On Sun, 11 Feb 2024 19:38:28 GMT, Phil Race <p...@openjdk.org> wrote:
>> Hi Reviewers, >> I have updated the test with 'PassFailJFrame' with programmatically >> generating image and print , please review and let me know your suggestions. > > I had a slack chat with Jon Gibbons (maintainer of jtreg) and he thinks FAIL > would be better here. > I think we should be consistent and now that all tests that need a printer > specify @key printer - > should FAIL if there's no printer. Not saying we have to rush off and change > them all immediately, > but if we are updating it anyway for some reason then do it. @prrace You're right. I was under impression that `jtreg.SkippedException` gives the test a distinct status from either *Pass* or *Fail*. Eventually, it's not the case. In verbose mode, I can see the exception: runner starting test: java/awt/PrintJob/test/FailedTest.java runner finished test: java/awt/PrintJob/test/FailedTest.java Failed. Execution failed: `main' threw exception: java.lang.Error: Fail the test runner starting test: java/awt/PrintJob/test/PassedTest.java runner finished test: java/awt/PrintJob/test/PassedTest.java Passed. Execution successful runner starting test: java/awt/PrintJob/test/SkippedTest.java runner finished test: java/awt/PrintJob/test/SkippedTest.java Passed. Skipped: jtreg.SkippedException: Skip the test Test results: passed: 2; failed: 1 This is why I thought it was recorded as a different status. At the same time, the report looks like this: ============================== Test summary ============================== TEST TOTAL PASS FAIL ERROR >> jtreg:open/test/jdk/java/awt/PrintJob/test 3 2 1 0 << ============================== TEST FAILURE If `jtreg.SkippedException` resulted in a status distinct from *Pass* or *Fail*, it would be more useful. ------------- PR Comment: https://git.openjdk.org/jdk/pull/17790#issuecomment-1939412090
