Just in case you missed it, .NET is all open source now and .NET Core is a free, open source, cross-platform framework that you can download and start with in <10 minutes. You can get it on Mac, Windows, and a half-dozen Unixes at http://dot.net. Take that along with the free, cross-platform Visual Studio Code and you'll be writing C# and F# all over the place.
Ok, that said, there's two ways to deploy a .NET Core application. There's FDD and SCD. Since TLAs (three letter acronyms) are stupid, that's Framework-dependent and Self-contained. When .NET Core is installed it ends up in C:\program files\dotnet on Windows, for example. In the "Shared" folder there's a bunch of .NET stuff that is, well, shared. There may be multiple folders, as you can see in my folder below. You can have many and multiple installs of .NET Core.
When you're installing your app and its dependencies BUT NOT .NET Core itself, you're dependent on .NET Core already being on the target machine. That's fine for Web Apps or systems with lots of apps, but what if I want to write an app and give it to you as zip or on a USB key and and I just want it to work. I'll include .NET Core as well so the whole thing is a Self Contained Deployment.
It will make my "Hello World" application larger than if I was using an existing system-wide install, but I know it'll Just Work because it'll be totally self-contained.
If I deploy my app along with .NET Core it's important to remember that I'll be responsible for servicing .NET Core and keeping it up to date. I also need to decide on my target platforms ahead of time. If I want it to run on Windows, Mac, and Linux AND just work, I'll need to include those target platforms and build deployment packages for them. This all makes mostly intuitive sense but it's good to know.
I'll take my little app (I'm just using a "dotnet new" app) and I'll modify project.json in a text editor.
My app is a .NETCore.App, but it's not going to use the .NET Core platform that's installed. It'll use a local version so I'll remove "type="platform"" from this dependency.
Once this is done, I've got my app self-contained in n folders, ready to deploy to whatever systems I want.
You can see in the Win10 folder there's my "MYAPPLICATION.exe" (mine is called scd.exe) that can be run, rather than running things like developers do with "dotnet run."
There's lots of good documentation about how you can tune and define exactly what gets deployed with your self contained application over at the .NET Core Docs. You can do considerable trimming to .NET Core, and there's talk of that becoming more and more automated in the future, possibly down to the method level.
Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.
I noticed today that Evernote was in the Windows Store. I went to the store, installed Evernote, and it ran. No nextnextnextfinish-style install, it just worked and it worked nicely. It's a Win32 app and it appears to use NodeWebKit for part of it's UI. But it's a Windows app, just like VB6 apps and just like .NET apps and just like UWP (Universal Windows Platform) apps, so I found this to be pretty cool. Now that the Evernote app is a store app it can use Windows 10 specific features like Live Tiles and Notifications and it'll be always up to date.
The Windows Store is starting (slowly) to roll out and include existing desktop apps and games by building and packaging those apps using the Universal Windows Platform. This was called "Project Centennial" when they announced it at the BUILD conference. It lets you basically put any Windows App in the Windows Store, which is cool. Apps that live there are safe, won't mess up your machine, and are quickly installed and uninstalled.
Here's some of the details about what's happening with your app behind the scenes, from this article. This is one of the main benefits of the Windows Store. Apps from the Store can't mess up your system on a global scale.
[The app] runs in a special environment where any accesses that the app makes to the file system and to the Registry are redirected. The file named Registry.dat is used for Registry redirection. It's actually a Registry hive, so you can view it in the Windows Registry Editor (Regedit). When it comes to the file system, the only thing redirected is the AppData folder, and it is redirected to the same location that app data is stored for all UWP apps. This location is known as the local app data store, and you access it by using the ApplicationData.LocalFolderproperty. This way, your code is already ported to read and write app data in the correct place without you doing anything. And you can also write there directly. One benefit of file system redirection is a cleaner uninstall experience.
The "DesktopAppConverter" is now packaged in the Windows Store as well, even though it runs at the command prompt! If your Windows Desktop app has a "silent installer" then you can run this DesktopAppConvertor on your installer to make an APPX package that you can then theoretically upload to the Store.
