Quarkus - Qute Reference Guide

This technology is considered experimental.

In experimental mode, early feedback is requested to mature the idea. There is no guarantee of stability nor long term presence in the platform until the solution matures. Feedback is welcome on our mailing list or as issues in our GitHub issue tracker.

For a full list of possible extension statuses, check our FAQ entry.

1. Hello World Example

In this example, we’d like to demonstrate the basic workflow when working with Qute templates. Let’s start with a simple hello world example. We will always need some template contents:

  <p>Hello {name}! (1)
1 {name} is a value expression that is evaluated when the template is rendered.

Then, we will need to parse the contents into a template definition Java object. A template definition is an instance of io.quarkus.qute.Template.

If using Qute "standalone" you’ll need to create an instance of io.quarkus.qute.Engine first. The Engine represents a central point for template management with dedicated configuration. Let’s use the convenient builder:

Engine engine = Engine.builder().addDefaults().build();
In Quarkus, there is a preconfigured Engine available for injection - see Quarkus Integration.

If we have an Engine instance we could parse the template contents:

Template helloTemplate = engine.parse(helloHtmlContent);
In Quarkus, you can simply inject the template definition. The template is automatically parsed and cached - see Quarkus Integration.

Finally, we will create a template instance, set the data and render the output:

// Renders <html><p>Hello Jim!</p></html>
helloTemplate.data("name", "Jim").render(); (1)
1 Template.data(String, Object) is a convenient method that creates a template instance and sets the data in one step.

So the workflow is simple:

  1. Create template contents (hello.html),

  2. Parse template definition (io.quarkus.qute.Template),

  3. Create template instance (io.quarkus.qute.TemplateInstance),

  4. Render output.

The Engine is able to cache the definitions so that it’s not necessary to parse the contents again and again.

In Quarkus, the caching is done automatically.

2. Core Features

2.1. Syntax and Building Blocks

The dynamic parts of a template include:

  • Comment

    • Starts with {! and ends with !}: {! This is a comment !},

    • Could be multi-line,

    • May contain expressions and sections: {! {#if true} !}.

  • Expression

    • Outputs the evaluated value,

    • Simple properties: {foo}, {item.name},

    • Virtual methods: {item.get(name)}, {name ?: 'John'},

    • With namespace: {inject:colors}.

  • Section

    • May contain expressions and nested sections: {#if foo}{foo.name}{/if},

    • The name in the closing tag is optional: {#if active}ACTIVE!{/},

    • Can be empty: {#myTag image=true /},

    • May declare nested section blocks: {#if item.valid} Valid. {#else} Invalid. {/if} and decide which block to render.

  • Unparsed Character Data

    • Starts with {[ and ends with ]}: {[ <script>if(true){alert('Qute is cute!')};</script> ]},

    • Could be multi-line,

    • Used to mark the content that should be rendered but not parsed.

2.2. Identifiers and Tags

Identifiers are used in expressions and section tags. A valid identifier is a sequence of non-whitespace characters. However, users are encouraged to only use valid Java identifiers in expressions.

You can use bracket notation if you need to specify an identifier that contains a dot, e.g. {map['my.key']}.

When parsing a template document the parser identifies all tags. A tag starts and ends with a curly bracket, e.g. {foo}. The content of a tag must start with:

  • a digit, or

  • an alphabet character, or

  • underscore, or

  • a built-in command: #, !, @, /.

If it does not start with any of the above it is ignored by the parser.

Tag Examples
   {_foo.bar}   (1)
   {! comment !}(2)
   {  foo}      (3)
   {{foo}}      (4)
   {"foo":true} (5)
1 Parsed: an expression that starts with underscore.
2 Parsed: a comment
3 Ignored: starts with whitespace.
4 Ignored: starts with {.
5 Ignored: starts with ".
It is also possible to use escape sequences \{ and \} to insert delimiters in the text. In fact, an escape sequence is usually only needed for the start delimiter, ie. \{foo} will be rendered as {foo} (no parsing/evaluation will happen).

