Identity Server: Using Entity Framework Core for Configuration Data

This post is a continuation of a series of posts that follow my initial looking into using IdentityServer4 in ASP.NET Core with an API and an Angular front end. The following are the related posts.

Identity Server: Introduction
Identity Server: Sample Exploration and Initial Project Setup
Identity Server: Interactive Login using MVC
Identity Server: From Implicit to Hybrid Flow
Identity Server: Using ASP.NET Core Identity
Identity Server: Using Entity Framework Core for Configuration Data (this post)
Identity Server: Usage from Angular

This post is going to take the existing solution this series has been using and switch from using hard coded configuration data, found in the  Config class of the Identity Application and moving it to a database using Entity Framework Core. As with prior entries, this will be following the intent of one of the official quick starts for Using Entity Framework Core for configuration data. This post is fairly different just because our example project already uses entity framework so a lot of steps can be skipped. The starting point of the code can be found here. All the changes in this post will be taking place in the Identity Application.

Identity Application

Thankfully the creators of IdentityServer provide a NuGet package that includes all the bits needed to move configuration data and operational data to Entity Framework Core. Start by added the following NuGet package.

  • IdentityServer4.EntityFramework
Startup

With the above NuGet package installed the  ConfigureServices function of the  Startup class needs to be changed to tell IdentityServer the new place to pull data from. The following is the new version of the  AddIdentityServer call updated to use Entity Framework Core.

Notice that the following have all been replaced by  AddConfigurationStore and  AddOperationalStore.

  • AddInMemoryPersistedGrants
  • AddInMemoryIdentityResources
  • AddInMemoryApiResources
  • AddInMemoryClients

The other thing of note is the  migrationsAssembly and its usage via  options.MigrationsAssembly. This is moving the management of the migrations from the assembly that the contexts are defined to the Identity Application. This is needed in this case since the two contexts in question are defined in a NuGet package.

Migrations

Now that the configuration is done for the new contexts migrations need to be added to them. As always there are two ways to handle this either via the Package Manager Console or from a command prompt. I am going to use the command prompt this round to match the IdentityServer docs. Run the following two commands from the same path as the Identity Application’s csproj file.

This is the first time I have used the  -o argument which controls where the migration is output and following the docs example I am putting the migrations that are for entities outside of the control of the application into a subdirectory. Speaking of the entities being outside of the control of the main application, this means anytime the NuGet package that contains the entity is updated a check will need to be made to see if new migrations are needed.

Database Migrations and Seed Data

Since the  DbContext classes that need migrations run are outside of the control our application if automatic migrations must be handled in a different way than with the identity-related context used previously in this series. Following the official docs, I am going to create an  InitializeDatabase function that will apply any needed migrations as well as add seed data. To do this I am adding a new IdentityServerDatabaseInitialization class in the Data/IdentityServer directory. The following is the full class.

The  InitializeDatabase takes an  IApplicationBuilder in order to be able to control the lifetime of the two  DbContext classes needed. Normally this wouldn’t be needed and the lifetime would be controlled automatically, but since this code is being called from the  Startup class instead of during a request (which is how the DI system does auto scoping) the scope is being created by the  app.ApplicationServices.GetService<IServiceScopeFactory>().CreateScope() call.

The  PerformMigrations function pulls the two  DbContext objects from the container and applies migrations. Finally in  SeedData if the  DbSets don’t already contain data then the seed data is pulled from the  Config class and saved to the database.

Back to Startup

In the  Configure function of the  Startup class add the following call to make sure migrations and seed data are run when the application starts.

Wrapping up

With the above changes, the Identity Application is now using the database for all its persistence. The missing bits are of course UI to manage the related data, but those can be built out as needed. The code in its completed state can be found here.

The next steps for this project will be utilizing IdentityServer from Angular in the Client Application instead of the temporary  IdentityController that has had to be used in all the examples so far.

Identity Server: Using ASP.NET Core Identity

This post is a continuation of a series of posts that follow my initial looking into using IdentityServer4 in ASP.NET Core with an API and an Angular front end. The following are the related posts.

