gRPC-Web for .NET is now officially released. We announced experimental support in January and since then we’ve been making improvements based on feedback from early adopters.

With this release gRPC-Web graduates to a fully supported component of the grpc-dotnet project and is ready for production. Use gRPC in the browser with gRPC-Web and .NET today.

Getting started

Developers who are brand new to gRPC should check out Create a gRPC client and server in ASP.NET Core. This tutorial walks through creating your first gRPC client and server using .NET.

If you already have a gRPC app then the Use gRPC in browser apps article shows how to add gRPC-Web to a .NET gRPC server.

What are gRPC and gRPC-Web

gRPC is a modern high-performance RPC (Remote Procedure Call) framework. gRPC is based on HTTP/2, Protocol Buffers and other modern standard-based technologies. gRPC is an open standard and is supported by many programming languages, including .NET.

It is currently impossible to implement the gRPC HTTP/2 spec in the browser because there are no browser APIs with enough fine-grained control over requests. gRPC-Web is a standardized protocol that solves this problem and makes gRPC usable in the browser. gRPC-Web brings many of gRPC’s great features, like small binary messages and contract-first APIs, to modern browser apps.

New opportunities with gRPC-Web

gRPC-Web is designed to make gRPC available in more scenarios. These include:

• Call ASP.NET Core gRPC apps from the browser – Browser APIs can’t call gRPC HTTP/2. gRPC-Web offers a compatible alternative.
  
  • JavaScript SPAs
  • .NET Blazor Web Assembly apps
• Host ASP.NET Core gRPC apps in IIS and Azure App Service – Some servers, such as IIS and Azure App Service, currently can’t host gRPC services. While this is actively being worked on, gRPC-Web offers an interesting alternative that works in every environment today.
• Call gRPC from non-.NET Core platforms – HTTP/2 is not supported by HttpClient on all .NET platforms. gRPC-Web can be used to call gRPC services from Blazor and Xamarin.

gRPC is a registered trademark of the Linux foundation. Blazor is compatible with gRPC-web

We’ve worked with the Blazor team to make gRPC-Web a great end-to-end developer experience when used in Blazor WebAssembly apps. Not only will gRPC tooling automatically generate strongly typed clients for you to call gRPC services from your Blazor app, but gRPC offers significant performance benefits over JSON.

A great example of the performance benefits in action is Blazor’s default template app. The data transferred on the fetch data page is halved when gRPC is used instead of JSON. Data size is reduced from 627 bytes down to 309 bytes.

The performance gain here comes from gRPC’s efficient binary serialization over traditional text-based JSON. gRPC-Web is an exciting opportunity to improve rich browser-based apps.

Try gRPC-Web with ASP.NET Core today

For more information about gRPC-Web, check out the documention, or try out a sample app that uses gRPC-Web.

gRPC-Web for .NET is out on NuGet now:

We look forward to seeing what you create with .NET, gRPC and now gRPC-Web!

Apple on Tuesday ceased signing code for iOS 13.6.1, effectively barring users who updated to the latest iOS 13.7 operating system from downgrading.

Bloomberg has reported that Nintendo is looking to increase Switch production, upping its previous total of 25 million units in 2020 to a whopping 30 million units.

According to the report, Nintendo boosted production in August to 25 million for the year in order to cope with demand triggered by the ongoing COVID-19 pandemic, which showcases just how successful Nintendo console has been during 2020.

While Nintendo’s release slate has been a little less busy than in previous years, the company has had a global smash hit in the form of Animal Crossing: New Horizons, a title which has sold 22 million copies so far (and that’s August’s figures – it’s no doubt even more now) and has caused a flood of new Switch hardware purchases.

Switch has now sold 61.44 million hardware units worldwide as of June 30th, according to Nintendo, and has already surpassed the lifetime sales of systems such as the NES, SNES, GameCube and N64. It is also rumoured that a Switch Pro model will launch in 2021, which will give Nintendo’s hybrid hardware even more momentum.

To sort and return a Python list in a single line of code, use the sorted(list) method that returns a new list of sorted elements. It copies only the references to the original elements so the returned list is not a deep but a shallow copy.

Let’s dive into the challenge to learn about more details and alternatives. Everything is not always simple. By studying the different methods that solves this, you’ll become a better coder!

Problem: Given a list of comparable objects such as integers or floats. Is there a way to sort the list and return the sorted list in a single line of Python code?