2.3. Removing Standalone Lines From the Template

By default, Qute parser removes standalone lines from the template output. A standalone line is a line that contains at least one section tag (e.g. {#each} and {/each}), parameter declaration (e.g. {@org.acme.Foo foo}) or comment but no expression and no non-whitespace character. In other words, a line that contains no section tag or a parameter declaration is not a standalone line. Likewise, a line that contains an expression or a non-whitespace character is not a standalone line.

Template Example
     {#for item in items} (1)
       <li>{item.name} {#if item.active}{item.price}{/if}</li>  (2)
     {/for}               (4)
1 This is a standalone line and will be removed.
2 Not a standalone line - contains an expression and non-whitespace characters
3 Not a standalone line - contains no section tag/parameter declaration
4 This is a standalone line.
Default Output
       <li>Foo 100</li>


The default behavior can be disabled by setting the property quarkus.qute.remove-standalone-lines to false. In this case, all whitespace characters from a standalone line will be printed to the output.

Output with quarkus.qute.remove-standalone-lines=false

       <li>Foo 100</li>


2.3.1. Expressions

An expression outputs a value. It consists of one or more parts separated by dot (dot notation) or square brackets (bracket notation). In the object.property (dot notation) syntax, the property must be a valid identifier>. In the object[property_name] (bracket notation) syntax, the property_name has to be a non-null <<literals, literal value. An expression could start with an optional namespace followed by a colon (:).

Expressions Example
{name} (1)
{item.name} (2)
{item['name']} (3)
{global:colors} (4)
1 no namespace, one part: name
2 no namespace, two parts: item, name
3 equivalent to {item.name} but using the bracket notation
4 namespace global, one part: colors

A part of an expression could be a virtual method in which case the name can be followed by a list of comma-separated parameters in parentheses. A parameter of a virtual method can be either a nested expression or a literal value.

Virtual Methods Example
{item.getLabels(1)} (1)
{name or 'John'} (2)
1 no namespace, two parts - item, getLabels(1), the second part is a virtual method with name getLabels and params 1
2 infix notation, translated to name.or('John'); no namespace, two parts - name, or('John') Supported Literals
Literal Examples


true, false




'value', "string"


1, -5


1l, -5L


1D, -5d


1f, -5F Resolution

The first part of the expression is always resolved against the current context object. If no result is found for the first part it’s resolved against the parent context object (if available). For an expression that starts with a namespace the current context object is found using all the available NamespaceResolvers. For an expression that does not start with a namespace the current context object is derived from the position of the tag. All other parts are resolved using ValueResolvers against the result of the previous resolution.

For example, expression {name} has no namespace and single part - name. The "name" will be resolved using all available value resolvers against the current context object. However, the expression {global:colors} has the namespace global and single part - colors. First, all available NamespaceResolvers will be used to find the current context object. And afterwards value resolvers will be used to resolve "colors" against the context object found. Current Context

If an expression does not specify a namespace the current context object is derived from the position of the tag. By default, the current context object represents the data passed to the template instance. However, sections may change the current context object. A typical example is the for/each loop - during iteration the content of the section is rendered with each element as the current context object:

{#each items}
  {count}. {it.name} (1)

{! Another form of iteration... !}
{#for item in items}
  {count}. {item.name} (2)
1 it is an implicit alias. name is resolved against the current iteration element.
2 Loop with an explicit alias item.

Data passed to the template instance are always accessible using the data namespace. This could be useful to access data for which the key is overridden:

{item.name} (1)
{#for item in item.getDerivedItems()} (2)
  {item.name} (3)
  is derived from
  {data:item.name} (4)
1 item is passed to the template instance as a data object.
2 Iterate over the list of derived items.
3 item is an alias for the iterated element.
4 Use the data namespace to access the item data object. Built-in Operators
Operator Description Examples


Outputs the default value if the previous part cannot be resolved or resolves to null.

