This document will show you how to use Caffeine’s AsyncLoadingCache
in a Vert.x application.
You will build an application that rotates images in a web page.
The images will be downloaded by the server from a public API exposed at https://http.cat.
Each image represents an HTTP status code as a 🐱.
Ready?
Note
|
Images created by Tomomi Imura (@girlie_mac) |
The code of this project contains Maven and Gradle build files that are functionally equivalent.
Here is the content of the pom.xml
file you should be using:
pom.xml
link:pom.xml[role=include]
The index.html
web page consists mainly of:
-
an
<img>
tag in the body, and -
a script that, after the page is loaded, changes the
src
attribute of the image.
link:src/main/resources/webroot/index.html[role=include]
-
img
tag in the body withid
set tocat-img
-
run the script function when the page is loaded
-
define some HTTP status codes
-
schedule a function to execute periodically at a fixed-delay of 250 milliseconds
-
retrieve the
img
element from the DOM using its id -
update the
src
attribute using an HTTP status code chosen randomly
We will need a Vert.x Web Client instance to fetch images:
link:src/main/java/io/vertx/howtos/caffeine/CatsVerticle.java[role=include]
We will also need a cache because we do not want to overload the backend API:
link:src/main/java/io/vertx/howtos/caffeine/CatsVerticle.java[role=include]
-
create a cache builder
-
configure the cache to expire items after 1 minute
-
enable statistics recording
-
define an executor which invokes tasks on the verticle context
-
create an asynchronous loader, which much return a
CompletableFuture
, using the cache executor -
fetch the cat image
-
convert the Vert.x
Future
to aCompletionStage
-
log cache statistics periodically
Note
|
The executor definition and the complex loader implementation are not strictly needed here. Indeed, we will deploy a single verticle instance, bind the cache to a field and always invoke its methods from the event-loop. If that is your use-case, you may simplify the setup to: link:src/main/java/io/vertx/howtos/caffeine/CatsVerticle.java[role=include] If, however, you plan to deploy several instances of the verticle and to share the cache between them, stick to the previous implementation. It guarantees that the asynchronous loader is always invoked on the right context. |
Fetching the cat image consists in sending the request to the backend using the corresponding HTTP status code as URI:
link:src/main/java/io/vertx/howtos/caffeine/CatsVerticle.java[role=include]
Using Vert.x Web, creating the HTTP server for our API and static file is pretty straightforward:
link:src/main/java/io/vertx/howtos/caffeine/CatsVerticle.java[role=include]
Here is how we will implement image request handling:
link:src/main/java/io/vertx/howtos/caffeine/CatsVerticle.java[role=include]
-
retrieve the specified code from the request path
-
invoke Caffeine (the image will be loaded from the backend transparently, if needed)
-
convert the
CompletableFuture
returned by Caffeine to a Vert.xFuture
-
on completion, send the image bytes (or the failure) to the client
-
instruct the browser to disable caching of the image (otherwise, it would query our server only once for a given code!)
The CatsVerticle
needs a main
method:
link:src/main/java/io/vertx/howtos/caffeine/CatsVerticle.java[role=include]
-
Create a Vert.x instance
-
Deploy
CatsVerticle
You can run the application from:
-
your IDE, by running the
main
method from theCatsVerticle
class, or -
with Maven:
mvn compile exec:java
, or -
with Gradle:
./gradlew run
(Linux, macOS) orgradle run
(Windows).
Browse to http://localhost:8080.
You should see the cat images rotating in the web page:
After some time, inspect the program output. You should read something like:
Mar 22, 2022 3:45:17 PM io.vertx.howtos.caffeine.CatsVerticle lambda$start$4 INFO: Stats: CacheStats{hitCount=52, missCount=28, loadSuccessCount=28, loadFailureCount=0, totalLoadTime=2514949257, evictionCount=0, evictionWeight=0} Mar 22, 2022 3:45:37 PM io.vertx.howtos.caffeine.CatsVerticle lambda$start$4 INFO: Stats: CacheStats{hitCount=132, missCount=28, loadSuccessCount=28, loadFailureCount=0, totalLoadTime=2514949257, evictionCount=0, evictionWeight=0} Mar 22, 2022 3:45:57 PM io.vertx.howtos.caffeine.CatsVerticle lambda$start$4 INFO: Stats: CacheStats{hitCount=212, missCount=28, loadSuccessCount=28, loadFailureCount=0, totalLoadTime=2514949257, evictionCount=0, evictionWeight=0} Mar 22, 2022 3:46:17 PM io.vertx.howtos.caffeine.CatsVerticle lambda$start$4 INFO: Stats: CacheStats{hitCount=267, missCount=53, loadSuccessCount=52, loadFailureCount=0, totalLoadTime=3337599348, evictionCount=28, evictionWeight=28} Mar 22, 2022 3:46:37 PM io.vertx.howtos.caffeine.CatsVerticle lambda$start$4 INFO: Stats: CacheStats{hitCount=344, missCount=56, loadSuccessCount=56, loadFailureCount=0, totalLoadTime=3480880213, evictionCount=28, evictionWeight=28}
Notice the changes in hitCount
, missCount
, or evictionCount
.
This document covered:
-
the Vert.x web client for making HTTP requests,
-
the Vert.x web server and router,
-
the integration of Caffeine’s asynchronous loading cache in a Vert.x application,
-
Vert.x periodic tasks.