Example: Say, you’ve got the following list.

a = [4, 2, 1, 3]

You want to sort this list and return the result in a single line. If you use the list.sort() method, the return value is None:

print(a.sort())

# None

The return value of the list.sort() method is None, but many coders expect it to be the sorted list. So they’re surprised finding out that their variables contain the None type rather than a sorted list.

However, returning None makes perfect sense for the list.sort() method. Why? Because you call the method on a list object and it modifies this exact list object. It doesn’t create a new list—there won’t be a new list object in memory.

So, how to sort and return a list in a single line of Python code? As a rule of thumb, there are always multiple ways to accomplish the same thing in Python. Let’s dive into the different ways to accomplish this!

Here’s a quick overview of the methods addressed in this article:

Exercise: Change the list to be sorted by adding negative floats. Does it still work?

You’ll now learn more about each of the methods.

Method 1: sorted()

The easiest way to accomplish this task is to call Python’s built-in sorted() function that takes an iterable and returns a new list with sorted elements.

a = [4, 2, 1, 3] # Method 1: sorted()

print(sorted(a))

The sorted() function generates a new sorted list that is put into the print() function that prints the sorted list to the shell. The output is the sorted list:

[1, 2, 3, 4]

This method is the most Pythonic one. But are there alternatives?

Method 2: list.sort() + Ternary Operator

The sorted() method leaves the original list unchanged. But what if you want to sort the original list and get this original list as an output that you can assign to a variable?

The answer is simple: use a combination of the list.sort() method and the ternary operator!

a = [4, 2, 1, 3] # Method 2: list.sort() + ternary

print(a if a.sort() else a)

# [1, 2, 3, 4]

You need to understand two concepts: (1) list.sort() and (2) the ternary operator:

1. The list.sort() method sorts the list elements in place in an ascending manner. To customize the default sorting behavior, use the optional key argument by passing a function that returns a comparable value for each element in the list.
2. The ternary operator x if c else y consists of three operands x, c, and y. It returns x if the Boolean expression c evaluates to True. Otherwise, if the expression c evaluates to False, the ternary operator returns the alternative y.

The beautiful thing is that the one-liner print(a if a.sort() else a) modifies the original list and returns it right away. How does it do this?

Explanation: First, the a.sort() method is called to check which “branch” of the ternary operator should be visited. The return value of a.sort() will always be None. The None value is automatically converted to the Boolean False. Thus, the ternary operator always returns the list object referred to by variable a.

Note that the only purpose of the ternary operator is to make sure to call the a.sort() method before returning the value a—to make sure it is sorted!

If you print the original list to the shell, you’ll see that it is now sorted:

>>> print(a)

[1, 2, 3, 4]

Method 3: Combining Multiple Statements in a Single Line with Semicolon

An alternative is chaining the statements with a semicolon ; to one-linerize a Python code snippet. This strategy works with flat Python programs without, possible nested, blocks:

a = [4, 2, 1, 3] # Method 3: semicolon

a.sort(); print(a)

# [1, 2, 3, 4]

If you need to sort a Python list and print its return value to the shell—for example because you run this command from the command line or terminal—you can use this excellent strategy.

You can learn more about how to one-linerize any Python program in my following video:

Python One-Liners Book

Python programmers will improve their computer science skills with these useful one-liners.

Python One-Liners will teach you how to read and write “one-liners”: concise statements of useful functionality packed into a single line of code. You’ll learn how to systematically unpack and understand any line of Python code, and write eloquent, powerfully compressed Python like an expert.

The book’s five chapters cover tips and tricks, regular expressions, machine learning, core data science topics, and useful algorithms. Detailed explanations of one-liners introduce key computer science concepts and boost your coding and analytical skills. You’ll learn about advanced Python features such as list comprehension, slicing, lambda functions, regular expressions, map and reduce functions, and slice assignments. You’ll also learn how to:

•  Leverage data structures to solve real-world problems, like using Boolean indexing to find cities with above-average pollution
•  Use NumPy basics such as array, shape, axis, type, broadcasting, advanced indexing, slicing, sorting, searching, aggregating, and statistics
•  Calculate basic statistics of multidimensional data arrays and the K-Means algorithms for unsupervised learning
•  Create more advanced regular expressions using grouping and named groups, negative lookaheads, escaped characters, whitespaces, character sets (and negative characters sets), and greedy/nongreedy operators
•  Understand a wide range of computer science topics, including anagrams, palindromes, supersets, permutations, factorials, prime numbers, Fibonacci numbers, obfuscation, searching, and algorithmic sorting

By the end of the book, you’ll know how to write Python at its most refined, and create concise, beautiful pieces of “Python art” in merely a single line.

Get your Python One-Liners Now!!

Last modified on August 13th, 2020.

This is fairly a straightforward topic. Just a few lines of PHP code will do the work of getting the client user IP address. Most of the times it is true, it is all about a few lines of PHP script.

Sometimes things get tricky. We shall see about those things and also an other aspect like the privacy issue involved with it.

Following one line PHP code will get you the client user’s IP address. Then why all the fuss? Read through the article and you will know about it.

$_SERVER['REMOTE_ADDR'];

$_SERVER is a PHP array which is set by the server. Depending on this one line PHP code may not yield the best results in many a situation. For example, if a client is behind a proxy server, then this variable will not be suitable. Also be aware that, these headers can be easily spoofed by the users by setting an IP address themselves.

Following PHP script covers majority of the scenario and returns the user’s IP address.

<?php class Request

{ /** * Retrieves IP address set in the request header. * * Each ISPs sets them following their own logic. There is also a possibility for the user * to easily spoof their IP address. * * So using this for mission critical situations is not advisable. * If you are getting the IP address for casual logging purposes, then this is fine. */ public function getIpAddress() { $ipAddress = ''; if (! empty($_SERVER['HTTP_CLIENT_IP']) && $this->isValidIpAddress($_SERVER['HTTP_CLIENT_IP'])) { // check for shared ISP IP $ipAddress = $_SERVER['HTTP_CLIENT_IP']; } else if (! empty($_SERVER['HTTP_X_FORWARDED_FOR'])) { // check for IPs passing through proxy servers // check if multiple IP addresses are set and take the first one $ipAddressList = explode(',', $_SERVER['HTTP_X_FORWARDED_FOR']); foreach ($ipAddressList as $ip) { if ($this->isValidIpAddress($ip)) { $ipAddress = $ip; break; } } } else if (! empty($_SERVER['HTTP_X_FORWARDED']) && $this->isValidIpAddress($_SERVER['HTTP_X_FORWARDED'])) { $ipAddress = $_SERVER['HTTP_X_FORWARDED']; } else if (! empty($_SERVER['HTTP_X_CLUSTER_CLIENT_IP']) && $this->isValidIpAddress($_SERVER['HTTP_X_CLUSTER_CLIENT_IP'])) { $ipAddress = $_SERVER['HTTP_X_CLUSTER_CLIENT_IP']; } else if (! empty($_SERVER['HTTP_FORWARDED_FOR']) && $this->isValidIpAddress($_SERVER['HTTP_FORWARDED_FOR'])) { $ipAddress = $_SERVER['HTTP_FORWARDED_FOR']; } else if (! empty($_SERVER['HTTP_FORWARDED']) && $this->isValidIpAddress($_SERVER['HTTP_FORWARDED'])) { $ipAddress = $_SERVER['HTTP_FORWARDED']; } else if (! empty($_SERVER['REMOTE_ADDR']) && $this->isValidIpAddress($_SERVER['REMOTE_ADDR'])) { $ipAddress = $_SERVER['REMOTE_ADDR']; } return $ipAddress; } /** * To validate if an IP address is both a valid and does not fall within * a private network range. * * @access public * @param string $ip */ public function isValidIpAddress($ip) { if (filter_var($ip, FILTER_VALIDATE_IP, FILTER_FLAG_IPV4 | FILTER_FLAG_IPV6 | FILTER_FLAG_NO_PRIV_RANGE | FILTER_FLAG_NO_RES_RANGE) === false) { return false; } return true; } } 

There are many options given all through this article to get the client user’s IP address. This is the best among all the possibilities given, but I highly recommend you to read the section “IP address reliability” below.

Flowchart to get client user’s IP address

I have tried my best minimize the flowchart. Please do not look for diagrammatic accuracy. It is just an attempt to present the above PHP script in a flowchart model. But after drawing I realized that the above PHP script is easier to understand than this flowchart ?

Minimal code to get user’s IP address

Below one line code will work in most of the situation. Read the ‘IP address reliability’ section below, must. It will give you a good idea.

$ipAddress = $_SERVER['REMOTE_ADDR']?:($_SERVER['HTTP_X_FORWARDED_FOR']?:$_SERVER['HTTP_CLIENT_IP']); 

Get user’s IP address for CloudFlare hosted websites

I have taken CloudFlare as example. There are many similar services. You need to talk to that service provider to know which request header they are setting and get the IP address accordingly.