{person.name ?: 'John'}, {person.name or 'John'}


Shorthand for if-then-else statement. Unlike in If Section nested operators are not supported.

{item.isActive ? item.name : 'Inactive item'} outputs the value of item.name if item.isActive resolves to true.

The condition in a ternary operator evaluates to true if the value is not considered falsy as described in the If Section.
In fact, the operators are implemented as "virtual methods" that consume one parameter and can be used with infix notation, i.e. {person.name or 'John'} is translated to {person.name.or('John')}. Character Escapes

For HTML and XML templates the ', ", <, >, & characters are escaped by default. If you need to render the unescaped value:

  1. Use the raw or safe properties implemented as extension methods of the java.lang.Object,

  2. Wrap the String value in a io.quarkus.qute.RawString.

<h1>{title}</h1> (1)
{paragraph.raw} (2)
1 title that resolves to Expressions & Escapes will be rendered as Expressions &amp; Escapes
2 paragraph that resolves to <p>My text!</p> will be rendered as <p>My text!</p> Virtual Methods

A virtual method is a part of an expression that looks like a regular Java method invocation. It’s called "virtual" because it does not have to match the actual method of a Java class. In fact, like normal properties a virtual method is also handled by a value resolver. The only difference is that for virtual methods a value resolver consumes parameters that are also expressions.

Virtual Method Example
<h1>{item.buildName(item.name,5)}</h1> (1)
1 buildName(item.name,5) represents a virtual method with name buildName and two parameters: item.name and 5 . The virtual method could be evaluated by a value resolver generated for the following Java class:
class Item {
   String buildName(String name, int age) {
      return name + ":" + age;
Virtual methods are usually evaluated by value resolvers generated for @TemplateExtension methods, @TemplateData or classes used in parameter declarations. However, a custom value resolver that is not backed by any Java class/method can be registered as well.

A virtual method with single parameter can be called using the infix notation:

Infix Notation Example
<p>{item.price or 5}</p>  (1)
1 item.price or 5 is translated to item.price.or(5).

Virtual method parameters can be "nested" virtual method invocations.

Nested Virtual Method Example
<p>{item.subtractPrice(item.calculateDiscount(10))}</p>  (1)
1 item.calculateDiscount(10) is evaluated first and then passed as an argument to item.subtractPrice().

2.3.2. Sections

A section:

  • has a start tag

    • starts with #, followed by the name of the section such as {#if} and {#each},

  • may be empty

    • tag ends with /, ie. {#emptySection /}

  • may contain other expression, sections, etc.

    • the end tag starts with / and contains the name of the section (optional): {#if foo}Foo!{/if} or {#if foo}Foo!{/},

The start tag can also define parameters. The parameters have optional names. A section may contain several content blocks. The "main" block is always present. Additional/nested blocks also start with # and can have parameters too - {#else if item.isActive}. A section helper that defines the logic of a section can "execute" any of the blocks and evaluate the parameters.

{#if item.name is 'sword'}
  It's a sword!
{#else if item.name is 'shield'}
  It's a shield!
  Item is neither a sword nor a shield.
{/if} Loop Section

The loop section makes it possible to iterate over an instance of Iterable, Map 's entry set, Stream and an Integer. It has two flavors. The first one is using the each name alias.

{#each items}
  {it.name} (1)
1 it is an implicit alias. name is resolved against the current iteration element.

The other form is using the for name alias and can specify the alias used to reference the iteration element:

{#for item in items}

It’s also possible to access the iteration metadata inside the loop:

{#each items}
  {count}. {it.name} (1)
1 count represents one-based index. Metadata also include zero-based index, hasNext, odd, even.