Identity Server: Introduction
Identity Server: Sample Exploration and Initial Project Setup
Identity Server: Interactive Login using MVC
Identity Server: From Implicit to Hybrid Flow
Identity Server: Using ASP.NET Core Identity (this post)
Identity Server: Using Entity Framework Core for Configuration Data
Identity Server: Usage from Angular

This post is going to cover using ASP.NET Core Identity instead of an in-memory user store like the previous examples. As I write this I am working through the Using ASP.NET Core Identity quick start from the docs. This isn’t going to differ a whole lot from the official docs, but I still want to document it to help solidify everything in my head. The starting point of the code for this post can be found here.

Identity Application

The Identity Application will be where the bulk of the changes happen. Since it is much easier to add IdentityServer to a project than it is to add ASP.NET Core Identity we are going to delete the existing Identity Application project and re-create it with Identity from the start. Right click on the IdentityApp project and click remove.

This removes the project from the solution, but the files also need to be deleted off of disk or use a different name. I chose to rename the old project folder on disk so I could still grab any files I might need.

Create a new Identity Application

Right-click on the solution and select Add > New Project.

On the Add New Project dialog under Visual C# > .NET Core select ASP.NET Core Web Application and enter the name of the project (IdentityApp in this example) and click OK.

On the next dialog select the Web Application template.

Next, click the Change Authentication button and select the Individual User Accounts option.

Click OK on the Change Authentication dialog and then click OK on the template dialog. After a few seconds, the solution will contain a new IdentityApp that is using ASP.NET Core Identity with Entity Framework Core.

Adding Identity Server to the Identity App Project

Using NuGet install the IdentityServer4.AspNetIdentity package which will also install  IdentityServer4 which the old project was using. Next, copy the  Config class from the old IdentityApp project and delete the  GetUsers function.

Startup Changes

In the  Startup class at the end of  ConfigureServices function add the following.

The only difference between this and the one used with the previous posts is instead of  AddTestUsers being used to pull a hard coded list of uses our of the  Config class users are pulled from the database using ASP.NET Core Identity using this  AddAspNetIdentity<ApplicationUser>() call. Identity Server is very flexible and this is only of the option for an identity store.

Next, in the  Configure function add  app.UseIdentityServer() after  app.UseIdentity().

Database

There are a couple of ways to make sure the database is created and migrated when changes happen. One is via the command line in the project directory using the following command.

The way I normally us when at this stage in development is to add code to the DB context to automatically apply migrations. The following is the full  ApplicationDbContext class modified to automatically run migrations when the context is constructed.

Now throw that all away

The above is good to go through to know how things work, but I got to this point and wanted the functionality to be on par with the previous entries. The docs made this sound simple, but it was not simple at all I had a list of at least 30 points of files to be moved and changes made to existing files. I am going to spare you all those details and recommend that you just pull the Identity Application from my GitHub repo for this project instead. I did basically the same thing out of the official samples repo to get things working as they should with contents, errors, and log out.

If you want the auto migrations then make sure and keep the version of the  ApplicationDbContext from above.

Client Application

No changes are actually required to the client application, but as in the official docs, I made changes to show hitting the API Application using both a user access token and client credentials. The following is the index action on the  IdentityController which has been changed to call two functions one for each type of API access.

The following is the function to access the API Application using a user access token.

Now the function to access the API Application using client credentials.

Finally,  Index.cshtml found in the Views/Identity directory has the following change.

Wrapping up

Now the Identity Application is using ASP.NET Core Identity with Entity Framework Core to store users in the database. The next post will cover moving the items now in the  Config class into the database. The completed version of the code can be found here.

Identity Server: From Implicit to Hybrid Flow

This post is a continuation of a series of posts that follow my initial looking into using IdentityServer4 in ASP.NET Core with an API and an Angular front end. The following are the related posts.

Identity Server: Introduction
Identity Server: Sample Exploration and Initial Project Setup
Identity Server: Interactive Login using MVC
Identity Server: From Implicit to Hybrid Flow (this post)
Identity Server: Using ASP.NET Core Identity
Identity Server: Using Entity Framework Core for Configuration Data
Identity Server: Usage from Angularsing MVC

This post is going to cover adding back in the API access that was lost in the last post by changing the MVC client to use a hybrid grant instead of an implicit grant. This post was written while working through Switching to Hybrid Flow and adding API Access back in the official docs.