NOTE: This "Centennial" technology is in Windows 10 AU, so if you haven't auto-updated yet, you can get AU now.
They are also working with install vendors like InstallShield and WiX to so their installation creation apps will create Windows Store apps with the Desktop Bridge automatically. This way your existing MSIs and stuff can turn into UWP packages and live in the store.
It looks like there are a few ways to make your existing Windows apps into Windows 10 Store-ready apps. You can use this DesktopAppConverter and run it in your existing silent installer. Once you've made your app a Store App, you can "light up" your app with Live Tiles and Notifications and other features with code. Check out the https://github.com/Microsoft/DesktopBridgeToUWP-Samples GitHub Repro with samples that show you how to add Tiles or Background tasks. You can use [Conditional("DesktopUWP")] compilation if you have both a Windows Store and Windows desktop version of your app with a traditional installer.
If your app is a simple Xcopy-deploy app that has no installer, it's even easier. To prove this I installed Visual Basic 6 on my Windows 10 machine. OH YES I DID.
NOTE: I am using VB6 as a fun but also very cool example. VB6 is long out of support but apps created with it still run great on Windows because they are win32 apps. For me, this means that if I had a VB6 app that I wanted to move into the Store and widen my audience, I could.
I made a quick little Project1.exe in VB6 that runs on its own.
In the folder is my Project1.exe long with an Assets folder with my logo and a few PNGs.
Now I can run the DesktopAppConverter if I have a quiet installer, but since I've just got a small xcopyable app, I'll run this to test on my local machine.
Add-AppxPackage -register .\AppxManifest.xml
And now my little VB6 app is installed locally and in my Start Menu.
NOTE: Make sure the Identity in the AppxManifest matches the code signing cert's CN=Identity. That's the FULL string from the cert. Otherwise you'll see weird stuff in your Event Viewer in Microsoft|Windows\AppxPackagingOM|Microsoft-Windows-AppxPackaging/Operational like "error 0x8007000B: The app manifest publisher name (CN=HanselmanVB6, O=Hanselman, L=Portland, S=OR, C=USA) must match the subject name of the signing certificate exactly (CN=HanselmanVB6)."
I'll use a command line like this. Remember that Visual Studio can hide a lot of this, but since I'm doing it manually it's good to understand the details.
The following certificates were considered: Issued to: HanselmanVB6 Issued by: HanselmanVB6 Expires: Sat Dec 31 00:00:00 2016 SHA1 hash: 19F384D1D0BD33F107B2D7344C4CA40F2A557749
After EKU filter, 1 certs were left. After expiry filter, 1 certs were left. After Private Key filter, 1 certs were left. The following certificate was selected: Issued to: HanselmanVB6 Issued by: HanselmanVB6 Expires: Sat Dec 31 00:00:00 2016 SHA1 hash: 19F384D1D0BD33F107B2D7344C4CA40F2A557749
The following additional certificates will be attached: Done Adding Additional Store Successfully signed: Project1.appx
Number of files successfully Signed: 1 Number of warnings: 0 Number of errors: 0
Now I've got a (local developer) signed, packaged Appx that has a VB6 app inside it. If I double click I'll get the Appx installer, but what I really want to do is sign it with a real cert and put it in the Windows Store!
Here's the app running. Pretty amazing UX, I know.
It's early days, IMHO, but I'm looking forward to a time when I can go to the Windows Store and get my favorite apps like WindowsOpen Live Writer, Office, Slack, and more! Now's the time for you to start exploring these tools.
Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.
Sarah Mei had a great series of tweets last week. She's a Founder of RailsBridge, Director of Ruby Central, and the Chief Consultant of DevMynd so she's experienced with work both "on the job" and "on the side." Like me, she organizes OSS projects, conferences, but she also has a life, as do I.
If you're reading this blog, it's likely that you have gone to a User Group or Conference, or in some way did some "on the side" tech activity. It could be that you have a blog, or you tweet, or you do videos, or you volunteer at a school.
With Sarah's permission, I want to take a moment and call out some of these tweets and share my thoughts about them. I think this is an important conversation to have.
My career has had a bunch of long cycles (months or years in length) of involvement & non-involvement in tech stuff outside of work.
This is vital. Life is cyclical. You aren't required or expected to be ON 130 hours a week your entire working life. It's unreasonable to expect that of yourself. Many of you have emailed me about this in the past. "How do you do _____, Scott?" How do you deal with balance, hang with your kids, do your work, do videos, etc.
I don't.
Sometimes I just chill. Sometimes I play video games. Last week I was in bed before 10pm two nights. I totally didn't answer email that night either. Balls were dropped and the world kept spinning.
Sometimes you need to be told it's OK to stop, Dear Reader. Slow down, breathe. Take a knee. Hell, take a day.
When we pathologize the non-involvement side of the cycle as "burnout," we imply that the involvement side is the positive, natural state.
Here's where it gets really real. We hear a lot about "burnout." Are you REALLY burnt? Maybe you just need to chill. Maybe going to three User Groups a month (or a week!) is too much? Maybe you're just not that into the tech today/this week/this month. Sometimes I'm so amped on 3D printing and sometimes I'm just...not.
Am I burned out? Nah. Just taking in a break.
But you know what? Your kids are only babies once (thank goodness). Those rocks won't climb themselves. Etc. And tech will still be here.
I realize that not everyone with children in their lives can get/afford a sitter but I do also want to point out that if you can, REST. RESET. My wife and I have Date Night. Not once a month, not occasionally. Every week. As we tell our kids: We were here before you and we'll be here after you leave, so this is our time to talk to each other. See ya!
Date Night, y'all. Every week. Self care. Take the time, schedule it, pay the sitter. You'll thank yourself.
Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.
A little Linux VM on Azure is like $13 a month. You can get little Linux machines all over for between $10-15 a month. On Linode they are about $10 a month so I figured it would be interesting to setup an ASP.NET Core website running on .NET Core. As you may know, .NET Core is free, open source, cross platform and runs basically everywhere.
Step 0 - Get a cheap host
I went to Linode (or anywhere) and got the cheapest Linux machine they offered. In this case it's an Ubuntu 14.04 LTS Profile, 64-bit, 4.6.5 Kernel.
Since I'm on Windows but I want to SSH into this Linux machine I'll need a SSH client. There's a bunch of options.
It's always a good idea to avoid being root. After logging into the system as root, I made a new user and give them sudo (super user do):
adduser scott usermod -aG sudo scott
Then I'll logout and go back in as scott.
Step 1 - Get .NET Core on your Linux Machine
Head over to http://dot.net to get .NET Core and follow the instructions. There's at least 8 Linuxes supported in 6 flavors so you should have no trouble. I followed the Ubuntu instructions.
To make sure it works after you've set it up, make a quick console app like this and run it.
mkdir testapp cd testapp dotnet new dotnet restore dotnet run
If it runs, then you've got .NET Core installed and you can move on to making a web app and exposing it to the internet.
Today, this default site uses npm, gulp, and bower to manage JavaScript and CSS dependencies. In the future there will be options that don't require as much extra stuff but for now, in order to dotnet restore this site I'll need npm and what not so I'll do this to get node, npm, etc.
Now I can dotnet restore easily and run my web app to test. It will startup on localhost:5000 usually.
$ dotnet restore $ dotnet run scott@ubuntu:~/dotnettest$ dotnet run Project dotnettest (.NETCoreApp,Version=v1.0) was previously compiled. Skipping compilation. info: Microsoft.Extensions.DependencyInjection.DataProtectionServices[0] User profile is available. Using '/home/scott/.aspnet/DataProtection-Keys' as key repository; keys will not be encrypted at rest. Hosting environment: Production Content root path: /home/scott/dotnettest Now listening on: http://localhost:5000
Of course, having something startup on localhost:5000 doesn't help me as I'm over here at home so I can't test a local website like this. I want to expose this site (via a port) to the outside world. I want something like http://mysupermachine -> inside my machine -> localhost:5000.
Step 3 - Expose your web app to the outside.
I could tell Kestrel - that's the .NET Web Server - to expose itself to Port 80, although you usually want to have another process between you and the outside world.
You can do this a few ways. You can open open Program.cs with a editor like "pico" and add a .UseUrls() call to the WebHostBuilder like this.
var host = new WebHostBuilder() .UseKestrel() .UseUrls("http://*:80") .UseContentRoot(Directory.GetCurrentDirectory()) .UseStartup<Startup>() .Build();
Here the * binds to all the network adapters and it listens on Port 80. Putting http://0.0.0.0:80 also works.
You might have permission issues doing this and need to elevate the dotnet process and webserver which is also a problem so let's just keep it at a high internal port and reverse proxy the traffic with something like Nginx or Apache. We'll pull out the hard-coded port from the code and change the Program.cs to use a .json config file.
public static void Main(string[] args) { var config = new ConfigurationBuilder() .SetBasePath(Directory.GetCurrentDirectory()) .AddJsonFile("hosting.json", optional: true) .Build();
var host = new WebHostBuilder() .UseKestrel() .UseConfiguration(config) .UseContentRoot(Directory.GetCurrentDirectory()) .UseStartup<Startup>() .Build();
NOTE: I'm doing this work a folder under my home folder ~ or now. I'll later compile and "publish" this website to something like /var/dotnettest when I want it seen.
sudo apt-get install nginx sudo service nginx start
I'm going to change the default Nginx site to point to my (future) running ASP.NET Core web app. I'll open and change /etc/nginx/sites-available/default and make it look like this. Note the port number. Nginx is a LOT more complex than this and has a lot of nuance, so when you are ready to go into Super Official Production, be sure to explore what the perfect Nginx Config File looks like and change it to your needs.
The website isn't up and running on localhost:5123 yet (unless you've run it yourself and kept it running!) so we'll need an app or a monitor to run it and keep it running. There's an app called Supervisor that is good at that so I'll add it.
sudo apt-get install supervisor
Here is where you/we/I/errbody needs to get the paths and names right, so be aware. I'm over in ~/testapp or something. I need to publish my site into a final location so I'm going to run dotnet publish, then copy the reuslts into /var/dotnettest where it will live.
dotnet publish publish: Published to /home/scott/dotnettest/bin/Debug/netcoreapp1.0/publish sudo cp -a /home/scott/dotnettest/bin/Debug/netcoreapp1.0/publish /var/dotnettest
Now I'm going to make a file (again, I use pico because I'm not as awesome as emacs or vim) called /src/supervisor/conf.d/dotnettest.conf to start my app and keep it running:
Now we start and stop Supervisor and watch/tail its logs to see our app startup!
sudo service supervisor stop sudo service supervisor start sudo tail -f /var/log/supervisor/supervisord.log #and the application logs if you like sudo tail -f /var/log/dotnettest.out.log
If all worked out (if it didn't, it'll be a name or a path so keep trying!) you'll see the supervisor log with dotnet starting up, running your app.
Remember the relationships.
Dotnet - runs your website
Nginx or Apache - Listens on Port 80 and forwards HTTP calls to your website
Supervisor - Keeps your app running
Next, I might want to setup a continuous integration build, or SCP/SFTP to handle deployment of my app. That way I can develop locally and push up to my Linux machine.
Of course, there are a dozen other ways to publish an ASP.NET Core site, not to mention Docker. I'll post about Docker another time, but for now, I was able to get my ASP.NET Core website published to a cheap $10 host in less than an hour. You can use the same tools to manage a .NET Core site that you use to manage any site be it PHP, nodejs, Ruby, or whatever makes you happy.
Sponsor: Aspose makes programming APIs for working with files, like: DOC, XLS, PPT, PDF and countless more. Developers can use their products to create, convert, modify, or manage files in almost any way. Aspose is a good company and they offer solid products. Check them out, and download a free evaluation.
About Scott
Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.
There's a lot of confusing terms in the Cloud space. And that's not counting the term "Cloud." ;)
IaaS (Infrastructure as a Services) - Virtual Machines and stuff on demand.
PaaS (Platform as a Service) - You deploy your apps but try not to think about the Virtual Machines underneath. They exist, but we pretend they don't until forced.
SaaS (Software as a Service) - Stuff like Office 365 and Gmail. You pay a subscription and you get email/whatever as a service. It Just Works.
"Serverless Computing" doesn't really mean there's no server. Serverless means there's no server you need to worry about. That might sound like PaaS, but it's higher level that than.
Serverless Computing is like this - Your code, a slider bar, and your credit card. You just have your function out there and it will scale as long as you can pay for it. It's as close to "cloudy" as The Cloud can get.
With Platform as a Service, you might make a Node or C# app, check it into Git, deploy it to a Web Site/Application, and then you've got an endpoint. You might scale it up (get more CPU/Memory/Disk) or out (have 1, 2, n instances of the Web App) but it's not seamless. It's totally cool, to be clear, but you're always aware of the servers.
New cloud systems like Amazon Lambda and Azure Functions have you upload some code and it's running seconds later. You can have continuous jobs, functions that run on a triggered event, or make Web APIs or Webhooks that are just a function with a URL.
I'm going to see how quickly I can make a Web API with Serverless Computing.
I'll go to http://functions.azure.com and make a new function. If you don't have an account you can sign up free.
You can make a function in JavaScript or C#.
Once you're into the Azure Function Editor, click "New Function" and you've got dozens of templates and code examples for things like:
Find a face in an image and store the rectangle of where the face is.
Run a function and comment on a GitHub issue when a GitHub webhook is triggered
Update a storage blob when an HTTP Request comes in
Load entities from a database or storage table
I figured I'd change the first example. It is a trigger that sees an image in storage, calls a cognitive services API to get the location of the face, then stores the data. I wanted to change it to:
Take an image as input from an HTTP Post
Draw a rectangle around the face
Return the new image
You can do this work from Git/GitHub but for easy stuff I'm literally doing it all in the browser. Here's what it looks like.
I code and iterate and save and fail fast, fail often. Here's the starter code I based it on. Remember, that this is a starter function that runs on a triggered event, so note its Run()...I'm going to change this.
#r "Microsoft.WindowsAzure.Storage"
#r "Newtonsoft.Json"
using System.Net;
using System.Net.Http;
using System.Net.Http.Headers;
using Newtonsoft.Json;
using Microsoft.WindowsAzure.Storage.Table;
using System.IO;
public static async Task Run(Stream image, string name, IAsyncCollector<FaceRectangle> outTable, TraceWriter log)
{
var image = await req.Content.ReadAsStreamAsync();
string result = await CallVisionAPI(image); //STREAM
log.Info(result);
if (String.IsNullOrEmpty(result))
{
return req.CreateResponse(HttpStatusCode.BadRequest);
}
ImageData imageData = JsonConvert.DeserializeObject<ImageData>(result);
foreach (Face face in imageData.Faces)
{
var faceRectangle = face.FaceRectangle;
faceRectangle.RowKey = Guid.NewGuid().ToString();
faceRectangle.PartitionKey = "Functions";
faceRectangle.ImageFile = name + ".jpg";
await outTable.AddAsync(faceRectangle);
}
return req.CreateResponse(HttpStatusCode.OK, "Nice Job");
}
static async Task<string> CallVisionAPI(Stream image)
{
using (var client = new HttpClient())
{
var content = new StreamContent(image);
var url = "https://api.projectoxford.ai/vision/v1.0/analyze?visualFeatures=Faces";
client.DefaultRequestHeaders.Add("Ocp-Apim-Subscription-Key", Environment.GetEnvironmentVariable("Vision_API_Subscription_Key"));
content.Headers.ContentType = new MediaTypeHeaderValue("application/octet-stream");
var httpResponse = await client.PostAsync(url, content);
if (httpResponse.StatusCode == HttpStatusCode.OK){
return await httpResponse.Content.ReadAsStringAsync();
}
}
return null;
}
public class ImageData {
public List<Face> Faces { get; set; }
}
public class Face {
public int Age { get; set; }
public string Gender { get; set; }
public FaceRectangle FaceRectangle { get; set; }
}
public class FaceRectangle : TableEntity {
public string ImageFile { get; set; }
public int Left { get; set; }
public int Top { get; set; }
public int Width { get; set; }
public int Height { get; set; }
}
GOAL: I'll change this Run() and make this listen for an HTTP request that contains an image, read the image that's POSTed in (ya, I know, no validation), draw rectangle around detected faces, then return a new image.