The for statement also works with integers, starting from 1. In the example below, considering that total = 3:

{#for i in total}

The output will be:

1:2:3: If Section

The if section represents a basic control flow section. The simplest possible version accepts a single parameter and renders the content if the condition is evaluated to true. A condition without an operator evaluates to true if the value is not considered falsy, i.e. if the value is not null, false, an empty collection, an empty map, an empty array, an empty string/char sequence or a number equal to zero.

{#if item.active}
  This item is active.

You can also use the following operators in a condition:

Operator Aliases Precedence (higher wins)

logical complement



greater than

gt, >


greater than or equal to

ge, >=


less than

lt, <


less than or equal to

le, <=



eq, ==, is


not equals

ne, !=


logical AND (short-circuiting)

&&, and


logical OR (short-circuiting)

||, or


A simple operator example
{#if item.age > 10}
  This item is very old.

Multiple conditions are also supported.

Multiple conditions example
{#if item.age > 10 && item.price > 500}
  This item is very old and expensive.

Precedence rules can be overridden by parentheses.

Parentheses example
{#if (item.age > 10 || item.price > 500) && user.loggedIn}
  User must be logged in and item age must be > 10 or price must be > 500.

You can also add any number of else blocks:

{#if item.age > 10}
  This item is very old.
{#else if item.age > 5}
  This item is quite old.
{#else if item.age > 2}
  This item is old.
  This item is not old at all!
{/if} With Section

This section can be used to set the current context object. This could be useful to simplify the template structure:

{#with item.parent}
  <h1>{name}</h1>  (1)
  <p>{description}</p> (2)
1 The name will be resolved against the item.parent.
2 The description will be also resolved against the item.parent.

This section might also come in handy when we’d like to avoid multiple expensive invocations:

{#with item.callExpensiveLogicToGetTheValue(1,'foo',bazinga)}
  {#if this is "fun"} (1)
    <h1>{this} is not fun at all!</h1>
1 this is the result of item.callExpensiveLogicToGetTheValue(1,'foo',bazinga). The method is only invoked once even though the result may be used in multiple expressions. Let/Set Section

This section allows you to define named local variables:

{#let myParent=order.item.parent}

The section tag is also registered under the set alias:

{#set myParent=item.parent price=item.price}
  <p>Price: {price}
{/set} Include Section

This section can be used to include another template and possibly override some parts of the template (template inheritance).

Simple Example
<meta charset="UTF-8">
<title>Simple Include</title>
  {#include foo limit=10 /} (1)(2)
1 Include a template with id foo. The included template can reference data from the current context.
2 It’s also possible to define optional parameters that can be used in the included template.

Template inheritance makes it possible to reuse template layouts.

Template "base"
<meta charset="UTF-8">
<title>{#insert title}Default Title{/}</title> (1)
  {#insert body}No body!{/} (2)
1 insert sections are used to specify parts that could be overridden by a template that includes the given template.
2 An insert section may define the default content that is rendered if not overridden.
Template "detail"
{#include base} (1)
  {#title}My Title{/title} (2)
      My body.
1 include section is used to specify the extended template.
2 Nested blocks are used to specify the parts that should be overridden.
Section blocks can also define an optional end tag - {/title}. User-defined Tags

User-defined tags can be used to include a template and optionally pass some parameters. Let’s suppose we have a template called itemDetail.html:

{#if showImage} (1)
  {it.image} (2)
  {nested-content} (3)
1 showImage is a named parameter.
2 it is a special key that is replaced with the first unnamed param of the tag.
3 (optional) nested-content is a special key that will be replaced by the content of the tag.

Now if we register this template under the name itemDetail.html and if we add a UserTagSectionHelper to the engine:

Engine engine = Engine.builder()
                   .addSectionHelper(new UserTagSectionHelper.Factory("itemDetail","itemDetail.html"))
In Quarkus, all files from the src/main/resources/templates/tags are registered and monitored automatically!

