tpr

Using Nette Tester with ProcessWire

Recommended Posts

This is a short tutorial on how to use Nette Tester with ProcessWire.

As you will see it's very easy to setup and use and it's perfect for testing your code's functionality. With bootstrapping ProcessWire it's also possible to check the rendered markup of pages using the API, checking page properties, etc. It's also a great tool for module developers for writing better code. 

While there will be nothing extraordinary here that you couldn't find in Tester's docs this can serve as a good starting point.

Prerequisites: PHP 5.6+

01 Download Tester

Go to https://github.com/nette/tester/releases and download the latest release (currently 2.0.2). Download from the link reading "Source code (zip)". You can use composer also if you wish.

02 Extract Tester files

Create a new directory in your site root called "tester". Extract the zip downloaded here, so it should look like this:

/site
/tester/src
/tester/tools
/tester/appveyor.yml
/tester/composer.json
/tester/contributing.md
/tester/license.md
/tester/readme.md
/wire
...

03 Create directory for test files

Add a new directory in "/tester" called "tests". Tester recognizes "*.Test.php" and "*.phpt" files in the tests directory, recursively. 

04 Create your first test

In the "tests" directory create a new "MyTest.php" file.

The first test is a very simple one that bootstraps ProcessWire and checks if the Home page name is "Home". This is not the smartest test but will show you the basics.

Add this to "/tester/tests/MyTest.php":

<?php namespace ProcessWire;

use \Tester\Assert;
use \Tester\DomQuery;
use \Tester\TestCase;
use \Tester\Environment;

require __DIR__ . '/../src/bootstrap.php'; // load Tester
require __DIR__ . '/../../index.php'; // bootstrap ProcessWire

Environment::setup();

class MyTest extends TestCase
{
    // first test (step 04)
    public function testHomeTitle()
    {
        $expected = 'Home'; // we expect the page title to be "Home"
        $actual = wire('pages')->get(1)->title; // check what's the actual title
        Assert::equal($expected, $actual); // check whether they are equal
    }

    // second test will go here (step 06)
    // third test will go here (step 07)
}

// run testing methods
(new MyTest())->run();

I've added comment placeholders for the second and third tests that we will insert later.

05 Run Tester

Tester can be run either from the command line or from the browser. The command line output is more verbose and colored while in the browser it's plain text only (see later).

Running from the command line

Navigate to the "/tester" directory in your console and execute this:

php src/tester.php -C tests

This will start "/tester/src/tester.php" and runs test files from the "/tester/tests" directory. The "-C" switch tells Tester to use the system-wide php ini file, that is required here because when bootstrapping ProcessWire you may run into errors (no php.ini file is used by default). You may load another ini file with the "-c <path>" (check the docs).

If the title of your Home page is "Home" you should see this:

tester-ok.png.7af030edf8d090a2d50c38ce87fa5d5a.png

If it's for example "Cats and Dogs", you should see this:

tester-failure.png.0e1a44f116602b4373221ddfacd4850d.png

Running from the browser

First we need to create a new PHP file in ProcessWire's root, let's call it "testrunner.php". This is because ProcessWire doesn't allow to run PHP files from its "site" directory.

The following code runs two test classes and produces a legible output. IRL you should probably iterate through directories to get test files (eg. with glob()), and of course it's better not allow tests go out to production.

<?php
ini_set('html_errors', false);
header('Content-type: text/plain');

echo 'Starting tests.' . PHP_EOL;
echo '--------------------------' . PHP_EOL;

$file = __DIR__ . '/PATH_TO/FirstTest.php';
echo basename($file) . ' ';
require $file;
echo '[OK]' . PHP_EOL;

$file = __DIR__ . '/PATH_TO/SecondTest.php';
echo basename($file) . ' ';
require $file;
echo '[OK]' . PHP_EOL;

echo '--------------------------' . PHP_EOL;
echo 'Tests finished.';

exit;

Navigate to "DOMAIN/testrunner.php" in your browser to execute the file. If every test succeeds you should get this:

tester-ok-browser.png.bd4dc2b5d69f1be26ea2b00eabdd286f.png

If there are failed tests the execution stops and you can read the error message. If there were more tests (eg. ThirdTest), those won't be displayed under the failed test.

tester-failure-browser.png.1a89ccae06b187ab56fa9482b61e9240.png

06 DOM test

This test will check if a page with "basic-page" template has a "h1" element. We will create the page on the fly with ProcessWire's API. To keep things simple we will add the new test as a new method to our MyTest class.

