DEV Community: Alex Ruzenhack

Reads of the week

Alex Ruzenhack — Wed, 03 Oct 2018 23:48:09 +0000

I consume some articles on Medium every week, some of them made me reevaluate my learning strategies, or improve my coding habit, or just let me know some useful career tip. Sometimes I feel the content could reach more people. So I'm bringing them to my audience here, first as a general test. Feel free to contact me or discuss one or two subjects of the reading list.

#Programming #Architecture

#Security

How I used a simple Google query to mine passwords from dozens of public Trello boards

#Ethic #Neuroscience #Future

Did you find any article useful? Which one? Why?

That's it!

Reads of the week

Alex Ruzenhack — Thu, 27 Sep 2018 00:03:12 +0000

#Learning #Writing #Productivity

#Programming

#Work #Careers

#AI #Blockchain

#Basic Income

The Real Story of Automation Beginning with One Simple Chart

#Life Lessons #Self Improvement

The 3 Keys to Becoming Irresistible

Did you find any article useful? Which one? Why?

That's it!

Using C# LINQ API to perform SQL IN logic at WHERE clause

Alex Ruzenhack — Thu, 30 Aug 2018 12:24:58 +0000

I was working on a project with User and Costumer identificação. The first is generated by the web app, and the second by the business department.

So I had to develop a logic to make those match between identifications.

The User entity needs to know the Costumer identification and, in order to make the match, first I selected the users without a costumer and extracted a list of emails. Then, I selected the customers based on this email list. This statement can be transcribed to SQL syntax, like this:

SELECT *
FROM Customers
WHERE email IN (SELECT email FROM Users WHERE CustomerId IS NULL)

But how to translate the SQL to C# LINQ API?

First, make the selection from Users, extracting the list of emails
Select the Customers with those emails

This way:

var emails = _userManager.Users
    .Where(user => user.CustomerId == null)
    .Select(user => user.Email) // extract the emails from users
    .ToList();

var customers = _applicationRepository.GetCustomers()
    .Where(customer => emails.Contains(customer.Email)) // the Contains method carry the IN logic when translated to SQL script
    .ToList();

Now we have the match programmatically.

That's it!

Reference

Add a unique nullable index to an entity property

Alex Ruzenhack — Wed, 29 Aug 2018 14:11:36 +0000

The problem

I was working on an MVC project and to simplify our life, the team chose to work with the ASP.NET Core Identity Assembly to create the authentication and authorization logic of the web app.

However, when a user registers yourself on the app, a new Id is generated for them, despite the company may already have one Id.

Then, to link together the new Id with the natural old Id, I extended the default IdentityUser to possess the MyCompanyId property, this way:

public class CustomUser : IdentityUser
{
    public int MyCompanyId { get; set; }
}

And in order to make this change effective I extended the IdentityDbContext to use my CustomUser and also to register the property MyCompanyId as a unique index, like this:

public class CustomDbContext : IdentityDbContext<CustomUser>
{
    public CustomDbContext(DbContextOptions<CustomDbContext> options)
        : base(options)
    {
    }

    protected override void OnModelCreating(ModelBuilder builder)
    {
        base.OnModelCreating(builder);

        builder.Entity<CustomUser>(user => 
        {
            user.HasIndex(x => x.MyCompanyId).IsUnique(true);
        });
    }
}

With this, when a user registers yourself a default value is assigned to MyCompanyId property, which is 0. When another user tries to register, the same value will be assigned to their MyCompanyId, and it will throw an error of duplicate key.

It occurs because when the entity is transcribed to a fluentish API for migration, by default, the column is set as non-nullable and a default value must be assigned, as we saw 0.

// hidden for brevity
...
migrationBuilder.CreateTable(
    name: "AspNetUsers",
    columns: table => new
    {
        Id = table.Column<string>(nullable: false),
        ...
        MyCompanyId = table.Column<int>(nullable: false)
    },
...

And the index assigns uniqueness to all values:

// hidden for brevity
...
migrationBuilder.CreateIndex(
    name: "IX_AspNetUsers_MyCompanyId",
    table: "AspNetUsers",
    column: "MyCompanyId",
    unique: true);
...

