使用 selenium webdriverjs 运行 mocha 测试时调用堆栈不是很有帮助

Question

When running a test in mocha using the selenium webdriver for JavaScript the call-stack it not really helpful.

As a simple example I've used the following simple test script test.js :

var webdriver = require('selenium-webdriver'),
    test = require('selenium-webdriver/testing');

const TIMEOUT = 30000;

test.describe('selenium webdriverjs', function () {
    var driver;

    this.timeout(TIMEOUT);

    test.before(function () {
        driver = new webdriver.Builder().withCapabilities(webdriver.Capabilities.firefox()).build();
        driver.manage().timeouts().pageLoadTimeout(TIMEOUT);
    });

    test.after(function () {
        driver.quit();
    });

    test.it('error stack', function () {
        driver.get('http://www.google.com');
        driver.findElement(webdriver.By.id('doesnotexit'));
    });
});

that generates the following error stack when (for example) run from Gulp:

  selenium webdriverjs
    1) error stack


  0 passing (4s)
  1 failing

  1) selenium webdriverjs error stack:
     NoSuchElementError: Unable to locate element: {"method":"id","selector":"doesnotexit"}
      at new bot.Error (...\node_modules\selenium-webdriver\lib\atoms\error.js:108:18)
      at Object.bot.response.checkResponse (...\node_modules\selenium-webdriver\lib\atoms\response.js:109:9)
      at ...\node_modules\selenium-webdriver\lib\webdriver\webdriver.js:379:20
      at promise.Promise.goog.defineClass.invokeCallback_ (.../node_modules/selenium-webdriver/lib/goog/../webdriver/promise.
js:1337:14)
      at promise.ControlFlow.goog.defineClass.goog.defineClass.abort_.error.executeNext_.execute_ (.../node_modules/selenium-
webdriver/lib/goog/../webdriver/promise.js:2776:14)
      at promise.ControlFlow.goog.defineClass.goog.defineClass.abort_.error.executeNext_ (.../node_modules/selenium-webdriver
/lib/goog/../webdriver/promise.js:2758:21)
      at goog.async.run.processWorkQueue (...\node_modules\selenium-webdriver\lib\goog\async\run.js:124:15)
  From: Task: selenium webdriverjs error stack
      at Context.ret (...\node_modules\selenium-webdriver\testing\index.js:128:10)
      at callFnAsync (...\node_modules\mocha\lib\runnable.js:306:8)
      at Test.Runnable.run (...\node_modules\mocha\lib\runnable.js:261:7)
      at Runner.runTest (...\node_modules\mocha\lib\runner.js:421:10)
      at ...\node_modules\mocha\lib\runner.js:528:12
      at next (...\node_modules\mocha\lib\runner.js:341:14)
      at ...\node_modules\mocha\lib\runner.js:351:7
      at next (...\node_modules\mocha\lib\runner.js:283:14)
      at Immediate._onImmediate (...\node_modules\mocha\lib\runner.js:319:5)

The stack consists of a series of deeply nested function in the mocha and selenium libraries of the local node_modules directory and actual test script test.js is not even part of the stack.

I'm wondering, if this is really the best information I can get from this setup if I just did something wrong?

Answer 1

Yes, this is really too much unnecessary and not really helpful information with all of the meaningless async traceback cruft, but the most important part of it, the webdriver error itself, is on top and is pretty self-descriptive :

NoSuchElementError: Unable to locate element: {"method":"id","selector":"doesnotexit"}

You may try to clean up the output with the help of mocha-clean package. When applied, it should really leave the error message itself and only relevant parts of the stack trace eliminating everything mocha - and selenium-webdriver -specific:

Mocha stack traces are riddled with frames that you don't want to see, like code from modules and Mocha internals. It strips away mocha internals, node_modules, absolute paths (based on cwd), and other unneccessary cruft.