Add this block to the MyTest class:

public function testBasicPageHeadline()
{
    $p = new Page();
    $p->template = 'basic-page';
    $html = $p->render();
    $dom = DomQuery::fromHtml($html);

    Assert::true($dom->has('h1'));
}

This will most likely be true but of course you can check for something more specific, for example "div#main". Note that we have used the DomQuery helper here (check the "use" statement on the top of the file).

07 Custom function test

You will probably want to make sure your custom functions/methods will work as they should so let's write a test that demonstrates this.

I don't want to complicate things so I'll check if the built-in "pageName" sanitizer works as expected. Add this to the myTest class:

public function testPageNameSanitizer()
{
    $expected = 'hello-world';
    $actual = wire('sanitizer')->pageName('Hello world!', true);

    Assert::equal($expected, $actual);
}

This should also be true. Try to change the expected value if you are eager to see a failure message.
 
08 Next steps

You can add more methods to the MyTest class or create new files in the "tests" directory. Check out the range of available Assertions and other features in the docs and see how they could help you writing more fail-safe code.

Once you make a habit of writing tests you'll see how it can assist making your code more bulletproof and stable. Remember: test early, test often 🙂

If you find out something useful or cool with Tester make sure to share.

  • Like 13
  • Thanks 2

Share this post


Link to post
Share on other sites

It seems like nette is a bit less crufty than phpunit, which is nice. However I'm wondering how you're handling state in the database. E.g. how do you make sure if one tests creates a new page it's not affecting the next test?

Share this post


Link to post
Share on other sites

Honestly I haven't went that far yet but the closest I found are the setUp and tearDown methods (see the docs). Of course this creates holes in the autoincrement ids but I don't think that matters much. 

Method calls order
------------------
setUp()
testOne()
tearDown()

setUp()
testTwo()
tearDown()
  • Like 1

Share this post


Link to post
Share on other sites

When using setUp and tearDown methods it's good to keep in mind that Tester runs tests in parallel threads. That is, if you except that a Page you create in the setUp method will be deleted in tearDown before the next test method begins, you may be wrong.

For example I've created and saved a new Page with the same title in setUp and deleted it in tearDown. I randomly got a ProcessWire error saying it could not generate a unique name for the page, and that was because the other tests have been started before the page could be deleted in the tearDown method.

The actual thread number can be retrieved, so appending it to the title solved the issue (or by adding a random suffix):

$p->title = 'Test ' . getenv(\Tester\Environment::THREAD);

Alternatively you can reduce the number of threads to 1 witthe th "-j 1" switch (runtime will increase a lot).

  • Like 1

Share this post


Link to post
Share on other sites

I've updated the first post with info on how to run tests from the browser, so there's no need to have command line access to the server.

  • Like 4

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Similar Content

    • By abdus
      While I was reading Smashing Magazine, I discovered this handy tool called XRespond that lets you load webpages in iframes that are designed to simulate different devices, (similar to how browsers' mobile simulators work).
      I think it's quite useful for local development. In fact it works for any website as long as it doesnt have `X-Frame-Options` headers set to `sameorigin`. I highly recommend using it with browser-sync with live loading as well. I use this settings when using it with PHPStorm to live reload my dev site on pw.dev.
      // remember set cwd to /site/templates/ or call it from templates directory path/to/browser-sync.cmd start --proxy pw.dev --port 8080 --files "**/*.php" --files "assets/**/*.css" --files "assets/**/*.css"
      What other tools/utilities do you use for testing your responsive designs?
    • By Richard Jedlička
      Tense    
      Tense (Test ENvironment Setup & Execution) is a command-line tool to easily run tests agains multiple versions of ProcessWire CMF.
      Are you building a module, or a template and you need to make sure it works in all supported ProcessWire versions? Then Tense is exactly what you need. Write the tests in any testing framework, tell Tense which ProcessWire versions you are interested in and it will do the rest for you.