We can include the tag like this:

{#for item in items}
  {#itemDetail item showImage=true} (1)
    = <b>{item.name}</b> (2)
1 item is resolved to an iteration element and can be referenced using the it key in the tag template.
2 Tag content injected using the nested-content key in the tag template.

2.4. Engine Configuration

2.4.1. Template Locator

Manual registration is sometimes handy but it’s also possible to register a template locator using EngineBuilder.addLocator(Function<String, Optional<Reader>>). This locator is used whenever the Engine.getTemplate() method is called and the engine has no template for a given id stored in the cache.

In Quarkus, all templates from the src/main/resources/templates are located automatically.

3. Quarkus Integration

If you want to use Qute in your Quarkus application add the following dependency to your project:


In Quarkus, a preconfigured engine instance is provided and available for injection - a bean with scope @ApplicationScoped, bean type io.quarkus.qute.Engine and qualifier @Default is registered automatically. Moreover, all templates located in the src/main/resources/templates directory are validated and can be easily injected.

import io.quarkus.qute.Engine;
import io.quarkus.qute.Template;
import io.quarkus.qute.api.ResourcePath;

class MyBean {

    Template items; (1)

    @ResourcePath("detail/items2_v1.html") (2)
    Template items2;

    Engine engine; (3)
1 If there is no ResourcePath qualifier provided, the field name is used to locate the template. In this particular case, the container will attempt to locate a template with path src/main/resources/templates/items.html.
2 The ResourcePath qualifier instructs the container to inject a template from a path relative from src/main/resources/templates. In this case, the full path is src/main/resources/templates/detail/items2_v1.html.
3 Inject the configured Engine instance.

3.1. Template Variants

Sometimes it’s useful to render a specific variant of the template based on the content negotiation. This can be done by setting a special attribute via TemplateInstance.setAttribute():

class MyService {

    Template items; (1)

    ItemManager manager;

    String renderItems() {
       return items.data("items",manager.findItems()).setAttribute(TemplateInstance.SELECTED_VARIANT, new Variant(Locale.getDefault(),"text/html","UTF-8")).render();
When using quarkus-resteasy-qute the content negotiation is performed automatically. See RESTEasy Integration.

3.2. Injecting Beans Directly In Templates

A CDI bean annotated with @Named can be referenced in any template through the inject namespace:

{inject:foo.price} (1)
1 First, a bean with name foo is found and then used as the base object.

All expressions using the inject namespace are validated during build. For the expression inject:foo.price the implementation class of the injected bean must either have the price property (e.g. a getPrice() method) or a matching template extension method must exist.

A ValueResolver is also generated for all beans annotated with @Named so that it’s possible to access its properties without reflection.

3.3. Parameter Declarations

It is possible to specify optional parameter declarations in a template. Quarkus attempts to validate all expressions that reference such parameters. If an invalid/incorrect expression is found the build fails.

{@org.acme.Foo foo} (1)
<!DOCTYPE html>
<meta charset="UTF-8">
<title>Qute Hello</title>
  <h1>{title}</h1> (2)
  Hello {foo.message}! (3)
1 Parameter declaration - maps foo to org.acme.Foo.
2 Not validated - not matching a param declaration.
3 This expression is validated. org.acme.Foo must have a property message or a matching template extension method must exist.
A value resolver is also generated for all types used in parameter declarations so that it’s possible to access its properties without reflection.

3.3.1. Overriding Parameter Declarations

{@org.acme.Foo foo}
<!DOCTYPE html>
<meta charset="UTF-8">
<title>Qute Hello</title>
  <h1>{foo.message}</h1> (1)
  {#for foo in baz.foos}
    <p>Hello {foo.message}!</p> (2)
1 Validated against org.acme.Foo.
2 Not validated - foo is overridden in the loop section.

3.4. Template Extension Methods

Extension methods can be used to extend the data classes with new functionality or resolve expressions for a specific namespace. For example, it is possible to add computed properties and virtual methods.