Identity Application

The changes to the Identity Application are pretty simple and only involve tweaking the settings on the MVC client found in the  GetClients function of the  Config class. First, change the  AllowedGrantTypes from  Implicit to  HybridAndClientCredentials. Next, a client secret should be added.

This is, of course, a bad secret, but this is only an example. Next, add  "apiApp" to the  AllowedScopes and finally add  AllowOfflineAccess = true. The following is the full client code.

Most of the above are straight forward.  AllowedGrantTypes is what is moving to the hybrid flow which then needs a client secret to ensure everything is on the up and up. This client should be able to hit the API application so it is added to the allowed scopes.  AllowOfflineAccess is less clear to me. According to the docs, it allows the requesting refresh tokens for long-lived API access. This would take some more digging before production to ensure authorization isn’t too long lived.

Client Application

Changes to the client application were pretty minimal as well. First, in the  Configure function of the  Startup class, the  UseOpenIdConnectAuthentication call must pass a few more items. The following is the full set up.

ClientSecret should match what was set up for the client in the Identity Application. According to the docs setting  ResponseType to  code id_token means use a hybrid flow. This is another point that I would want to dig more on.  Scope is requesting access to the API Application and offline access which is the matching part to the offline access set up in the Identity Application.  GetClaimsFromUserInfoEndpoint tells the middleware to go to the user info endpoint to retrieve additional claims after getting an identity token.

Identity Controller

The  Index action in the  IdentityController ends up being much simpler than it was in the previous posts. The following is the full function.

In the new version, the token can be retrieved from the HTTP context instead of using the  DiscoveryClient and  TokenClient like the previous version of this code did. The general idea is the same in both which is to get a token, use the token as part of a request to the API application, and finally display the response in a view.

Identity View

The last set of changes is to the  Index.cshtml file in the View/Identity directory which is the view that goes with the Index action of the  IdentityController. The view displays the access token, refresh token, results of the API call, and the logged in user’s claims.

Wrapping up

Adding back API access was pretty easy and the new setup will make managing other resources pretty simple. The identity space is still pretty new to me but working through the IdentityServer quickstarts are helping get me up to a basic level of knowledge. The finished code for this post can be found here. Come back next week to convert this example to use ASP.NET Core Identity.

Identity Server: Interactive Login using MVC

This post is a continuation of a series of posts that follow my initial looking into using IdentityServer4 in ASP.NET Core with an API and an Angular front end. The following are the related posts.

Identity Server: Introduction
Identity Server: Sample Exploration and Initial Project Setup
Identity Server: Interactive Login using MVC (this post)
Identity Server: From Implicit to Hybrid Flow
Identity Server: Using ASP.NET Core Identity
Identity Server: Using Entity Framework Core for Configuration Data
Identity Server: Usage from Angular

As before the end goal will be having authorization happen from Angular, but in the short term, the Client Application is using MVC/Razor for testing and verifications. The code as it stood before this post can be found here. If you are following along with the official docs I wrote this post while working through the Adding User Authentication with OpenID Connect quickstart.

The main point of this post is to add a way for a user to enter their username and password and get access to a page that requires authorization using the OpenID Connect protocol.

Identity Application

To enable this scenario the Identity Application will need MVC added along with some UI that will be used to handle login, permissions, and log off. First, using NuGet install the following two packages.

  • Microsoft.AspNetCore.Mvc
  • Microsoft.AspNetCore.StaticFiles

Next, in the ConfigureServices of the Startup class MVC needs to be added as a service.

Then in the Configure function use static files and use MVC should be added after the use statement for IdentityServer.

UI Changes

For the type of flow being used in this sample, the Identity Application will be in control of the login, grant, log out, and related UI. This is not a small amount of thing to get set up properly. Thankfully the IdentityServer team provides a Quickstart UI for use with the in-memory items we are currently using. The files can be downloaded from the repo linked in the previous line or an easier way is to open a Powershell prompt in the same directory of the Identity application as the Startup.cs file and run the following command.

After the download the project will contain a Quickstart folder with the needed controllers, a Views with of course the needed views, and wwwroot will have all the related files that need to be served with the views.

Config Changes

The Config class needs to be changed to return some more in-memory information to make this new process work. The first is to add a new client for MVC to the GetClients function. The following is the full function, but it is the second Client is the new one.

