Category Archives: Go

Google auth with Beego

As I move forward with my Beego project, I have reached a point where I need to authenticate my users. In the flow that I’m looking for, the client (web app, mobile app, etc…), will communicate directly with the Auth provider (Google, Facebook, etc…) and get a JWT. The only thing the server needs to do is validate the JWT. If the validation succeeds, it means the user is logged in.

If you are not familiar with JWT, you can read my previous article that explains how JWT works.

Authentication flow

Since the authentication with Google is going to happen entirely on the client, the server logic becomes a lot simpler. For my application, all endpoints will require the user to be logged in, so I will create a middleware to verify this. The middleware will expect a valid JWT in the Authorization header. If this requirement is not met, the server will return a 401.

Read more »

Models in Beego

Before we start creating models, we need to make sure our app can connect to our database. At the time of this writing Beego supports MySQL, PostgreSQL and Sqlite3.

We can have the ORM connect to our database from our main.go file:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
package main

import (
    "os"
    "strings"

    _ "app/routers"

    "github.com/astaxie/beego"
    "github.com/astaxie/beego/orm"
    _ "github.com/go-sql-driver/mysql"
)

func init() {
    orm.RegisterDriver("mysql", orm.DRMySQL)
    parts := []string{os.Getenv("MYSQL_USER"), ":", os.Getenv("MYSQL_PASSWORD"),
            "@tcp(", os.Getenv("MYSQL_HOST"), ":3306)/", os.Getenv("MYSQL_DATABASE")}
    orm.RegisterDataBase("default", "mysql", strings.Join(parts, ""))
}

func main() {
    if beego.BConfig.RunMode == "dev" {
        beego.BConfig.WebConfig.DirectoryIndex = true
        beego.BConfig.WebConfig.StaticDir["/swagger"] = "swagger"
    }
    beego.Run()
}

Read more »

Database migrations in Beego

A few weeks ago I wrote an article showing how to start a simple Beego server. Today I’m going to go a little further and explain how to start building your database using migrations. For my examples I will use a MySQL server with an empty database (no tables) named beego.

It is possible to just log in to a database server and type the SQL commands necessary to create or modify tables when we need to, but using migrations helps keep track of the changes made to the database over time. This provides something similar to version control at the database level.

If we need to create a new table, we would start by creating a new migration file using the bee tool:

1
bee generate migration create_user_table

Read more »

Introduction to Beego

I tried GoLang for the first time a little less than two years ago and I didn’t get a very good first impression of it. One of the things that I disliked the most is that back then the community seemed to be overly focused in performance even when that meant creating unreadable code.

One of the things that bothered me the most was that the community was against creating web frameworks with opinions. Because of this lack of standardization, I had to relearn how things are done every time I looked at a new web project written in Go.

Today I decided to look again into the state of web frameworks and there seems to be a few promising options out there. I saw a good amount of people recommending Beego, so I decided to give it a try.

Read more »

Dependency management with golang/dep

It has been more than a year since I wrote an article about dependency management with Glide. It seems like things have changed a little since then. An official package manager has been started by the community, which hopefully will make things easier for developers.

Install

For go applications more than with any other language (because of the necessity of GOPATH), I highly recommend using docker. This is a minimal Dockerfile I’m using that includes both Go and Dep:

Read more »

Golang: Sane dependency management with Glide

In a previous article I wrote an article explaining how to do dependency management wrong by following Go’s recommendations. This week I’m going to explore a better way to manage your dependencies.

Last year the Go community decided to try to fix the dependency management problem they had. Since this problem came from the root, the solution had to come from the same place. The big problem came from the fact that dependencies were pulled from GOPATH. This gave go users no way to have two versions of the same library or application installed in the same computer.

To fix this the vendor folder was created. This allows projects to store dependencies in a folder named vendor inside the project folder. This can be done recursively, so dependencies can store their own dependencies and so on. This allows each project to have it’s own dependencies without affecting other projects.

This resembles same dependency management systems, like npm. The problem is that the community didn’t provide any tooling to help you manage the dependencies. It is your responsibility to download the dependencies and put them in the vendor folder. Luckily other projects were born to help make this easier.

Read more »

Golang: Dependency management done wrong

I have just begun my journey in the Go universe and so far I have found a few things that I don’t really like. I consider this natural because as I get familiar with a way of working I find it hard to accept other ways without questioning them very heavily first.

I’m not an expert in doing dependency management, but when a friend told me how Go decided to do it, it really hurt my soul. Before I begin telling you why it did and why I believe it is the wrong way to do dependency management let me add a disclaimer:

The Go team realized that the out of the box way of doing dependency management was not ideal so they came up with a solution. If you are going to start a project that has dependencies in other projects you should use Golang’s new proposal for package management.

Read more »

Go Language: Concurrency

According to wikipedia concurrency can be defined like this:

The property of program, algorithm, or problem decomposability into order-independent or partially-ordered components or units.[1] This means that even if the concurrent units of the program, algorithm, or problem are executed out-of-order or in partial order, the result will remain determinate.

or this:

A form of computing in which several computations are executing during overlapping time periods—concurrently—instead of sequentially (one completing before the next starts)

An example of a concurrent program could be a web server that can receive multiple requests “at the same time”. If the server wasn’t concurrent, users would have to wait for another request to be processed before the server could handle their request.

Go is said to make concurrency very easy to program so today I’m going to explore it.

To explain concurrency it’s useful to have parts of a program that can execute asynchronously or in parallel. As an example I’m going to use a pair of functions that pretend to get some information from a database or service. Lets look at a simple example where everything is executed without concurrency:

Read more »

Go Language: The For loop

One of the first things I heard about Go that sounded interesting is that it has only one way to create loops. Go only provides the for statement. This sounds a little weird, but the truth is that they just decided to use the for keyword for while loops instead of having a separate while keyword. Lets see an example:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
package main

import (
  "fmt"
)

func main() {
  people := []string{"Hugo", "Paco", "Luis"}

  // Like a for
  numPeople := len(people)
  for i := 0; i < numPeople; i++ {
    fmt.Println(people[i])
  }

  // Like a while
  numPeople = len(people)
  i := 0
  for i < numPeople {
    fmt.Println(people[i])
    i++
  }
}

Read more »

Go Language: Slices

A few weeks ago I talked about arrays in Go. This time I’m going to cover slices, which are built on top of arrays.

In my previous post I showed the difference between doing this:

1
2
3
4
5
6
7
8
9
10
11
12
func doSomething(arr [3]int) {
    arr[0] = 5
}

func main() {
    a := [3]int{1, 2, 3}
   
    doSomething(a)

    // Prints [1 2 3]
    fmt.Println(a)
}

Read more »