Software Quality Assurance & Testing Stack Exchange is a question and answer site for software quality control experts, automation engineers, and software testers. It only takes a minute to sign up.
Sign up to join this community
Navigate to the custom elements example, open a console and type in document.body.getElementsByTagName("*").length - you should get 6 returned as the result.
Using webdriver, try (def d (execute-script "var n = document.body.getElementsByTagName(\"*\").length; return n;")) and a count on that returns 5.
Why is webdriver auto-filtering my results? The custom element hangout-module is not returned.
As per the rules of W3C custom elements are registered by script using document.registerElement(). Now, it may be possible that they are declared or created before their definition is registered by the browser. For example, you can declare on the page but end up invoking document.registerElement('hangout-module') much later.
This although this is an HTML element that has a valid custom element name but haven't been registered. Due to this, the element remains an unresolved element until it is upgraded to its definition.
And it may be possible that Webdriver is not yet designed to read unresolved elements. So you might want to look in to it, whether your custom element is registered before it is declared.
Can you try and check whether WebdriverJS reads the element at all or not. Would you try isDisplayed(), or try to find the element by xpath or id or CSS and assert whether it is present(read by Webdriver) or not?
getElementsByTagNameNS is an alternative means to accomplish what you are trying to do. I know that getElementsByTagName changes all values to lowercase to process. Could be its either not counting a capital letter.
Example from the API
elements = document.getElementsByTagNameNS(namespace, name)
I had similar problem when our designer used custom elements (with custom tag) in Angular/material. Webdriver was not able to interact with such elements (IAmMilinPatel could be right), but I did not spent much time on researching real underlying reason for this - changing the approach was much quicker :-)
I ended up tweaking design a bit, generating span elements with custom names. Webdriver was able to find and interact with elements generated such way.
Webdriver can find all elements with a given name, returns you a list, you can loop through that list and interrogate items to find which you really want by any custom rules (instead of relying on WebDriver to find you needed elements). Or as I once needed, match items from two such lists to pairs based on autogenerated IDs. It is a cool trick to have in your toolbox.
(def d (execute-script "var n = document.body.getElementsByTagName(\"*\").length; return n;"))getElementsByTagNamebefore hangout-module exists. Adding a sleep beforehand might help determine if that's the problem.