Notice that for the OpenID Connect implicit flow there are URLs that are needed that so this flow knows how to call back into the client application. At this point, I haven’t dug into everything that is going on in the client. The ClientId, ClientName, and URLs related properties are pretty clear. I am not 100% on the AllowedGrantTypes and AllowedScopes, but at this point, I am not going to dive into on these two options.

Next, add a  GetIdentityResources function matching the following. This fall in the same category as the two properties above, we are using them without fully digging into them.

The last change to the Config class is to add a function to return the in-memory users.

Startup Changes

The last change in the Identity Application is to add the new in-memory items to the IdentityServer service in the ConfigureServices function. The following is the full function.

Client Application

In order to get the client application to play well with the changes in the Identity Application, a few changes need to be made. First, the following NuGet packages need to be installed.

  • Microsoft.AspNetCore.Authentication.Cookies
  • Microsoft.AspNetCore.Authentication.OpenIdConnect

Next, in the  Configure function of the  Startup class, the application’s middleware pipeline needs some changes. Add the following line to turn off the JWT claim type mapping. This must be done before calling  UseOpenIdConnectAuthentication.

Now add in the cookie authentication middleware.

The last change is to add OpenID Connect authentication to the pipeline placed after the cookies middleware.

Notice that the URL of the authority is the URL the Identity Application runs on as well as the client ID match the one we set up in the  GetClients function of the  Config class in the Identity Application.

Identity Controller

Now that the above is set up we can switch over to the  IdentityController and add the  Authorize attribute to the  Index function.

This means if a user hits the index action of this controller and isn’t logged in they will be presented with the login page and after login, they will be redirected back to the above index action. That whole process is handled by the OpenId Connect Authentication middleware. The first time I tested the flow and it just worked was magical.

 

The final set of changes for this post is going to be added a way to log out. In the  IdentityController add a  Logout function.

Identity View Changes

The last change is to add a logout button to the  Index.cshtml found in the Views/Identity directory. At the bottom of the page, the following was added to call the  Logout action.

Wrapping Up

I already liked the idea of IdentityServer before this post, but after playing with it with the changes above it is emphasized how nice it is. I am very happy I am going down this path instead of trying to work this all out on my own. Stay tuned as this exploration will continue in future posts.

The code in the finished state can be found here.

Update

Turns out there is a bug in the code that goes with this example. In the client application’s  IdentityConroller the call to get a token is using  clientApp instead of  mvc for the client ID when requesting a token. With that change, the call to the API Application will fail since the MVC client doesn’t have access to the API scope. Look for next weeks post where API access will be added to the MVC client.

Entity Framework Core with SQLite Scaffolding

This is the third in what is turning into a series of post about using SQLite with Entity Framework Core. This post is going to cover adding a migration, scaffolding a controller and related views, and a few things that are harder to do using SQLite. The following are the first two post.

Entity Framework Core with SQLite
Entity Framework Core Errors Using Add-Migration

Adding Model, DbContext, Controller, and Views

If you have any experience with Entity Framework Core or have read any of my past entries on the subject this section is going to repeat some of the same information, but I am including it so someone who is looking for a full example will have it.

Model

In the Models folder add a Contact class similar the following.

DbContext

In the Data folder add a ContactsDbContext that inherits from DbContext. The following is an example that auto applies migrations to a database, if you don’t need that functionality it can be dropped out.

Now that the application has a model and a related DbContext the following can be used to add a migration that will create a Contacts in the SQLite database. Run from the Package Manager console.

Add-Migration is a Powershell command to add a migration (surprise!), AddContacts is the name of the migration and -Context ContactsDbContext is an argument that lets the command know which  DbConext to use. The Context is only needed if your application has more than one DbContext.

Controller and Views

With the above complete Visual Studio provides some tooling that makes it very fast to create a controller with views for listing, adding, editing, and deleting items. To begin right-click on the Controllers folder and select Add > New Scaffolded Item.

Select the MVC Controller with views, using Entity Framework option and click Add.

On the next dialog use the drop downs to select a model class and a data context class. Then verify the controller name and click add.

When the process completes the following items will have been added to your project.

Add to nav bar

To add a link to the new section of the app to the nav bar open the _Layout.cshtml in the Views/Shared/ directory. The following is the section of the file that needs to be changed to add an item to the nav bar.

Specifically, the following line was added to provide access to the contact list page.

Wrapping up

With the above, the application will be runnable. The code for this post can be found here. The next post in this series will cover the limitations of migrations when using SQLite with Entity Framework Core.

 

Add Aurelia to an ASP.NET Core Project

In this post I am going to add a new project to the my existing ASP.NET Core Basics solution which can be found in this repository. The new project will be MVC 6  to which I will add in Aurelia. With both ASP.NET and Aurelia now being at RTM I thought this would be a good time to cover getting a new project setup.

Over time the ASP.NET Core Basics repo used in this post is going to be replacing my ASP.NET Core SPAs repo based on some feedback that having Aurelia and Angular 2 in the same project made it harder to see how each individual framework is setup.

Part 1 – Add Aurelia to an ASP.NET Core Project (This Post)
Part 2 – Aurelia with an ASP.NET Core API
Part 3 – Aurelia with an ASP.NET Core API: Modeling and Displaying Data in Aurelia
Part 4 – Aurelia with ASP.NET Core: Host Aurelia from a Controller
Github repo with the code for all of the parts with a release to go with each post

 Adding a new project to an existing solution

To add a new project to the existing solution right click on the solution and then click Add > New Project.

AddProjectExistingSolution

On the Add New Project dialog select ASP.NET Core Web Application (.NET Core), enter a name and then click OK.

AddNewProjectDialog

On the New ASP.NET Core Web Application (.NET Core) dialog select Web Application. This application doesn’t need authentication so leave it set to No Authentication. Finally click OK.

NewASPNetCoreWebApplication

After a few seconds the project creation will complete and the solution will contain two projects. The existing Contacts project that contains both a razor/normal implementation of a contacts list as well as an API implementation. The second project is the newly created Aurelia project.

SolutionWithTwoProjects

Changing the startup project

Notice in the screenshot above that the Contacts project is in bold. This means that the Contacts project is set as the startup project and it will be the project that starts when the application is run (F5 or Cntrl + F5). In this post we will just be working with the Aurelia project so we need to make it the startup project. To do this right click the Aurelia project and select Set as StartUp Project.

SetAsStartupProject

Now if you hit F5 the Aurelia project will run. Visual Studio provides a lot of flexibility around which projects start up. You can select a single, have which ever project you have to have select, or even multiple projects.  In a later post we will need both projects to start up and I will cover that when we have the need.

Install the Aurelia CLI

Make sure you have a minimum of NodeJs 4.x or above installed. If you need the installer it can be found here. After the install is complete open a command prompt and run the following command to install the Aurelia CLI.

Add Aurelia to existing ASP.NET Core Project

In a command prompt navigate to the folder that contains the xproj file for the ASP.NET Core project created above. Now the Aurelia CLI can be used to setup a new Aurelia project at the current location using the following command.

There will be a series of prompts the first of which is the selection of which platform to use. Select the option for ASP.NET Core (option 2). I used the defaults for most of the remaining prompts. The exception was for unit testing which I selected no on just to keep the project simpler not because I think testing is a bad idea.

When the Aurelia CLI finishes its file creation and dependency restore your project will contain the highlighted new files and folders.

auaddedfiels

Notice that I have a warning on Dependencies that something is not installed. There is a quirky issue with Visual Studio that Scott Hanselman has blogged about here. He goes in to a good bit of detail about what is going on as well as suggesting a work around. It has to do with npm and not being about to restore an optional package that isn’t meant for Windows machines.

gulp

The Aurelia CLI creates a set of tasks to help with building, transpiling the Aurelia part of the applications. I wrote a couple of posts over the couple few weeks dealing with converting a project to use glup as well as how to get gulp working with ES 2015.

I am going to cover the abbreviated version of those two post here. Add a new file called  gulpfile.babel.js in the root of the project, where your  project.json is located. The Aurelia CLI added all the needed items in the  devDependencies section of  package.json.

gulp no go

At this point I attempted to include the tasks under  aurelia_project/tasks using  require('require-dir')('aurelia_project/tasks');. This failed completely. I couldn’t get any of the items in the tasks folder to show up. I am not sure why this didn’t work. My best guess is that the tasks in the tasks folder are exporting  gulp.series and not  gulp.task. I just don’t know enough about gulp at this point to now how to fix/work around this or if what I am trying to do is just not the right way it should be used.