When the website is hosted using CloudFlare service, $_SERVER['REMOTE_ADDR'] will return the CloudFlare server’s IP address and not the user’s IP address. So in such a situation, you can make use of the variable set by the CloudFlare server to get the IP address.

if (isset($_SERVER["HTTP_CF_CONNECTING_IP"])) { $_SERVER['REMOTE_ADDR'] = $_SERVER["HTTP_CF_CONNECTING_IP"]; $ipAddress = $_SERVER['REMOTE_ADDR'];

} 

IP address reliability

$_SERVER['REMOTE_ADDR'] is the most reliable variable from the above list of variables. REMOTE_ADDR may not contain the real IP address of the TCP connection as well. It depends on the SAPI configuration.

If the user is behind a proxy server, HTTP_X_FORWARDED_FOR will be set. But this can be easily spoofed by the user by setting an IP address he wishes to expose.

Even the variable X-Forwarded-For or the Client-IP can be set by the user to any value he wishes. So, when you maintain a log of user’s IP address, it is better to store the IP address that is returned by $_SERVER['REMOTE_ADDR'] also.

Take an example of a shopping cart. You may display the product price according to the user’s country. In such a scenario, you may use the user’s IP address to determine the user’s country. In such a scenario, the code to get user’s IP address should be fool-proof.

So in summary, if you are building a mission critical application that depends on the user’s IP address, then this is not the way to go. For audit log of the user’s IP address, this may suffice. In essence, understand each line of these PHP code and variables used. Know your use-case well and use it accordingly.

IP address field length

When you are storing the IP address in the database, remember to have the field size as 45 for single IP. Because there is a possibility for IPv6 IP addresses. It its full form the IPv6 IP address can be 45 characters in length.

How to get the IP address of a host

Using a hostname, we can get the server’s IP address using the PHP’s built-in function gethostbyname. This returns the IPv4 address for the host given in argument.

$ipAddress = gethostbyname('www.example.com');

The reverse is possible too by using gethostbyaddr().

Client user’s IP address and privacy issues

Client’s IP address can be considered as private information in certain legislation. It depends on the land of law. Internet is public in general. You never know from which geographic region you will get users.

As per GDPR (Article 4, Point 1; and Recital 49), IP address is a personal data. I would suggest to inform the user and get consent to log the IP address. Also, call it out explicitly in your “Privacy Policy” document.

Explicitly state that you are collecting and logging the client user’s IP address and the reason for doing so.

Conclusion

Understand your use-case well. A good understanding of the scenario in which you are going to get and use the client user’s IP address is important. Then understand that the end user can easily spoof his IP address and present you something else. Third understand each line of the code, the server variables, from where the IP address is fetched, who sets it etc. Then decide how it will impact your use-case. Last but not the least, explicitly call out in your privacy policy that you are getting the client user’s IP address and storing it. Also state why you are doing it. As per GDPR, IP address is user’s personal information and so it would be better to get user’s consent before processing it.

↑ Back to Top

Quarkus is, in its own words, “Supersonic subatomic Java” and a “Kubernetes native Java stack tailored for GraalVM & OpenJDK HotSpot, crafted from the best of breed Java libraries and standards.” For the purpose of illustrating how to modernize an existing Java application to Quarkus, I will use the Red Hat JBoss Enterprise Application Platform (JBoss EAP) quickstarts helloworld quickstart as sample of a Java application builds using technologies (CDI and Servlet 3) supported in Quarkus.

It’s important to note that both Quarkus and JBoss EAP rely on providing developers with tools based—as much as possible—on standards. If your application is not already running on JBoss EAP, there’s no problem. You can migrate it from your current application server to JBoss EAP using the Red Hat Application Migration Toolkit. After that, the final and working modernized version of the code is available in the https://github.com/mrizzi/jboss-eap-quickstarts/tree/quarkus repository inside the helloworld module.

This article is based on the guides Quarkus provides, mainly Creating Your First Application and Building a Native Executable.

Get the code

To start, clone the JBoss EAP quickstarts repository locally, running:

$ git clone https://github.com/jboss-developer/jboss...starts.git Cloning into 'jboss-eap-quickstarts'... remote: Enumerating objects: 148133, done. remote: Total 148133 (delta 0), reused 0 (delta 0), pack-reused 148133 Receiving objects: 100% (148133/148133), 59.90 MiB | 7.62 MiB/s, done. Resolving deltas: 100% (66476/66476), done. $ cd jboss-eap-quickstarts/helloworld/

Try plain, vanilla helloworld

The name of the quickstart is a strong clue about what this application does, but let’s follow a scientific approach in modernizing this code, so first things first: Try the application as it is.