A value resolver is automatically generated for a method annotated with @TemplateExtension. If a class is annotated with @TemplateExtension then a value resolver is generated for every non-private static method declared on the class. Method-level annotations override the behavior defined on the class. Methods that do not meet the following requirements are ignored.

A template extension method:

  • must not be private

  • must be static,

  • must not return void,

  • must accept at least one parameter, unless the namespace is specified.

The class of the first parameter is used to match the base object unless the namespace is specified. In such case, the namespace is used to match an expression.

The method name is used to match the property name by default. However, it is possible to specify the matching name with TemplateExtension#matchName(). A special constant - TemplateExtension#ANY - may be used to specify that the extension method matches any name. It is also possible to match the name against a regular expression specified in TemplateExtension#matchRegex(). In both cases, a string method parameter is used to pass the property name. If both matchName() and matchRegex() are set the regular expression is used for matching.

If a namespace is specified the method must declare at least one parameter and the first parameter must be a string. If no namespace is specified the method must declare at least two parameters and the second parameter must be a string.

Extension Method Example
package org.acme;

class Item {

    public final BigDecimal price;

    public Item(BigDecimal price) {
        this.price = price;

class MyExtensions {

    static BigDecimal discountedPrice(Item item) { (1)
        return item.getPrice().multiply(new BigDecimal("0.9"));
1 This method matches an expression with base object of the type Item.class and the discountedPrice property name.

This template extension method makes it possible to render the following template:

{item.discountedPrice} (1)
1 item is resolved to an instance of org.acme.Item.

3.4.1. Method Parameters

An extension method may accept multiple parameters. If no namespace is specified the first parameter is always used to pass the base object, i.e. org.acme.Item in the first example. Other parameters are resolved when rendering the template and passed to the extension method.

Multiple Parameters Example
class BigDecimalExtensions {

    static BigDecimal scale(BigDecimal val, int scale, RoundingMode mode) { (1)
        return val.setScale(scale, mode);
1 This method matches an expression with base object of the type BigDecimal.class, with the scale virtual method name and two virtual method parameters.
{item.discountedPrice.scale(2,mode)} (1)
1 item.discountedPrice is resolved to an instance of BigDecimal.

3.4.2. Namespace Extention Methods

If TemplateExtension#namespace() is specified then the extension method is used to resolve expressions with the given namespace. Template extension methods that share the same namespace are grouped in one resolver ordered by TemplateExtension#priority(). The first matching extension method is used to resolve an expression.

Namespace Extension Method Example
@TemplateExtension(namespace = "str")
public static class StringExtensions {

   static String format(String fmt, Object... args) {
      return String.format(fmt, args);

   static String reverse(String val) {
      return new StringBuilder(val).reverse().toString();

These extension methods can be used as follows.

{str:format('%s %s!','Hello', 'world')} (1)
{str:reverse('hello')} (2)
1 The output is Hello world!
2 The output is olleh

3.4.3. Built-in Template Extension

Quarkus provides a set of built-in extension methods. Maps
  • keys or keySet: Returns a Set view of the keys contained in a map

    • {#for key in map.keySet}

  • values: Returns a Collection view of the values contained in a map

    • {#for value in map.values}

  • size: Returns the number of key-value mappings in a map

    • {map.size}

  • isEmpty: Returns true if a map contains no key-value mappings

    • {#if map.isEmpty}

  • get(key): Returns the value to which the specified key is mapped

    • {map.get('foo')}

A map value can be also accessed directly: {map.myKey}. Use the bracket notation for keys that are not legal identifiers: {map['my key']}. Collections
  • get(index): Returns the element at the specified position in a list

    • {list.get(0)}

A list element can be accessed directly: {list.10} or {list[10]}. Numbers
  • mod: Modulo operation

    • {#if counter.mod(5) == 0} Config
  • config:<name> or config:[<name>]: Returns the config value for the given property name

    • {config:foo} or {config:['property.with.dot.in.name']}

  • config:property(name): Returns the config value for the given property name; the name can be obtained dynamically by an expression

    • {config:property('quarkus.foo')} or {config:property(foo.getPropertyName())}

3.5. @TemplateData

A value resolver is automatically generated for a type annotated with @TemplateData. This allows Quarkus to avoid using reflection to access the data at runtime.

