Repository navigation
Feature request: report failing tests at the bottom of the spec reporter #47110
Description
Activity
- changed the title
[-]report failing tests at the bottom of the spec reporter[/-][+]Feature request: report failing tests at the bottom of the spec reporter[/+]on Mar 15, 2023 Hi, can I take this?
Hi,
I am interested to work on the changes.
Based on my understanding:
- We need to output the failing tests report in root
Test(lib/internal/test_runner/test.js). - Since
#handleReportItem(lib/internal/test_runner/runner.js) immediately emits the test output to stream, we need to store the error output somewhere. - Eventually, all children
Tests need to propagate the error outputs to their parents. ArrayPrototypePushis concurrency safe.
Please correct me if I am wrong.
Draft commit hereI need suggestions about the output format.
I feel that if we output the whole diagnostics (including stack trace) again, it will make the output too cluttered since too many duplicates. Maybe we can omit the stack trace before the summary?
And I am worried about the memory usage, should we allow this feature to be disabled by choice (using flag)?Also may I know how to properly re-generate
test_runner_default_reporter.out?
Thanks!- We need to output the failing tests report in root
@HinataKah0 I would expect the change to only affect the spec reporter
whenever it receives atest:failevent, it can store the failed test, and when completed the reporting, it can print the failutes in the same format, at the end of the stream and with no indentationHi @MoLow ,
Thanks for the kind reply! Now the implementation is quite clear for me.
However, I noticed that we need to introduce a new event to mark the end of tests (e.g.test:eot).
Becausetest:diagnosticis sent a few times so it doesn't signify the ending. Actually I can maintain a counter but I don't think it's clean.To illustrate what I am talking about here
Does this match to what you have in mind?
I think adding file path will be very useful.It took me a while to understand how nodejs' build testing works (and to understand why I broke some tests). 🤣
Thanks!
Reacted by Moshe Atlow- added a commit that references this issue
on Mar 23, 2023 - added 2 commits that reference this issue
on Apr 5, 2023 - added a commit that references this issue
on Jul 6, 2023
What is the problem this feature will solve?
when test runs fail, it is currently needed to scroll up and manually search for the failing tests,
What is the feature you are proposing to solve the problem?
this behavior exists in some popular test runners (mocha, jest, playwright, etc),
re-printing the list of failing tests after the summary can be very useful
What alternatives have you considered?
scrolling to find the failing tests manually