I'm not using the

runner.cfm

I want to use the

test-browser

that lets you move up and down your tests directory to find the actual test cfc you want to run and then it opens a new window just for that cfc which, for me, is showing an empty page

yeah. that's what I do to. And the cfc is the file in the URL, and if I run that page without

method=runRemote

then it runs my tests, and displays the test report.

Hummm, I'm not sure how that would run? My understanding is that by having your tests extend

testbox.system.BaseSpec

it gets access to the

runRemote

method which you add to your url, that then handles the processing and presentation of the test results.

Hi, I am fairly new to TestBox and Coldbox, but I run into an issue which I don’t understand. I have written a BDD test which tests some Coldbox code. The Coldbox code makes use of a Service. Services need a reinit when you change some code, so every time I run my test, I need the Coldbox application to reinit. The strange thing is, when I run my test on the commandbox cli, a reinit takes place automatically since every change in my Service is visible immediately. However when I run the same test in the browser, a reinit does not take place. I have to do a reinit manually first and then I can run my test in the browser. The URL called from the commandbox CLI is: http://myhost/tests/runner.cfm?&labels=mytest&recurse=true&reporter=json&verbose=false The URL called from the browser is: http://myhost/tests/specs/integration/forms/StoreFormDataTest.cfc?method=runRemote Both run the exact same test, but the first one does a reinit and the second one does not. How can I get the test from the browser to initiate a reinit before running the test?

Ohhhh cbPlaywright - this looks interesting 🙂

Good morning. Hope a good weekend was had by all. After a few false starts, I feel like I'm actually and finally building up some momentum with testing. Eric Peterson gave a great presentation on cbPlaywright for end-to-end testing at ITB, which was the match that lit the fuse. I've been spending as much time as I can working on various kinds of testing since then. I have a fairly decent handle on end-to-end tests with cbPlaywright (seriously, what a great tool), and I have a decent-ish (but growing) handle on unit tests and mocking with Testbox. So right now I am working on integration tests with Testbox. I'm trying to test a typical form submission and the associated results on a results page. To pass in all the form fields for my integration tests I'm doing something like

getRequestContext().setValue( "employeeName", "Joseph Chastain" );

before the

execute

call. I know I can get the rendered results page and check for various things with

expect( event.getRenderedContent() ).toInclude(  "whatever" );

But I'm hoping I can get specific values that are calculated in the handler. It looks like I can do this if I stored them in the PRC scope with

event.getPrivateCollection()

but I don't typically use the PRC scope. I pass the values to the results view via the

args

param in the

event.setView()

call in the handler. Is there a way to get the values in the args param in the test? Or do I need to start using the PRC scope if I want to do this? I checked the docs but, if there is anything, I didn't see it. Appreciate any insights on this. Thanks.

Happy Friday all. How does everyone handle their tests as far as version control is concerned? Same repo as the code or a separate repo? If the same, what's the best way to keep the tests out of production? The best I've come up with so far is adding something to our deployment script to just remove the tests directory in prod.

Is there a testbox header that is the failures + errors count? Using AWS CodeBuild (which does a curl to the CFM runner) and have to check both currently. Or should I be looking at something else instead of the

x-testbox-totalError

and

x-testbox-totalFail

?

Just in case yer interested in testing outside of CFML: https://blog.adamcameron.me/2022/10/data-driven-tests-in-junit-and-kotest.html

It's Kotlin rather than CFML, but the concepts and the techniques are the same, so I figured it's worth sharing: https://blog.adamcameron.me/2022/10/kotlin-tdd-writing-tests-for-small-web.html

Is there a way to override the mockbox stub directory universally, via Application.cfc or something? I see that I can do it within an individual test via

getMockBox()

. But I don't see a single location that would affect all of my tests.

hmm, good point

I don’t think there is

but we can do a setting somewhere

good point

maybe when isntantiating testbox

Using Testbox is ok to use

expect()

without using

describe

and

it

? meaning just like an `assert()`:

function testExpectationExamples () {
		var message = "A custom message can be prepended to the error message!";

		// toBe( expected, [message] ) : Assert something is equal to each other, no case is required
		var value = "myValue";
		expect(value).toBe("myvalue", message);
		expect(value).toBe("myValue", message);
		expect(value).toBe("MYVALUE", message);
}

Anyone using testbox, you might like something I've been working on, a shareable bunch of tools, allows me to do things like this. New matchers, new helpers, easy to add into testbox, but saves lots of code, lots more readable, way more readable messages when tests do not pass. @Adam Cameron might have given me an idea or two 😉

var oResponse = deserializeJSON( event.getRenderedContent() );
expect( oResponse ).toBeStruct()
    .toHaveKeyWithCase( "error" )
    .toHaveKeyWithCase( "messages" );
expect( oResponse.error ).toBeFalse();                        
expect( oResponse.messages )
    .toBeArray()
    .toHaveLengthGTE( 1 );
    expect( oResponse.messages[1] ).toBe( "postPandaDocWebhookRecipientCompletedADocument Interception point announced" );

to

expect( event ).toHaveAPISuccessWithMessage(    "postPandaDocWebhookRecipientCompletedADocument Interception point announced" );

Yeah implementing one's own bespoke matchers is something ppl should always consider if it can give a name to a commonly-used expectation. The second version there is much clearer as to what the expectation is.

(https://testbox.ortusbooks.com/in-depth/expectations/custom-matchers)

I have to admit it's not something I have done. And I ought to have.

guess it's slightly OT, but which testing framework do people use for their javascript side of their cfml apps?

I believe I found a bug in cbplaywright v1.0.2 for ACF (I'm using ACF 2021.0.05). When I use the

route

helper function, I get an error message that states

Variable STRUCTVALUEARRAY is undefined

in

cbPlaywright\models\PlaywrightMixins.cfm

on line 73. That line is

var pathArray = structValueArray( arguments );

—

structValueArray

is not a function in ACF.

This looks like an interesting testing ide tool from JetBrains - seems like it's combining the browser, db and code ide all in one... https://www.jetbrains.com/aqua/

anyone know of a tool to compare junit test results, lucee's build outputs them, I'd like to be able to compare runs

qq for those of you using testbox in your ci/cd deployments and also using docker: do you use a non-finalized commandbox image so you can run commandbox with

testbox run

or do you launch your tests some other way?

I've been able to successfully write some integration and unit tests for an app that doesn't require authentication (hooray!). However, now I'm starting some tests on an app that does require authentication. Unit tests are still no problem, however, I'm struggling to figure out how to write an integration test when I need

cgi.remote_user

populated in

main.index

as I'm doing something like

userId = cgi.remote_user

and then passing

userID

into a function to retrieve data and then display it on the screen based on who is logged in. What is the best approach here?

@johnbarrett has left the channel

In Testbox, for integration testing, is it possible to mock objects that are called from a handler prior to the

execute

call of the handler?