Non-public members, constructors, static initializers, static, synthetic and void methods are always ignored.
package org.acme;

class Item {

    public final BigDecimal price;

    public Item(BigDecimal price) {
        this.price = price;

    public BigDecimal getDiscountedPrice() {
        return price.multiply(new BigDecimal("0.9"));

Any instance of Item can be used directly in the template:

{#each items} (1)
  {it.price} / {it.discountedPrice}
1 items is resolved to a list of org.acme.Item instances.

Furthermore, @TemplateData.properties() and @TemplateData.ignore() can be used to fine-tune the generated resolver. Finally, it is also possible to specify the "target" of the annotation - this could be useful for third-party classes not controlled by the application:

@TemplateData(target = BigDecimal.class)
class Item {

    public final BigDecimal price;

    public Item(BigDecimal price) {
        this.price = price;
{#each items} (1)
  {it.price.setScale(2, rounding)} (1)
1 The generated value resolver knows how to invoke the BigDecimal.setScale() method.

3.6. RESTEasy Integration

If you want to use Qute in your JAX-RS application, you’ll need to add the quarkus-resteasy-qute extension first. In your pom.xml file, add:


This extension registers a special ContainerResponseFilter implementation so that a resource method can return a TemplateInstance and the filter takes care of all necessary steps. A simple JAX-RS resource may look like this:

package org.acme.quarkus.sample;

import javax.inject.Inject;
import javax.ws.rs.GET;
import javax.ws.rs.Path;
import javax.ws.rs.QueryParam;

import io.quarkus.qute.TemplateInstance;
import io.quarkus.qute.Template;

public class HelloResource {

    Template hello; (1)

    public TemplateInstance get(@QueryParam("name") String name) {
        return hello.data("name", name); (2) (3)
1 If there is no @ResourcePath qualifier provided, the field name is used to locate the template. In this particular case, we’re injecting a template with path templates/hello.txt.
2 Template.data() returns a new template instance that can be customized before the actual rendering is triggered. In this case, we put the name value under the key name. The data map is accessible during rendering.
3 Note that we don’t trigger the rendering - this is done automatically by a special ContainerResponseFilter implementation.

The content negotiation is performed automatically. The resulting output depends on the Accept header received from the client.

class DetailResource {

    Template item; (1)

    @Produces({ MediaType.TEXT_HTML, MediaType.TEXT_PLAIN })
    public TemplateInstance item() {
        return item.data("myItem", new Item("Alpha", 1000)); (2)
1 Inject a variant template with base path derived from the injected field - src/main/resources/templates/item.
2 For text/plain the src/main/resources/templates/item.txt template is used. For text/html the META-INF/resources/templates/item.html template is used.

3.7. Development Mode

In the development mode, all files located in src/main/resources/templates are watched for changes and modifications are immediately visible.

3.8. Type-safe Message Bundles

3.8.1. Basic Concepts

The basic idea is that every message is potentially a very simple template. In order to prevent type errors a message is defined as an annotated method of a message bundle interface. Quarkus generates the message bundle implementation at build time. Subsequently, the bundles can be used at runtime:

  1. Directly in your code via io.quarkus.qute.i18n.MessageBundles#get(); e.g. MessageBundles.get(AppMessages.class).hello_name("Lucie")

  2. Injected in your beans via @Inject; e.g. @Inject AppMessages

  3. Referenced in the templates via the message bundle namespace:

     {msg:hello_name('Lucie')} (1) (2) (3)
     {msg:message(myKey,'Lu')} (4)
    1 msg is the default namespace.
    2 hello_name is the message key.
    3 Lucie is the parameter of the message bundle interface method.
    4 It is also possible to obtain a localized message for a key resolved at runtime using a reserved key message. The validation is skipped in this case though.
Message Bundle Interface Example
import io.quarkus.qute.i18n.Message;
import io.quarkus.qute.i18n.MessageBundle;

@MessageBundle (1)
public interface AppMessages {

    @Message("Hello {name}!") (2)
    String hello_name(String name); (3)
1 Denotes a message bundle interface. The bundle name is defaulted to msg and is used as a namespace in templates expressions, e.g. {msg:hello_name}.
2 Each method must be annotated with @Message. The value is a qute template.
3 The method parameters can be used in the template.

3.8.2. Bundle Name and Message Keys

Keys are used directly in templates. The bundle name is used as a namespace in template expressions. The @MessageBundle can be used to define the default strategy used to generate message keys from method names. However, the @Message can override this strategy and even define a custom key. By default, the annotated element’s name is used as-is. Other possibilities are:

  1. De-camel-cased and hyphenated; e.g. helloName()hello-name

  2. De-camel-cased and parts separated by underscores; e.g. helloName()hello_name.

3.8.3. Validation

  • All message bundle templates are validated:

    • All expressions without a namespace must map to a parameter; e.g. Hello {foo} → the method must have a param of name foo

    • All expressions are validated against the types of the parameters; e.g. Hello {foo.bar} where the parameter foo is of type org.acme.Fooorg.acme.Foo must have a property of name bar

      A warning message is logged for each unused parameter.
  • Expressions that reference a message bundle method, such as {msg:hello(item.name)}, are validated too.

3.8.4. Localization

The default locale of the Java Virtual Machine used to build the application is used for the @MessageBundle interface by default. However, the io.quarkus.qute.i18n.MessageBundle#locale() can be used to specify a custom locale. Additionaly, there are two ways to define a localized bundle:

  1. Create an interface that extends the default interface that is annotated with @Localized

  2. Create an UTF-8 encoded file located in src/main/resources/messages; e.g. msg_de.properties.

A localized interface is the preferred solution mainly due to the possibility of easy refactoring.
Localized Interface Example
import io.quarkus.qute.i18n.Localized;
import io.quarkus.qute.i18n.Message;

@Localized("de") (1)
public interface GermanAppMessages {

    @Message("Hallo {name}!") (2)
    String hello_name(String name);
1 The value is the locale tag string (IETF).
2 The value is the localized template.

Message bundle files must be encoded in UTF-8. The file name consists of the relevant bundle name (e.g. msg) and underscore followed by the locate tag (IETF). The file format is very simple: each line represents either a key/value pair with the equals sign used as a separator or a comment (line starts with #). Keys are mapped to method names from the corresponding message bundle interface. Values represent the templates normally defined by io.quarkus.qute.i18n.Message#value(). We use .properties suffix in our example because most IDEs and text editors support syntax highlighting of .properties files. But in fact, the suffix could be anything.

Localized File Example - msg_de.properties
hello_name=Hallo {name}! (1) (2)
1 Each line in a localized file represents a message template.
2 Keys and values are separated by the equals sign.

3.9. Configuration Reference

Configuration property fixed at build time - All other configuration properties are overridable at runtime

Configuration property



The set of suffixes used when attempting to locate a template file.

By default, engine.getTemplate("foo") would result in several lookups: foo, foo.html, foo.txt, etc.

list of string


Specify whether the parser should remove standalone lines from the output. A standalone line is a line that contains at least one section tag, parameter declaration, or comment but no expression and no non-whitespace character.



The additional map of suffixes to content types. This map is used when working with template variants. By default, the java.net.URLConnection#getFileNameMap() is used to determine the content type of a template file.