The solution

How to solve this all, and allow a nullable unique index? It's so simple, man! Just add the ? symbol after the type of the property, this way:

public class CustomUser : IdentityUser
{
    public int? MyCompanyId { get; set; }
}

Now the property is optional and it will be propagated all way down. Let's see the migration fluentish API:

// hidden for brevity
...
migrationBuilder.CreateTable(
    name: "AspNetUsers",
    columns: table => new
    {
        Id = table.Column<string>(nullable: false),
        ...
        MyCompanyId = table.Column<int>(nullable: true) // now it's true
    },
...

And the index has a new variable filter which skip the null values as unique. It means you can have as many null values as you want.

// hidden for brevity
...
migrationBuilder.CreateIndex(
    name: "IX_AspNetUsers_MyCompanyId",
    table: "AspNetUsers",
    column: "MyCompanyId",
    unique: true,
    filter: "[MyCompanyId] IS NOT NULL");
...

That's it!

What do you do in your two hours out of home?

Alex Ruzenhack — Sun, 26 Aug 2018 21:24:18 +0000

I read the StackOverflow statistics for this year (2018) and it shows the time developers spend outside is about 30min to 2h per day.

This number is very reasonable, by my self. Now I'm curious about: What you do at this time?

During the work week I spend time riding a bike and exercising me, and at the weekend I try to know the cool places in my city, like the "Two Brothers Hill" in Rio de Janeiro.

Passing serialized C# object in JSON to Razor Page

Alex Ruzenhack — Fri, 24 Aug 2018 16:18:50 +0000

I was working in a way to convey a dataset coming from the database to Google Chart JavaScript framework, using C# and .NET Core MVC framework under the Razor Page.

The first step was easy, a trivial query to the database in order to get the data, but Google Chart has their special way to handle a dataset and to realize a simple conversion from POCO was not sufficient, then I wrote a converter:

private IEnumerable<Array> poco2GoogleChartDataset(IEnumerable<MyData> dataset)
{
    List<Array> googleDataset = new List<Array>();
    foreach (MyData data in dataset) {
        Array arr = new object[] { data.Label, data.Value };
        googleDataset.Add(arr);
    }
    return googleDataset.ToList();
}

The result of this conversion was serialized to JSON using the Newtonsoft.Json assembly:

// Top of the class
using Newtonsoft.Json;

// Hidden for brevity
ReportJson = JsonConvert.SerializeObject(poco2GoogleChartDataset(myDataset));

On the other hand, in Razor Page, I need to pass this JSON object to a JavaScript variable. If you just pass the model value using a simple Razor syntax, it will not work because in the Razor Render process will pass the value through an HTML encoding. Like this:

// Razor Page
...
@section Scripts {
    function drawChart() {
        var jsonDataset = @Model.ReportJson;
        ...
    }
...
}

jsonDataset result:
[[\&#x22;Label\&#x22;, value]]

To bypass the HTML encoding wrap the model variable into the HtmlHelper directive with their Raw method. This way:

// Razor Page
...
@section Scripts {
    function drawChart() {
        var jsonDataset = @Html.Raw(Model.ReportJson);
        ...
    }
...
}

Now the value result will be a JSON object, ready to be used in Google Chart framework.

I suffered hours to achieve this comprehension, in part because I do not master Razor syntax, nor ASP.NET Core MVC, neither the C# 😂

That's it!

Git merge specific file from another branch

Alex Ruzenhack — Thu, 23 Aug 2018 21:04:12 +0000

There are two teams working on the same project and they realize a common problem that was already resolved by one side, but the status of work in each branch is in progress and to make a merge can be unsafe.

How to solve this problem?

There is a safe solution: merge a specific file from another branch. And there is the steps:

$ git checkout <another-branch> <path-to-file> [<one-more-file> ...]
$ git status
$ git commit -m "'Merge' specific file from '<another-branch>'"

That's it!

Here is a more detailed approach.

