[5.0] store the source of the browser #705
Conversation
when the test completes it will store the final source code in a directory, same as it does for the console log output. - add a new static property to store the path for where the files should be stored - new method that gets the source from the driver and saves it to the filesystem - store the source after a test has completed. there could be some discussion here on whether we do it only for failures or all tests.
|
Does this also work for SPAs? Or does this only work for "normal" blade/html pages? |
|
I don't think we would want to the source code every test? Seems like a lot of wasted space? |
|
@SjorsO I don't see why it wouldn't. I guess I'm not positive how the driver gets the source code, whether it's the source at the load of the page or if it takes into account all the changes made during the execution of the page. @taylorotwell I can move this into the |
|
do you guys handle new merge conflicts that pop up, or should I go in and do that? |
|
@browner12 best if you rebase yourself |
|
Personally I think I would make this a method on the browser that immediately takes a snapshot of the source and stores it. Then, if you have a test that fails you can just pop this method onto the browser somewhere and grab the source. Then we don't have to worry about storing the source on every test or every failure, etc. You would just end up using when you need to similar to "dd" etc. |
I can't tell you how many times I've clicked on the screenshot of a failed test trying to interact with it...
This will allow us to dig through the full source code of the browser at the point in time of the failures, which can help with debugging.
2 questions to ask:
.txt, but I'm considering.html.My grand master plan is to maybe proxy this code somehow so people can pull up the failure in their browser and actually interact with it.