kcctl is the missing Command Line Interface (CLI) for kafka connect implementations like e.g. debezium. kcctl is a fast and convenient Connect REST Interface wrapper, completely written in Java:

A Java regex pattern "java or duke" matches on a regular Java string:

import static org.junit.jupiter.api.Assertions.assertFalse;
import static org.junit.jupiter.api.Assertions.assertTrue;

import java.util.regex.Pattern;

import org.junit.jupiter.api.Test;

public class RegexTest {

    @Test
    public void singleLine(){
        var pattern = Pattern.compile(".*(java|duke).*");
        var text = "java duke";
        assertTrue(pattern.matcher(text).matches());
        
    }
    

...but fails on multiline Java String / Text Blocks:

@Test
public void multiLine() {
    var pattern = Pattern.compile(".*(java|duke).*");
    var message = """
        java
        duke
    """;
    assertFalse(pattern.matcher(message).matches());
}

The option Pattern.DOTALL:

(...) In dotall mode, the expression . matches any character, including a line terminator. By default this expression does not match line terminators (...)

makes the regular expression work on multiline strings:

@Test
public void multiLineDotAll() {
    var pattern = Pattern.compile(".*(java|duke).*",Pattern.DOTALL);
    var message = """
        java
        duke
    """;
    assertTrue(pattern.matcher(message).matches());
}

Subscribe to airhacks.fm podcast via: spotify| iTunes| RSS

Web Components are supported by all major browsers (https://caniuse.com/custom-elementsv1) and therefore do not require any additional shims or polyfills. Custom Elements combined with modern JavaScript versions (ES 6+) already provide a Java-like "look and feel":

class JavaDev extends HTMLElement {
    connectedCallback() {
        const message = "Web Components"
        this.innerHTML = `Hello, ${message}`;
    }
    }
    customElements.define('java-dev',JavaDev);

However, the "vanilla" Web Component approach would require a complete re-rendering of the entire element and will result in slower performance in more sophisticated applications. Because we are setting a raw string, which can also contain user input and data fetched from a server, the above approach also comes with a security risk. Thankfully, JavaScript provides a way to declare functions "tagged templates" (https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Template_literals#tagged_templates), which process the raw String and can realize a faster or more secure rendering.

A tagged template function receives the entire String with all "bindings," which allows the implementation of additional syntactic sugar. lit-html is an example of such implementation, which optimizes the rendering, escapes JavaScript code, and adds convenient event handlers' attachment:

const template = html`
<button @click="${_ => this.newBookmark()}">new bookmark</button>
`;

Redux (https://redux.js.org) is a lean (single file) implementation of the "mediator" pattern. It is a central singleton that manages the entire state of the application and notifies all listeners about the changes. The Redux library's use may lead to a significant boilerplate, but it integrates well with lit-html and Web Components. Web Components combined with lit-html, redux, and a few conventions are productive and scale for more complex, interactive applications.

The "Boundary Control Entity" is one of the oldest architectural patterns (https://en.wikipedia.org/wiki/Entity-control-boundary), is similar to MVC, and even provides icons available in all popular design and sketching tools. Equipped with BCE, you don't have to argue about the naming conventions, can conveniently ignore the "Parkinson's law of triviality" (https://en.wikipedia.org/wiki/Law_of_triviality), and focus on the semantics instead.

• Boundary: contains the views implemented with Custom Elements ("Web Components"). The custom elements react to user input and forward the extracted, relevant data to the "controls." All views are listening to redux store changes and are re-rendered at any state mutation.
• Control: are the glue between the visual views and the redux "store." A control package contains functions that implement business logic and backend communication. Controls are asynchronous, "fire-and-forget" functions that process the user input, call external services, and dispatch the data as "redux actions" to the store.
• Entity: the entity folder contains reducers that mutate the state. Any state change leads to notification of all listeners. Maintaining object lists, filtering, or changing the state are the typical responsibilities of functions residing in the "entity" package.

