I wanted to start a discussion around the idea of abstracting web driver implementations into "adapters" which have a specified interface.
A little background: We have scripts already written utilizing selenium-webdriver however we do not currently have the ability to run against a grid of browsers. We examined the source code of this library and loved the implementation. The only issue is that it is dependnent on utilizing wd rather than selenium-webdriver.
By agreeing on a common interface, the webdriver implementations could be abstracted and specified as a configuration option. The library would default to using wd.
At a high level if you are ok with this approach, I can dive deeper and propose what the interface will look like. Upon approval of the interface, I will make the necessary code changes and submit a PR.
If you are not longer maintaining this library, or have little interest in the proposed changes, we will move ahead by forking and simply swapping wd out.
I am hugely in favor of this approach. It had been discussed in the past but was considered too difficult to pursue - but if you're willing to do it, go for it, I'd happily bring it to master.
I wanted to start a discussion around the idea of abstracting web driver implementations into "adapters" which have a specified interface.
A little background: We have scripts already written utilizing
selenium-webdriver
however we do not currently have the ability to run against a grid of browsers. We examined the source code of this library and loved the implementation. The only issue is that it is dependnent on utilizingwd
rather thanselenium-webdriver
.By agreeing on a common interface, the webdriver implementations could be abstracted and specified as a configuration option. The library would default to using
wd
.At a high level if you are ok with this approach, I can dive deeper and propose what the interface will look like. Upon approval of the interface, I will make the necessary code changes and submit a PR.
If you are not longer maintaining this library, or have little interest in the proposed changes, we will move ahead by forking and simply swapping wd out.
Thanks!