public static async Task<HttpResponseMessage> Run(HttpRequestMessage req, TraceWriter log) { var image = await req.Content.ReadAsStreamAsync();
As for the body of this function, I'm 20% sure I'm using too many MemoryStreams but they are getting disposed so take this code as a initial proof of concept. However, I DO need at least the two I have. Regardless, happy to chat with those who know more, but it's more subtle than even I thought. That said, basically call out to the API, get back some face data that looks like this:
Then take that data and DRAW a Rectangle over the faces detected.
public static async Task<HttpResponseMessage> Run(HttpRequestMessage req, TraceWriter log)
{
var image = await req.Content.ReadAsStreamAsync();
MemoryStream mem = new MemoryStream();
image.CopyTo(mem); //make a copy since one gets destroy in the other API. Lame, I know.
image.Position = 0;
mem.Position = 0;
string result = await CallVisionAPI(image);
log.Info(result);
if (String.IsNullOrEmpty(result)) {
return req.CreateResponse(HttpStatusCode.BadRequest);
}
ImageData imageData = JsonConvert.DeserializeObject<ImageData>(result);
MemoryStream outputStream = new MemoryStream();
using(Image maybeFace = Image.FromStream(mem, true))
{
using (Graphics g = Graphics.FromImage(maybeFace))
{
Pen yellowPen = new Pen(Color.Yellow, 4);
foreach (Face face in imageData.Faces)
{
var faceRectangle = face.FaceRectangle;
g.DrawRectangle(yellowPen,
faceRectangle.Left, faceRectangle.Top,
faceRectangle.Width, faceRectangle.Height);
}
}
maybeFace.Save(outputStream, ImageFormat.Jpeg);
}
var response = new HttpResponseMessage()
{
Content = new ByteArrayContent(outputStream.ToArray()),
StatusCode = HttpStatusCode.OK,
};
response.Content.Headers.ContentType = new MediaTypeHeaderValue("image/jpeg");
return response;
}
I also added a reference to System. Drawing using this syntax at the top of the file and added a few namespaces with usings like System.Drawing and System.Drawing.Imaging. I also changed the input in the Integrate tab to "HTTP" as my input.
#r "System.Drawing
Now I go into Postman and POST an image to my new Azure Function endpoint. Here I uploaded a flattering picture of me and unflattering picture of The Oatmeal. He's pretty in real life just NOT HERE. ;)
So in just about 15 min with no idea and armed with just my browser, Postman (also my browser), Google/StackOverflow, and Azure Functions I've got a backend proof of concept.
Azure Functions supports Node.js, C#, F#, Python, PHP *and* Batch, Bash, and PowerShell, which really opens it up to basically anyone. You can use them for anything when you just want a function (or more) out there on the web. Send stuff to Slack, automate your house, update GitHub issues, act as a Webhook, etc. There's some great 3d party Azure Functions sample code in this GitHub repo as well. Inputs can be from basically anywhere and outputs can be basically anywhere. If those anywheres are also cloud services like Tables or Storage, you've got a "serverless backed" that is easy to scale.
I'm still learning, but I can see when I'd want a VM (total control) vs a Web App (near total control) vs a "Serverless" Azure Function (less control but I didn't need it anyway, just wanted a function in the cloud.)
Sponsor: Aspose makes programming APIs for working with files, like: DOC, XLS, PPT, PDF and countless more. Developers can use their products to create, convert, modify, or manage files in almost any way. Aspose is a good company and they offer solid products. Check them out, and download a free evaluation.
About Scott
Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.