Combining vanilla Web Components with lit-html and redux comes with the following advantage:

1. You can start with Web Components without learning any additional framework. Web Components are built-in into the browser, they are well documented, stable, and not expected to change. Migrations or upgrades are not necessary--you only have to learn them once and never migrate.
2. lit-html and redux are small utilities with several alternatives available.
3. BCE comes with a clear and consistent structure, which scales for large frontends.
4. The combination of Web Components, lit-html, and redux toolkit leads to code similar to React without any framework involved.
5. The necessary tooling is minimal. No transpilations, build tools, or bundling is required. Only a simple web server with hot-reload functionality like, e.g., https://browsersync.io, is recommended.

Check the following quickstarter, just "git clone && go." https://github.com/adambien/bce.design

See the BCE quickstarter in action:

The first airhacks.tv in 2021, live streamed from youtube.com/c/bienadam with the following topics:

2021 in review, killer use cases for Event Driven Architectures, EJB 2 to EJB 3 and EJB to CDI migrations, JPA alternatives, dynamic log configuration, DDOS prevention on AWS Lambda and AWS Fargate, Asynchronous Processing, JavaScript obfuscation, HTTP PATCH and Java 17, GlassFish vs. Payara, MicroProfile and JDBC

Any questions left? Ask now: https://gist.github.com/AdamBien/a0b7e3fb17efc4c2fff4208641c6bc68 and get the answers at the next airhacks.tv.

The method Collector#joining is overloaded with the parameters CharSequence prefix, CharSequence suffix which make List formatting easier:

import java.util.List;
import java.util.stream.Collectors;
import org.junit.jupiter.api.Test;


public class CollectorsJoiningTest {

    @Test
    public void withPrefixAndSuffix() {
        var words = List.of("java","duke","jvm");
        var message = words.stream().collect(Collectors.joining(",", "[", "]"));
        System.out.println(message);
    }
}

The unit test prints: [java,duke,jvm]

Questions and topics https://gist.github.com/AdamBien/2578630397cdab8f07e811627240fa5f for the 94th airhacks.tv live stream on youtube.com/c/bienadam/live:

1. 2021 in review
2. Event stores, event driven architectures, traditional CRUD and 2022 predictions
3. Migrating from EJB to CDI
4. Circular dependencies, bundling with JavaScript
5. JPA AttributeConverters and RequestScoped
6. Dynamically changing log level
7. DDOS prevention in the AWS clouds
8. Private vs. public clouds
9. Asynchronous processing, sessions and concurrency
10. Obfuscating JavaScript code
11. HTTP Patch and Java 17
12. MicroProfile and JDBC-based persistence

Any questions left? Ask now: gist.github.com/AdamBien/ea678489163f90a4bed43dbfeb4d97b6 and get the answers at the next airhacks.tv.

In 2013 I wrote a post: "In Case Dozer Or BeanUtils Are Working Well You Have A Problem." Most projects back then copied identical object hierarchies, usually persistent JPA entities and transient Data Transfer Objects, back and forth with the decoupling as the primary goal.

"The project is currently not active and will more than likely be deprecated in the future. If you are looking to use Dozer on a greenfield project, we would discourage that."

Dozer suggest the use of mapstruct or modelmapper frameworks instead.

However, consider removing any superfluous 1:1 mapping logic altogether before migrating dozer to another mapping framework. With Java Records, JSON-P, or Java Records and JSON-B you get interesting out-of-the-box alternatives to an external library and so dependency.

Subscribe to airhacks.fm podcast via: spotify| iTunes| RSS

The following Java "script" converts the src/main/java/airhacks/App.java script into an executable "Single-File Source-Code Program"

#!/usr/bin/java --source 17
import java.nio.file.Path;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.StandardOpenOption;

public class CreateJavaScript {

    public static void main(String ...args) throws IOException{
        var scriptDeclaration = "#!/usr/bin/java --source 17";
        var inputFile = Path.of("src/main/java/airhacks/App.java");
        var outputFile = Path.of("target","App.sh");

        var content = Files.readString(inputFile);
        var contentWithDeclaration = """
        $declaration
        $content
        """.replace("$declaration", scriptDeclaration)
            .replace("$content", content);
        Files.writeString(outputFile,contentWithDeclaration,StandardOpenOption.CREATE);
    }
}

This Java "script" ships with Maven Java 17 Quickstarter: https://github.com/AdamBien/java17-plain