
      </div>
      {% endblock %}
      

      Refresh the page. You’ll notice that the navigation menu still has the register and login links, even though we are already logged in. We’ll need to modify it to show a logout link when a user is already authenticated. We will also include a Hi, username! message in the nav bar:

      <!-- app/templates/base.html -->
      
      <!-- In the head tag, include link to Font Awesome CSS so we can use icons -->
      <link href="https://maxcdn.bootstrapcdn.com/font-awesome/4.7.0/css/font-awesome.min.css" rel="stylesheet">
      
      <!-- Modify nav bar menu -->
      <ul class="nav navbar-nav navbar-right">
          {% if current_user.is_authenticated %}
            <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("home.dashboard') }}">Dashboard</a></li>
            <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.logout') }}">Logout</a></li>
            <li><a><i class="fa fa-user"></i>  Hi, {{ current_user.username }}!</a></li>
          {% else %}
            <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("home.homepage') }}">Home</a></li>
            <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.register') }}">Register</a></li>
            <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.login') }}">Login</a></li>
          {% endif %}
      </ul>
      
      <!-- Modify footer menu -->
      <ul class="list-inline">
          <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("home.homepage') }}">Home</a></li>
          <li class="footer-menu-divider">⋅</li>
          {% if current_user.is_authenticated %}
            <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.logout') }}">Logout</a></li>
          {% else %}
            <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.register') }}">Register</a></li>
            <li class="footer-menu-divider">⋅</li>
            <li><a href="https://www.digitalocean.com/community/tutorials/{{ url_for("auth.login') }}">Login</a></li>
          {% endif %}
      </ul>
      

      Note how we use if-else statements in the templates. Also, take note of the current_user proxy provided by Flask-Login, which allows us to check whether the user is authenticated and to get the user’s username.

      Logging out will take you back to the login page:

      Attempting to access the dashboard page without logging in will redirect you to the login page and display the message we set in the app/__init__.py file:

      Notice that the URL is configured such that once you log in, you will be redirected to the page you initially attempted to access, which in this case is the dashboard.

      Conclusion

      That’s it for Part One! We’ve covered quite a lot: setting up a MySQL database, creating models, migrating the database, and handling registration, login, and logout. Good job for making it this far!

      Watch this space for Part Two, which will cover the CRUD functionality of the app, allowing admin users to add, list, edit, and delete departments and roles, as well as assign them to employees.



      Source link