      See example or see usage in a real project.
      How to use?
      1. Install it: 
      composer global require uiii/tense 2. Create tense.yml config:
      tense init 3. Run it:
      tense run  
      For detailed instructions see Github page: https://github.com/uiii/tense
       
      This is made possible thanks to the great wireshell tool by @justb3a, @marcus and others.
       
      What do you think about it? Do you find it useful? Do you have some idea? Did you find some bug? Tell me you opinion. Write it here or in the issue tracker.
    • By hettiger
      I'm using GitLab CI for continuous integration right now. Other people might use something like Travis CI. For the frontend part things are pretty easy:
      image: node:6.10.3 cache: paths: - node_modules/ build: script: - npm install - node_modules/.bin/gulp build Now I could extend my setup to run tests or whatever I want. It just works.
      In this post I'm not interested into finding out how I could go about testing and stuff. I just want to know how I could accomplish the equivalent to the above setup task for ProcessWire. It doesn't seem to be too easy to me since the installation process is running in the browser asking you a lot of stuff. How could I go about it using the CLI?
      That's what I currently have:
      git clone git@github.com:processwire/processwire.git cd processwire mv site-blank site rm .gitignore git submodule add -b develop git@gitlab.local:path/to/MyModule.git site/modules/MyModule From here on I'm stuck.
      The only possible solution I see is maintaining a super repository including a database dump that has it's module dependencies defined in a .gitmodules file. This could then be installed using:
      git clone --recursive git@... Once that's completed all that needs to be done would be to create the database using the dump.
      Probably that's a pretty solid solution but it adds the overhead of having to maintain an additional repository +  database dump. I'm trying to find an alternative that's using the original processwire repository. That way I could just rerun my build once a new version get's merged into master and see if everything would be still working as well.
      I'm looking forward to your replies. Thanks!
    • By FrancisChung
      I'm writing this to give back something to the community that has given so much up front over the past year.
      I noticed there's hardly any discussion about testing in these forums so I decided to write this quick primer to get some discussion going.
      I'm by no means an expert on phpunit or selenium but I had to jump through a few hoops to get it working (especially with PHPStorm), so I thought I figured I should share my experiences with the community. 

      Also, I'm hoping non Phpstorm users can still  pick something up useful in this post.

      Prerequisites : It is assumed Phpunit (https://phpunit.de/) is installed via Composer,   Selenium (http://www.seleniumhq.org/) and Php-webdriver for Selenium (https://github.com/facebook/php-webdriver) is preinstalled.
      For Phpstorm users, there's a fairly detailed installation and unit testing instructions here (https://www.jetbrains.com/help/phpstorm/2016.1/testing.html)
      I found some parts of it leaving me with unanswered questions, so I'm hoping this post will supplement any questions that you might encounter.

      Rather than writing a single monolithic post, I will write several posts covering different topics.
    • By gebeer
      Hello,
      I'm in the process of building a web application with PW that delivers data to mobile clients.
      There will be up to 1000 requests per minute to my webapp (later maybe more). Every request triggers a search through up to 1000 pages and compares timestamps that are sent by the mobile clients with the request to timestamps that are saved with each page that is being searched.
      The timestamps are saved in PW in their own table in the DB together with a page reference id which makes searching pretty fast.
      For my search I use:
      $ads = $pages->find("template=advertisement, ad_server=$serverID, ad_publish_at.date<$tsHigh, ad_publish_at.date>$tsLow"); I want to do some load testing for my webapp to ensure it can handle that many requests per minute and further optimize it.
      What I need is a testing framework that lets me simulate hundreds of requests/minute.
      Have you ever done this and what testing framework would you use?
      Here are some apps that I briefly took a look at:
      http://jmeter.apache.org/
      http://www.pylot.org/
      https://code.google.com/p/httperf/
      https://github.com/JoeDog/siege