We should not stall selector actions because of dialogs
and properly timeout instead. For this, we should not await
the handle.dispose() call because it will never happen
while dialog is shown.
Also, log information about dialogs to make it easier to debug.
- Gave all possible dom errors distinct names, and throw them on the node side.
- Separated errors into FatalDOMError and RetargetableDOMError.
Fatal errors are unrecoverable. Retargetable errors
could be resolved by requerying the selector.
- This exposed a number of unhandled 'notconnected' cases.
- Added helper functions to handle errors and ensure TypeScript catches
unhandled ones.
WebKit and Firefox are only able to continue redirects.
Firefox is faking it on the backend, so you can't even stall it.
Instead, we just do not fire routes for redirects on all browsers,
to avoid surprises.
This is a speculative fix to the following issue from the bots:
NON-TEST ERROR #0: UNHANDLED ERROR
TypeError: Cannot read property 'url' of undefined
at WKPage._onTargetCreated (/Users/runner/runners/2.169.1/work/playwright/playwright/src/webkit/wkPage.ts:274:12)
at process._tickCallback (internal/process/next_tick.js:68:7)
I assume that _initializeSession did swallow an error, so we erroneously
consider Page to be fully initialized (and having main frame).
Since we are recommending Promise.all pattern anyway, this special
logic just adds to the possibility of timeout if something goes wrong.
For example, Firefox sometimes send Page.willOpenNewWindowAsynchronously
later than the new target arrives and input action just hangs.
Currently, whenever the page crashes, it emits an `'error'` event.
Error event is a special type of event in node.js; if unhandled,
it crashes the process.
Instead of emitting `'error'` event, this patch switches to emitting
`'crash'` event. Playwright users are free to handle the event
however they like, or just to ignore it.
This encapsulates selectors logic in one place, in a preparation for more complex scenarios like main-world selectors or piercing frames.
Note: we had `Page.fill should wait for visible visibilty` test, but we do not actually wait for visible in page.fill(). It happened to pass due to lucky evaluation order.
References #1316.