Como criar uma aplicação REST API básica com Spark

Alex Ruzenhack — Mon, 14 May 2018 00:58:41 +0000

Através do micro framework Spark é possível criar micro serviços web com grande facilidade, envelopar num jar gordo e botar pra rodar numa micro instância Docker.

1. Crie um projeto maven

Configure o arquivo pom.xml com as seguintes dependências:

<dependencies>
    <dependency>
        <groupId>com.sparkjava</groupId>
        <artifactId>spark-core</artifactId>
        <version>2.7.1</version>
    </dependency>

    <dependency>
        <groupId>org.slf4j</groupId>
        <artifactId>slf4j-simple</artifactId>
        <version>1.7.21</version>
    </dependency>
</dependencies>

2. Configure a montagem do jar

Construa um jar gordo que contenha todas as dependências do seu projeto, assim você pode rodar sua aplicação de forma independente.

<plugin>
    <groupId>org.apache.maven.plugins</groupId>
    <artifactId>maven-assembly-plugin</artifactId>
    <executions>
        <execution>
            <goals>
                <goal>attached</goal>
            </goals>
            <phase>package</phase>
            <configuration>
                <finalName>${project.name}</finalName>
                <descriptorRefs>
                    <descriptorRef>jar-with-dependencies</descriptorRef>
                </descriptorRefs>
                <archive>
                    <manifest>
                        <mainClass>yourgroupid.sparkrest.HelloWorld</mainClass>
                    </manifest>
                </archive>
            </configuration>
        </execution>
    </executions>
</plugin>

3. Crie um endpoint para sua REST API

Crie uma classe chamada HelloWorld onde você irá configurar um endpoint que receba requisições GET e devolva o texto simples Hello World.

import static spark.Spark.*;

public class HelloWorld {
    public static void main(String[] args) {
        get("/hello", (req, res) -> "Hello World");
    }
}

4. Rode sua aplicação

Pra finalizar, execute esse script para compilar sua aplicação

$> mvn dependency:resolver
$> mvn verify
$> mvn package

Agora execute o jar gordo compilado

$> java -jar nomedoprojeto-jar-with-dependencies.jar

Faça uma requisição na URL http://localhost:4567/hello

// Resposta
Hello World

Conclusão

O Spark oferece uma solução de micro serviço para aplicações web REST API com requisições bloqueantes, isto é, uma requisição fica presa numa tread. Portanto, a recomendação é que as aplicações construídas sobre esse framework sejam simples.

Para escalar aplicações que utilizem essa tecnologia é preciso usar uma arquitetura de microserviços com load balancer e criar novas instâncias de acordo com a demanda.

ConfiguraÃ§Ã£o bÃ¡sica de processos em paralelo do PostgreSQL

Alex Ruzenhack — Tue, 24 Jan 2017 18:04:42 +0000

Para gerar uma execuÃ§Ã£o paralela de planos Ã© acrescentado um nÃ³-de-reuniÃ£o ao nÃ³-geral do plano de execuÃ§Ã£o. Na etapa de execuÃ§Ã£o, quando o nÃ³-de-reuniÃ£o Ã© atingido, entÃ£o o processo principal faz a requisiÃ§Ã£o de alguns outros processos de execuÃ§Ã£o para trabalhar o plano do nÃ³-de-reuniÃ£o em paralelo ao plano do nÃ³-geral.

No entanto, hÃ¡ um limite de processos em paralelo estabelecido pelo sistema e isso pode impactar negativamente a performance da query em execuÃ§Ã£o.

A soluÃ§Ã£o Ã© alterar a variÃ¡vel de sistema max_worker_processes para um limite superior, ou reduzir o nÃºmero mÃ¡ximo de processos em cada nÃ³-de-reuniÃ£o atravÃ©s da variÃ¡vel de sistema max_parallel_workers_per_gather.

Hi, I'm Alex Ruzenhack

Alex Ruzenhack — Fri, 13 Jan 2017 19:25:06 +0000

I have been coding for 2 years.

You can find me on GitHub as alexruzenhack

I live in Rio de Janeiro, Brazil.

I work for Nas11

I mostly program in these languages: Java, SQL, Ruby and Python.

I am currently learning more about Cyber Security.

Nice to meet you.