The gulp work around

I spent more time that I would have like working on getting gulp to pick up the items in the tasks folder, but I don’t want to have to run a CLI command every time I do a build to make sure all the Aurelia related files are up to date. As a work around I decided to add a gulp task to invoke the CLI command for me.

To start open  package.json  and add the following to the  devDependencies section which allows shell commands to be run from gulp.

Next in  gulpfile.babel.js added the proper imports and created tasks for the CLI commands I wanted to run. In the case I am just showing the build command.

Using the Task Runner Explorer this task can now be set to run after a build of the MVC project.

treafterbuild

This accomplishes what I wanted, but it feels like a hack. If anyone knows a better way please let me know.

It’s Alive!

At this point if you run the application it will go to the normal default home page that gets created by the Visual Studio template. For me that address is  http://localhost:37472/. From there if you add index.html, the full address is  http://localhost:37472/index.html, you will be invoking the Aurelia application.

At this point all you will see is “Hello World!”. Not that impressive I know, but it is a starting point that we will build on in future posts.

The associated code can be found here.

ASP.NET Core Basics: Project Creation

With ASP.NET Core released it seems like a good time to do a series of posts on the basics of this new platform starting with getting a new project up and running.  In this post I am going to walk through installation of ASP.NET Core and then move to project creation. This project will end up being a basic contact list application although there will not be much specific to that end goal in this post.

Installation

All the software needed for this post can be found at http://dot.net. I will be using Visual Studio 2015 in my examples and if you don’t have it installed already it can be downloaded using the Download Visual Studio 2015 button on the right of the above page or by clicking here. If you already have Visual Studio 2015 installed please ensure you have installed Update 3.

The next bit of software you will need is found by clicking the Download .NET Core 1.0 or clicking here. This page has a good write up getting started with .NET Core, but all you need is the install for .NET Core 1.0 for Visual Studio which can be found here.

To verify that .NET Core is installed open a command prompt and run  dotnet --version which will print the current version of .NET Core you have installed. As of this writing this my version is 1.0.0-preview2-003121. While .NET Core its self has been officially released the tooling is still in preview which is why the version contains preview2-003121.

Project Creation

Launch Visual Studio and from the File > New menu select Project…

NewProject

This will load the New Project dialog. On the left side under Templates > Visual C# > .NET Core select ASP.NET Core Web Application (.Net Core) or you can use the search box in the upper right of the screen to search for “ASP.NET Core Web Application (.Net Core)”.

NewProjectDialogContacts

Next on the New ASP.NET Core Web Application dialog select Web Application. This option creates an application with example MVC Views (razor) and an example controller. Notice the note that this template can be used for RESTful HTTP services as well. Next click the Change Authentication button.

NewAspNetCoreWebApplication

For this example I am going to use Individual User Accounts. This option adds UI, models, controllers, etc. to allow registration and management of user accounts. Since the option needs a database to store account information it include Entity Framework Core. Click OK on the Change Authentication dialog and then click ON new web application dialog.

ChangeAuthentication

Project Overview

After the creation process is finished got to the Solution Explorer window and you will see a set up similar to the following.

SolutionExplorer

I am going to point out a few of the files and folder that are part of a newly created application. Fist is  wwwroot which is where static files will be severed as long you are using the static files middleware.  This is where images, CSS and JavaScript should go.

appsettings.json is where you will find connection strings and logging settings by default. Your own settings can be added here as well.

project.json is where you will find all your project’s dependencies (using the NuGet UI or Package Manager console both write to this file), tools, frameworks, build options, runtime options, publish options and scripts are all defined. Thankfully this file has good support for intellisense if you decide to edit it manually. This file covers a lot, but for this introduction I am going to avoid digging into the specifics.

Startup.cs is the last file I want to call out. Its generated contents are fine for this example project, but it has a couple of functions you should be aware of. First ASP.NET Core comes with dependency injection built in and the  ConfigureServices function is where items are registered with the built in container. The second function is  Configure and this is the function where the HTTP request pipeline for your application is configured using various middleware.