Deploy helloworld

1. Open a terminal and navigate to the root of the JBoss EAP directory EAP_HOME (which you can download).
2. Start the JBoss EAP server with the default profile by typing the following command:

$ EAP_HOME/bin/standalone.sh 

Note: For Windows, use the EAP_HOME\bin\standalone.bat script.

After a few seconds, the log should look like:

[org.jboss.as] (Controller Boot Thread) WFLYSRV0025: JBoss EAP 7.2.0.GA (WildFly Core 6.0.11.Final-redhat-00001) started in 3315ms - Started 306 of 527 services (321 services are lazy, passive or on-demand)

3. Open http://127.0.0.1:8080 in a browser, and a page like Figure 1 should appear:

4. Following instructions from Build and Deploy the Quickstart, deploy the helloworld quickstart and execute (from the project root directory) the command:

$ mvn clean install wildfly:deploy 

This command should end successfully with a log like this:

[INFO] ------------------------------------------------------------------------ [INFO] BUILD SUCCESS [INFO] ------------------------------------------------------------------------ [INFO] Total time: 8.224 s 

The helloworld application has now been deployed for the first time in JBoss EAP in about eight seconds.

Test helloworld

Following the Access the Application guide, open http://127.0.0.1:8080/helloworld in the browser and see the application page, as shown in Figure 2:

Make changes

Change the createHelloMessage(String name) input parameter from World to Marco (my ego is cheap):

writer.println("<h1>" + helloService.createHelloMessage("Marco") + "</h1>");

Execute again the command:

$ mvn clean install wildfly:deploy 

and then refresh the web page in the browser to check the message displayed changes, as shown in Figure 3:

Undeploy helloworld and shut down

If you want to undeploy (optional) the application before shutting down JBoss EAP, run the following command:

$ mvn clean install wildfly:undeploy 

To shut down the JBoss EAP instance, enter Ctrl+C in the terminal where it’s running.

Let’s modernize helloworld

Now we can leave the original helloworld behind and update it.

Create a new branch

Create a new working branch once the quickstart project finishes executing:

$ git checkout -b quarkus 7.2.0.GA 

Change the pom.xml file

The time has come to start changing the application. starting from the pom.xml file. From the helloworld folder, run the following command to let Quarkus add XML blocks:

$ mvn io.quarkus:quarkus-maven-plugin:0.23.2:create 

This article uses the 0.23.2 version. To know which is the latest version is, please refer to https://github.com/quarkusio/quarkus/releases/latest/, since the Quarkus release cycles are short.

This command changed the pom.xml, file adding:

• The property <quarkus.version> to define the Quarkus version to be used.
• The <dependencyManagement> block to import the Quarkus bill of materials (BOM). In this way, there’s no need to add the version to each Quarkus dependency.
• The quarkus-maven-plugin plugin responsible for packaging the application, and also providing the development mode.
• The native profile to create application native executables.

Further changes required to pom.xml, to be done manually:

1. Move the <groupId> tag outside of the <parent> block, and above the <artifactId> tag. Because we remove the <parent> block in the next step, the <groupId> must be preserved.
2. Remove the <parent> block: The application doesn’t need the JBoss parent pom anymore to run with Quarkus.
3. Add the <version> tag (below the <artifactId> tag) with the value you prefer.
4. Remove the <packaging> tag: The application won’t be a WAR anymore, but a plain JAR.
5. Change the following dependencies:
   
   1. Replace the javax.enterprise:cdi-api dependency with io.quarkus:quarkus-arc, removing <scope>provided</scope> because—as stated in the documentation—this Quarkus extension provides the CDI dependency injection.
   2. Replace the org.jboss.spec.javax.servlet:jboss-servlet-api_4.0_spec dependency with io.quarkus:quarkus-undertow, removing the <scope>provided</scope>, because—again as stated in the documentation—this is the Quarkus extension that provides support for servlets.
   3. Remove the org.jboss.spec.javax.annotation:jboss-annotations-api_1.3_spec dependency because it’s coming with the previously changed dependencies.

The pom.xml file’s fully changed version is available at https://github.com/mrizzi/jboss-eap-quickstarts/blob/quarkus/helloworld/pom.xml.

