Summary
In this episode, we create the application entry point by setting up a main.go file in the CMD directory with a main function and a testable run function. We configure an HTTP server with the router, set timeouts, and successfully start the application on localhost:8080, confirming that our blog is working by viewing sample articles in the browser. However, we still need to set up a proper Postgres database before we can build out the full application functionality.
Transcript
So far, we have set up models, views, controllers, we have configured a router, we have made a way to configure the environment we are operating in, but we still have one critical question left, and that is how do we actually start and run the application? In Go projects, you typically have an entry point that lives under a CMD that is short for command directory, which is where we put the binaries of our project, the actual programs that we run. So for our blog, this means that we're going to create this command slash app slash main, so we have the entry point to our app in this place. And this is going to contain files. For now, at least two elements. It's going to contain the main function that we have to have there, and then it's going to contain a run function. But let me create the main function first, and then I'm going to explain the run function after. So we're going to do func main. We're going to create our context. Set it in the background. Then we're going to grab our configuration, calling .new. If there is an error, we are not going to return an error. We're going to say OS exit with status 1, and then we're going to say if error equals to run. And this function that we haven't created is going to accept context and the configuration. And if there is an error, we're going to say OS exit 1. Now, what this does is that we can actually create tests for our entry point for running the application. So we know that whenever we make changes that the application can run because this is the only entry point. And by having a run function that returns an error, we can create tests around this. And this is really, really nice when you start to get to larger environments. We have the option to pre-configure a test for a test environment that we can actually validate that our app compiles and runs. Now, this run function here is going to contain a few things. So let me just flesh out what the signature is going to look like. So we're going to have the context. We're going to have the config, and then we're going to return the error. And then inside of here, you're simply going to say router equals router new. And then we're going to create a server variable or SRV, which is just short for server. Then we're going to use an HTTP.server from the standard library. And in here, we're going to have an address, which we can now grab from our config, the get TCP address. We're going to have a handler, which is simply just going to be the router. And then the field called handler, which is technically just an echo instance. But the naming here makes it make sense. Finally, we're going to add our read timeout, which is just going to be one times one. Let's just go for a minute. And then our write timeout. I misspelled timeout here. Just go for two minutes. So that if a read or write operation takes longer than this, we will then return an error that a timeout has occurred and then accept new requests again. So we don't hang in the event of something going wrong. Finally, we're going to set a base context. It's going to be a func. We're going to say net listener. It's going to return context. And it's just going to return the context that we are passing. So we actually use the same context for everything. And then underneath here, we're going to say slug info. Starting server. And we're going to say error. And then serve. Finally, we're going to say return serve listen and serve. With this, we now have a proper and testable application. And we can now run it by saying go run command app main.go. But nothing great is really going to happen yet. We haven't styled it. We haven't set up a proper database. So we're not really storing or retrieving any proper data. But if we were to say source.env, and I say go run command app main.go, we can see that it now starts on localhost:8080. And if I just quickly jump into our browser here and then go to localhost:8080, we can see we have a sample article readme and another sample article readme. We can click the readme links to see the article. So our app is working. Our blog is technically working. But we still need a database. And then we can actually start to properly build out this application. So in the next module, we're going to focus on Postgres, setting up a database, and talking through a bunch of Postgres-specific concepts.