I'm using qunit 2.3.3 with qunit-parameterize. I'm interested in preparing a modal dialog html at start of the suite/runner.html which is used to select the test cases for execution. Is there a way to do this or any thoughts or leads around this please
Note: I found qunit.config.autostart(false) & start()/stop() for execution. But unable to call start() & stop() as they are not recognized as qunit functions. How to solve this ?
Related
I am trying to debug in nightwatch.
When I put in a console.log it prints before the test runs - while its doing some sort of construction/compiling of the test.
I tried also visual studio code debugger and same thing - the breakpoint hits before the tests are actually run.
thanks for help - mark
"nightwatch" is built on node.js. Node.js executes statements asynchronously. If you write console.log();, it will be executed asynchronously.
The statements you write using browser (or client) object (e.g. browser.url(); , browser.click(); etc.) will be queued up at Selenium Server. They are also executed asynchronously by node.js but are queued up at selenium server.
To perform console.log() in sync with other statements in nightwatch, use .perform(). You will get output in sync with other statements.
Example
var elementValue;
browser
.getValue('.some-element', function(result) {
elementValue = result.value;
})
// other stuff going on ...
//
// self-completing callback
.perform(function() {
console.log('elementValue', elementValue);
// without any defined parameters, perform
// completes immediately (synchronously)
})
.end();
For debugging purpose, you can stop the execution to find out element values or check the browser related data. Use this command: browser.pause(); (or client.pause();). Do not pass any timer here. It will stop the execution.
As you don't really have an example, or some code that you run, i can only give you the general direction you can try in order to fix the issue.
This happens because your code is ran asynchronously (cause that's how javascript works) and everything that is not browser/html related will run before your browser starts your actual test. I suggest you should try using callbacks in order avoid printing your check msgs first (this way you link the console.log to your code that depends on the browser and it will run only after this finishes. If you do that, the debugging with console log will actually work.
Maybe you can learn here how to do that, as an example.
I'm familiar with python unittest tests where if an assertion fails, that test is marked as "failed" and it moves on to other tests. Jasmine on the other hand will continue through all expects even if the one of them fails. How can I make Jasmine stop processing a test after the first expectation fails?
it ("shouldn't need to test other expects if the first fails", function() {
expect(array.length).toBe(1);
// don't need to check this if the first failed.
expect(array[0]).toBe("foo");
});
Am I thinking about it wrong? I have some tests with lots of expect's and it seems like a waste to show all the stack traces when only the first is wrong really.
#Gregg's answer was correct for the latest version of Jasmine at that time (v2.0.0).
However, since then, this new feature was added in v2.3.0:
Allow user to stop a specs execution when an expectation fails (Fixes #577)
It's activated by adding throwFailures=true to the query string of the runner page, eg:
http://localhost:8000/?throwFailures=true
Jasmine doesn't support failing early, in a single spec. The idea is to give you all of the failures in case that helps figure out what is really wrong in your spec.
Jasmine has stop on failure feature and you can check it here:
https://plnkr.co/plunk/Ko5m6MQz9VUPMMrC
This starts jasmine with oneFailurePerSpec property.
According to the comments of https://github.com/jasmine/jasmine/issues/414 I figured out that 2 solutions exists for this:
https://github.com/radialanalytics/protractor-jasmine2-fail-whale
https://github.com/Updater/jasmine-fail-fast
I just started to use the protractor-jasmine2-fail-whale because it seems to have more features. Although to take screenshots in case of test failures I currently use protractor-jasmine2-html-reporter.
I'm using Jasmine in Appium (a tool to test React Native apps).
I fixed the issue by adding stopSpecOnExpectationFailure=true to Jasmine configs
jasmine v3.8.0 & jasmine-core v3.8.0
I recall having seen at some point screen shots of a select list of QUnit test modules in the test runner toolbar of QUnit. My impression was that selecting one of the modules in the select list would cause that module's tests to be run.
Question: Does such a feature actually exist OOB for QUnit? I know one can set filter via the URL but I would like a more "discoverable" option.
Thanks!
The select list only shows itself if you have defined more than one module in your test suite.
Also, make sure that your test suite is ready before QUnit initializes itself. i.e. QUnit initializes itself when the page finishes loading (the onload event). If you happen to define your test suite after this, then you have to call the (undocumented) QUnit.load() method to notify QUnit that your test suite has been defined.
Demo: http://jsfiddle.net/brianpeiris/98fc8/show/
So lets say I am doing a Test Complete keyword test. If something fails in it the text stops . Actually what I have founded out is that if i have 8 checkpoints if the 4th one fails the rest will always fail after it. So i get a "test execution was interrupted" error. Thats fine but it doesnt finish out the test and close the application. The reason this is an issue is because any tests after it will fail because the application is still left open. I could rewrite these tests so that the application is open when they start but is there a way to kill and application after your tests fail? If the tests pass the application is closed.
You need to organize your tests with test items. In this case, you create at least 3 test items: the first one starts the application, the second performs the test and the third closes the application. If an error occurs during execution of the second test, this second test execution is ended and TestComplete runs the third finalization test item.
Information on test items can be found in the Tests and Test Items help topic. Please note that you need to specify the Test Item value in the Stop on error column for the needed test item (the second one in the above example). Information on this and other columns can be found here. The column is hidden by default and you need to add it: right-click the header of the test items list and select Field Chooser. After this, drag the needed column to the header from the Field Chooser dialog.
Find more information on this solution in Stopping Tests on Errors and Exceptions.
Alternative solution is using the OnLogError or OnStopTest event handlers. Find description of how to handle standard TestComplete events in the Creating Event Handlers for TestComplete Events help topic.
Perhaps I'm oversimplifying, but could it be the setting for the test playback? Pls check the following page and let me know if it helps: http://support.smartbear.com/viewarticle/28751/.
If that doesn't work feel free to repost in the SmartBear Forum: http://community.smartbear.com/
The support team is monitoring the forum and I'm sure they'll be happy to help.
It appears that Selenium has a feature called JavascriptExecutor which makes it possible to execute JavaScript directly on the page. However it appears that there is no such thing available to my Cucumber/Capybara tests. How can I execute arbitrary JavaScript from my Cucumber tests?
Capybara has two methods to execute javascript #execute_script and #evaluate_script.
Both can be found at: http://rubydoc.info/github/jnicklas/capybara/master/Capybara/Selenium/Driver
However, do note that the readme has this statement about the above methods:
For simple expressions, you can return the result of the script. Note
that this may break with more complicated expressions:
See http://rubydoc.info/github/jnicklas/capybara/master/file/README.md.