Wrapping Up

At this point you have a web application that can be run (press F5 to run with the debugger attached or Ctrl + F5 to run without the debugger). With no code changes you now have a web application that has basic navigation, controllers, views and authentication.

With this series of post things are being kept intentionally short and focused on one or two main topics. Next week will build on this basic project by adding in the ability to manage contacts, using a new controller and associated razor view which will be persisted to a database using Entity Framework Core.

The code that goes with this post can be found in this GitHub repository.

 

Enable Scaffolding without Entity Framework in ASP.NET Core

Background

While working on a MVC 6 application backed by a web API application which resides in a separate project I found myself wanting a quick way to use a model as a base to generate a controller and associated CRUD razor views. If a project is using entity framework this is simple to accomplish, but since my project is using web API calls instead of entity framework scaffolding is unavailable. Instead of manually creating all the needed parts I decided to try and enable the scaffolding bits that are available in projects using entity framework.

Getting Started

The first step I took was to look at the  project.json from an application using entity framework. This helped me identity the dependencies and tools that might be needed for the scaffolding process. The following are the new items from the dependencies section.

And the new items from the tools section (spoiler Microsoft.VisualStudio.Web.CodeGenerators.Mvc isn’t actually needed).

Failed dotnet restore

With the above changes  dotnet restore fails with this error:

Package Microsoft.Composition 1.0.27 is not compatible with netcoreapp1.0 (.NETCoreApp,Version=v1.0). Package Microsoft.Composition 1.0.27 supports: portable-net45+win8+wp8+wpa81 (.NETPortable,Version=v0.0,Profile=Profile259)
One or more packages are incompatible with .NETCoreApp,Version=v1.0.

I double checked that the entity framework project was using the same versions. It was indeed using the same versions, but its restore worked fine.

The fix

This is a known issue which can be found on github here. The issues is caused by  Microsoft.VisualStudio.Web.CodeGenerators.Mvc in the tools sections which I finally found out is not actually needed to enable scaffolding. To be clear the only change needed in the  tools section the addition of the following and Microsoft.VisualStudio.Web.CodeGenerators.Mvc is not needed.

Scaffolding

Right click the Controllers folder and select Add > Controllers… (this is the menu enabled by the changes above).

AddController

The above launches the Add Scaffold dialog. Here select “MVC Controller with views, using Entity Framework”. This is the only built in option that creates views. Since this project isn’t using entity framework it will require a little clean up, but it is still faster than creating the views manually.

AddScaffold

After clicking add you will see the following screen. Select the model class to be used during the scaffold. For the data context class click the plus (+) button and enter a name (don’t spend any time on the name the resulting file will be deleted in the next step. Finally enter a controller name and click Add.

AddScaffoldController

Cleaning up the scaffold

In the models folder delete the data context class that was created, TestContext in this case. The removal of the TestContext will cause errors in the TestController which can all be removed (the constructor and private field used to hold the context) and replaced with calls to whatever back end system you happen to be using in your application. Finally in the Startup class remove the last reference to the TestContext.

Wrapping up

That is all the changes that were required other than adding a reference to the test index page to the home page of the application. Not sure this is the most efficient method ever, but for me it is easyer than creating the razor pages myself manually. In the future I may do some digging and determine what is required to create my own scaffolding option.

Emails using Mailgun in ASP.NET Core

Updated version of this post can be found here.

At last month’s Nashville .Net Users Group meeting Michael McCann when over some of the aspects of ASP.NET’s membership provider (non-core version). One of the things he talked about was enabling email as part of the user sign up process and for use in password recovery. This post is going to cover the same emailing aspect but in ASP.NET core using mailgun to actually send emails.

Account Controller

In the account controller most of the code needed is already present and just needs to be uncommented. In the Register function uncomment the following which will send the user an email asking that the address be confirmed. This of course stops users from signing up with email addresses they don’t actually have access to.

And then comment out this next line which would sign the user in before they have used the email above to confirm their account.

Next in the Login function add the following bit of code just before the call to _signInManager.PasswordSignInAsync. This looks up the user by email address and returns an error if the account has not been confirmed.

