Updates to storage ui test suite #1433
Conversation
| authenticationPage.showMoreButton().click() | ||
| authenticationPage.detectedWallet().click() | ||
| authenticationPage.web3SignInButton().safeClick() | ||
| bucketsPage.bucketsHeaderLabel().should("be.visible") |
There was a problem hiding this comment.
Just an FYI - The last line here is required so that we don't exit the session block early before we have fully logged in, prevents us exiting before the session data is created and saved.
We haven't been waiting for the header label on Files in the session block because we also have to enter a password and not save the browser so I think that gives us a little bit of extra time to save the token. To avoid any potential timing issues though I’ve added a similar check for Files too
There was a problem hiding this comment.
Thanks, the save browser modal in Files is certainly the one that helped (just a FYI because before the password is entered we haven't reconstructed the key, and the there's prob. not much of interest in the local and session storages.)
There was a problem hiding this comment.
@Tbaut Ah, good to know the detail. Thanks 👍
|
Your Render PR Server URL is https://storage-ui-stage-pr-1433.onrender.com. Follow its progress at https://dashboard.render.com/static/srv-c49ftu1g7hpcpoptjb4g. |
|
Your Render PR Server URL is https://gaming-ui-stage-pr-1433.onrender.com. Follow its progress at https://dashboard.render.com/static/srv-c49ftuhg7hpcpoptjbh0. |
|
Your Render PR Server URL is https://files-ui-stage-pr-1433.onrender.com. Follow its progress at https://dashboard.render.com/static/srv-c49ftv1g7hpcpoptjbqg. |
|
Just a note, the 1st test is always failing
|
Oops, that was my mistake. I didn't re-add the |
…files-ui into mnt/update-storage-tests-1283
asnaith
added
the
Status: Review Needed 👀
Work towards #1283 (still a wip but submitting these changes now to keep the PRs small).
Notes:
Still not ready to enable these tests because of issues with the apiTestHelper (see #1434). It is either not clearing pins (or it is but they aren't removed straight away because of the queuing system, I'm really not sure what's happening with it at the moment). This going to make it hard to build reliable ui tests for CID Pinning at least in the current structure. For example at the moment
I can't assert a CID was successfully pinned because there might already be one pinned at the beginning of the test so counting on the rows could be unreliable
I can't assert that the CID was deleted with the current test because the pin deletion is usually queued for a indeterminate period of time and not instantly removed from the UI (although sometimes it is so we can't just assert on "queued" status either!).