Note that the above mvn io.quarkus:quarkus-maven-plugin:0.23.2:create command, besides the changes to the pom.xml file, added components to the project. The added file and folders are:

• The files mvnw and mvnw.cmd, and .mvn folder: The Maven Wrapper allows you to run Maven projects with a specific version of Maven without requiring that you install that specific Maven version.
• The docker folder (in src/main/): This folder contains example Dockerfile files for both native and jvm modes (together with a .dockerignore file).
• The resources folder (in src/main/): This folder contains an empty application.properties file and the sample Quarkus landing page index.html (more in the section “Run the modernized helloworld“).

Run helloworld

To test the application, use quarkus:dev, which runs Quarkus in development mode (more details on Development Mode here).

Note: We expect this step to fail as changes are still required to the application, as detailed in this section.

Now run the command to check if and how it works:

$ ./mvnw compile quarkus:dev [INFO] Scanning for projects... [INFO] [INFO] ----------------< org.jboss.eap.quickstarts:helloworld >---------------- [INFO] Building Quickstart: helloworld quarkus [INFO] --------------------------------[ war ]--------------------------------- [INFO] [INFO] --- maven-resources-plugin:2.6:resources (default-resources) @ helloworld --- [INFO] Using 'UTF-8' encoding to copy filtered resources. [INFO] Copying 2 resources [INFO] [INFO] --- maven-compiler-plugin:3.1:compile (default-compile) @ helloworld --- [INFO] Nothing to compile - all classes are up to date [INFO] [INFO] --- quarkus-maven-plugin:0.23.2:dev (default-cli) @ helloworld --- Listening for transport dt_socket at address: 5005 INFO [io.qua.dep.QuarkusAugmentor] Beginning quarkus augmentation INFO [org.jbo.threads] JBoss Threads version 3.0.0.Final ERROR [io.qua.dev.DevModeMain] Failed to start quarkus: java.lang.RuntimeException: io.quarkus.builder.BuildException: Build failure: Build failed due to errors [error]: Build step io.quarkus.arc.deployment.ArcProcessor#validate threw an exception: javax.enterprise.inject.spi.DeploymentException: javax.enterprise.inject.UnsatisfiedResolutionException: Unsatisfied dependency for type org.jboss.as.quickstarts.helloworld.HelloService and qualifiers [@Default] - java member: org.jboss.as.quickstarts.helloworld.HelloWorldServlet#helloService - declared on CLASS bean [types=[javax.servlet.ServletConfig, java.io.Serializable, org.jboss.as.quickstarts.helloworld.HelloWorldServlet, javax.servlet.GenericServlet, javax.servlet.Servlet, java.lang.Object, javax.servlet.http.HttpServlet], qualifiers=[@Default, @Any], target=org.jboss.as.quickstarts.helloworld.HelloWorldServlet] at io.quarkus.arc.processor.BeanDeployment.processErrors(BeanDeployment.java:841) at io.quarkus.arc.processor.BeanDeployment.init(BeanDeployment.java:214) at io.quarkus.arc.processor.BeanProcessor.initialize(BeanProcessor.java:106) at io.quarkus.arc.deployment.ArcProcessor.validate(ArcProcessor.java:249) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:498) at io.quarkus.deployment.ExtensionLoader$1.execute(ExtensionLoader.java:780) at io.quarkus.builder.BuildContext.run(BuildContext.java:415) at org.jboss.threads.ContextClassLoaderSavingRunnable.run(ContextClassLoaderSavingRunnable.java:35) at org.jboss.threads.EnhancedQueueExecutor.safeRun(EnhancedQueueExecutor.java:2011) at org.jboss.threads.EnhancedQueueExecutor$ThreadBody.doRunTask(EnhancedQueueExecutor.java:1535) at org.jboss.threads.EnhancedQueueExecutor$ThreadBody.run(EnhancedQueueExecutor.java:1426) at java.lang.Thread.run(Thread.java:748) at org.jboss.threads.JBossThread.run(JBossThread.java:479) Caused by: javax.enterprise.inject.UnsatisfiedResolutionException: Unsatisfied dependency for type org.jboss.as.quickstarts.helloworld.HelloService and qualifiers [@Default] - java member: org.jboss.as.quickstarts.helloworld.HelloWorldServlet#helloService - declared on CLASS bean [types=[javax.servlet.ServletConfig, java.io.Serializable, org.jboss.as.quickstarts.helloworld.HelloWorldServlet, javax.servlet.GenericServlet, javax.servlet.Servlet, java.lang.Object, javax.servlet.http.HttpServlet], qualifiers=[@Default, @Any], target=org.jboss.as.quickstarts.helloworld.HelloWorldServlet] at io.quarkus.arc.processor.Beans.resolveInjectionPoint(Beans.java:428) at io.quarkus.arc.processor.BeanInfo.init(BeanInfo.java:371) at io.quarkus.arc.processor.BeanDeployment.init(BeanDeployment.java:206) ... 14 more 

It failed. Why? What happened?

The UnsatisfiedResolutionException exception refers to the HelloService class, which is a member of the HelloWorldServlet class (java member: org.jboss.as.quickstarts.helloworld.HelloWorldServlet#helloService). The problem is that HelloWorldServlet needs an injected instance of HelloService, but it can not be found (even if the two classes are in the very same package).

It’s time to return to Quarkus guides to leverage the documentation and understand how @Inject—and hence Contexts and Dependency Injection (CDI)—works in Quarkus, thanks to the Contexts and Dependency Injection guide. In the Bean Discovery paragraph, it says, “Bean classes that don’t have a bean defining annotation are not discovered.”

Looking at the HelloService class, it’s clear there’s no bean defining annotation, and one has to be added to have Quarkus to discover the bean. So, because it’s a stateless object, it’s safe to add the @ApplicationScoped annotation:

@ApplicationScoped public class HelloService { 

Note: The IDE should prompt you to add the required package shown here (add it manually if need be):

import javax.enterprise.context.ApplicationScoped; 

If you’re in doubt about which scope to apply when the original bean has no scope defined, please refer to the JSR 365: Contexts and Dependency Injection for Java 2.0—Default scope documentation.

Now, try again to run the application, executing again the ./mvnw compile quarkus:dev command:

$ ./mvnw compile quarkus:dev [INFO] Scanning for projects... [INFO] [INFO] ----------------< org.jboss.eap.quickstarts:helloworld >---------------- [INFO] Building Quickstart: helloworld quarkus [INFO] --------------------------------[ war ]--------------------------------- [INFO] [INFO] --- maven-resources-plugin:2.6:resources (default-resources) @ helloworld --- [INFO] Using 'UTF-8' encoding to copy filtered resources. [INFO] Copying 2 resources [INFO] [INFO] --- maven-compiler-plugin:3.1:compile (default-compile) @ helloworld --- [INFO] Changes detected - recompiling the module! [INFO] Compiling 2 source files to /home/mrizzi/git/forked/jboss-eap-quickstarts/helloworld/target/classes [INFO] [INFO] --- quarkus-maven-plugin:0.23.2:dev (default-cli) @ helloworld --- Listening for transport dt_socket at address: 5005 INFO [io.qua.dep.QuarkusAugmentor] (main) Beginning quarkus augmentation INFO [io.qua.dep.QuarkusAugmentor] (main) Quarkus augmentation completed in 576ms INFO [io.quarkus] (main) Quarkus 0.23.2 started in 1.083s. Listening on: http://0.0.0.0:8080 INFO [io.quarkus] (main) Profile dev activated. Live Coding activated. INFO [io.quarkus] (main) Installed features: [cdi] 

This time the application runs successfully.

Run the modernized helloworld

As the terminal log suggests, open a browser to http://0.0.0.0:8080(the default Quarkus landing page), and the page shown in Figure 4 appears:

This application has the following context’s definition in the WebServlet annotation:

@WebServlet("/HelloWorld") public class HelloWorldServlet extends HttpServlet { 

Hence, you can browse to http://0.0.0.0:8080/HelloWorldto reach the page shown in Figure 5:

It works!

Make changes

Please, pay attention to the fact that the ./mvnw compile quarkus:dev command is still running, and we’re not going to stop it. Now, try to apply the same—very trivial—change to the code and see how Quarkus improves the developer experience:

writer.println("<h1>" + helloService.createHelloMessage("Marco") + "</h1>");

Save the file, and then refresh the web page to check that Hello Marco appears, as shown in Figure 6:

Take time to check the terminal output:

INFO [io.qua.dev] (vert.x-worker-thread-3) Changed source files detected, recompiling [/home/mrizzi/git/forked/jboss-eap-quickstarts/helloworld/src/main/java/org/jboss/as/quickstarts/helloworld/HelloWorldServlet.java] INFO [io.quarkus] (vert.x-worker-thread-3) Quarkus stopped in 0.003s INFO [io.qua.dep.QuarkusAugmentor] (vert.x-worker-thread-3) Beginning quarkus augmentation INFO [io.qua.dep.QuarkusAugmentor] (vert.x-worker-thread-3) Quarkus augmentation completed in 232ms INFO [io.quarkus] (vert.x-worker-thread-3) Quarkus 0.23.2 started in 0.257s. Listening on: http://0.0.0.0:8080 INFO [io.quarkus] (vert.x-worker-thread-3) Profile dev activated. Live Coding activated. INFO [io.quarkus] (vert.x-worker-thread-3) Installed features: [cdi] INFO [io.qua.dev] (vert.x-worker-thread-3) Hot replace total time: 0.371s 

Refreshing the page triggered the source code change detection and the Quarkus automagic “stop-and-start.” All of this executed in just 0.371 seconds (that’s part of the Quarkus “Supersonic Subatomic Java” experience).

Build the helloworld packaged JAR

Now that the code works as expected, it can be packaged using the command:

$ ./mvnw clean package

This command creates two JARs in the /target folder. The first is helloworld-<version>.jar, which is the standard artifact built from the Maven command with the project’s classes and resources. The second is helloworld-<version>-runner.jar, which is an executable JAR.

Please pay attention to the fact that this is not an uber-jar, because all of the dependencies are copied into the /target/lib folder (and not bundled within the JAR). Hence, to run this JAR in another location or host, both the JAR file and the libraries in the /lib folder have to be copied, considering that the Class-Path entry of the MANIFEST.MF file in the JAR explicitly lists the JARs from the lib folder.

To create an uber-jar application, please refer to the Uber-Jar Creation Quarkus guide.

Run the helloworld packaged JAR

Now, the packaged JAR can be executed using the standard java command:

$ java -jar ./target/helloworld-<version>-runner.jar INFO [io.quarkus] (main) Quarkus 0.23.2 started in 0.673s. Listening on: http://0.0.0.0:8080 INFO [io.quarkus] (main) Profile prod activated. INFO [io.quarkus] (main) Installed features: [cdi] 

As done above, open the http://0.0.0.0:8080 URL in a browser, and test that everything works.

Build the helloworld quickstart-native executable

So far so good. The helloworld quickstart ran as a standalone Java application using Quarkus dependencies, but more can be achieved by adding a further step to the modernization path: Build a native executable.

Install GraalVM

First of all, the tools for creating the native executable have to be installed:

1. Download GraalVM 19.2.0.1 from https://github.com/oracle/graal/releases/tag/vm-19.2.0.1.
2. Untar the file using the command:

$ tar xvzf graalvm-ce-linux-amd64-19.2.0.1.tar.gz

3. Go to the untar folder.
4. Execute the following to download and add the native image component:

$ ./bin/gu install native-image

5. Set the GRAALVM_HOME environment variable to the folder created in step two, for example:

$ export GRAALVM_HOME={untar-folder}/graalvm-ce-19.2.0.1)

More details and install instructions for other operating systems are available in Building a Native Executable—Prerequisites Quarkus guide.

Build the helloworld native executable

As stated in the Building a Native Executable—Producing a native executable Quarkus guide, “Let’s now produce a native executable for our application. It improves the startup time of the application and produces a minimal disk footprint. The executable would have everything to run the application including the ‘JVM’ (shrunk to be just enough to run the application), and the application.”

To create the native executable, the Maven native profile has to be enabled by executing:

$ ./mvnw package -Pnative

The build took me about 1:10 minutes and the result is the helloworld-<version>-runner file in the /target folder.

Run the helloworld native executable

The /target/helloworld-<version>-runner file created in the previous step. It’s executable, so running it is easy:

$ ./target/helloworld-<version>-runner INFO [io.quarkus] (main) Quarkus 0.23.2 started in 0.006s. Listening on: http://0.0.0.0:8080 INFO [io.quarkus] (main) Profile prod activated. INFO [io.quarkus] (main) Installed features: [cdi] 

As done before, open the http://0.0.0.0:8080 URL in a browser and test that everything is working.

Next steps

I believe that this modernization, even of a basic application, is the right way to approach a brownfield application using technologies available in Quarkus. This way, you can start facing the issues and tackling them to understand and learn how to solve them.

In part two of this series, I’ll look at how to capture memory consumption data in order to evaluate performance improvements, which is a fundamental part of the modernization process.

The post Quarkus: Modernize “helloworld” JBoss EAP quickstart, Part 1 appeared first on Red Hat Developer.