The last change is in the ForgotPassword function. Uncomment the following code to send the user an email to reset their password.

Forgot Password View

In ForgotPassword.cshtml uncomment the following section to show the UI associated with email based password reset.

Caution for existing sites

With the changes above if a user has not confirmed their email address then they will not be able to log in or reset their password. Any existing users would need to have their accounts marked as confirmed manually by updating the EmailConfirmed bit field in the AspNetUsers table or be provided away to confirm their account.

Mailgun

Mailgun is an email service that provides a simple API for sending emails and allows up to 10,000 emails to be sent free every month. I have only used mailgun for sending test emails so I can’t speak to how it holds up at scale.

After signing up for an account click on the domains tab and select the only existing active domain which should start with something like sandbox.

Storing Configuration

In my project I created an EmailSettings class that will be loaded from user secrets in the start up of the application. For more details on general configuration in ASP.NET Core check out this post and thenthis post for more details on user secrets. The following is my email settings class.

If using mailgun the above fields map to the following from the mailgun domain page.

EmailSettings Mailgun Example
ApiKey API Key key-*
BaseUri API Base URL https://api.mailgun.net/v3/
RequestUri API Base URL sandbox*.mailgun.org
From Default SMTP Login [email protected]*.mailgun.org

A couple of notes to the above table on what I actually saved in my config files.

EmailSettings Field Note Example
ApiKey Used with basic auth and needs username api:key-*
RequestUri Needs the API end point to call sandbox*.mailgun.org/messages

The following is what my actual config files ends up looking like.

In the ConfigureServices function Startup.cs I added a reference to the new settings class so it would be available for dependency injection.

Message Services

In the Services folder there is a MessageServices.cs file which contains the AuthMessageSender class that has an empty implementation for sending email base on an IEmailSender interface which defines a single SendEmailAsync method. This function is already being called in the code that was uncommented above so I am going to use it to call mailgun’s API.

First I need to get the email settings defined above injected into the AuthMessageSenderClass by adding a class level field and a constructor. The only thing the constructor is doing is saving a reference to the injected settings class.

Next is the SendEmailAsync function mentioned above which I changed to an async function and added the code to send an email using mailgun’s API.

This code is using the HttpClient to send a request to mailgun’s API using basic authorization and form url encoded content to pass the API the relevant bit of information.

With that your application will now email account conformations and password resets.

Other Email Options

Mailgun is obviously not the only option for sending emails. This post from Mashape lists 12 API providers. In addition SMTP is also an option which this post by Steve Gordon covers.

ASP.NET 5 Web API Put

As part of last week’s post on Aurelia’s Event Aggregator I needed a way to update existing contacts. This was a hole in my existing API and it was time to fill it.  See last week’s post for usage on the client side this post will only be covering the server side.

First off here is the code I added to the ContactsController in the API folder of my project. I thought the overview would be helpful a helpful context as we break down the different parts of the function.

First off the Update function is decorated to only handle HttpPut actions with an ID. The function itself of course requires an ID. The FormBody attribute on the updatedContact parameter tells the framework to build a contact object from the request body.

Next if the updatedContact was unable to be built from the request body or if the ID passed to the update function doesn’t match the ID in the updatedContact then a bad request status is returned.

Next I query the database for a matching contact. This query is limited to the current users so it should help prevent a contact for the wrong user being updated. If no matching contact is found then not found status is returned.

Next check to see if the mode state is valid and return bad request object result if it is not.

This next bit was a little bit of a challenge. Since entity framework was queried to see if the updated contact already exists I was getting an error when trying to use the DB context’s update function. One way to address this error would have been to update the contact pulled from the database with the values from the updated contact from the request body. If I had of gone this route something like automapper would have made the job fairly easy. Instead I used the DB context entry on the contact returned from the database to set its state to detached which tells the context to stop tracking that item. After that the context’s update function worked like a charm and I could save changes based on the updatedContact parameter.

Finally I returned the client the updated contact. This is the same thing I did for post and may or may not be the best thing to do on a put.

I am not sure how much of this is the “correct” way to handle a put. After seeing how some of this is handled when using razor with anti-forgery tokens and the ability to only allow certain fields to be updated this implementation seems lacking. This is a jumping off point, but I need to do more research on how to properly secure an API.