New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
How will parallel option work with event protocol? #1038
Comments
I think each event should contain enough information to be able to link it to previous events rather than buffering. |
Not sure how that would work if only 'end' events are hooked as it would miss beginning of test's output. To use pretty formatter and IntelliJ-based editors' test runners as examples, they work as follows:
The order is important as to have opening statement before anything that the test logs and closing statement afterwards and no other test should log anything in parallel to that or they'll mix up which output belongs to which test. Buffering would seem to take care of that as each output will be isolated and logged on master in atomic operation. |
We can pass a |
We need more discussion on this issue. Currently, using any formatter at all has a risk of crashing the cucumber runner due to I/O conflicts while trying to write to a terminal. The progress bar formatter would be ideal for the parallel runner, if it only moved the bar along and then reported a summary at the end. |
We need this feature very much!!! |
Hello! Is there any chance to run cucumber js in parallel with allure reporter? When I try to run with cucumber-js --parallel 2 -t @debug --format reporter.js:./dummy.txt I just got TypeError: Cannot read property 'sourceLocation' of undefined |
Same issue as @yevgen-getalo here. |
Closing as v7.0.0 uses the new messages protocol mentioned above, where |
Custom formatters that only hook on
test-run-finished
like JSON formatter work as intended when parallelised but formatters that use earlier events liketest-(step|case)-started
and are ordered have their output logged out of sequence invalidating the output document structure. Is there anything planned to address that? May be buffer the output on per worker bases and log it all at once on master once completed?The text was updated successfully, but these errors were encountered: