Repository navigation
test: do not depend on the api name prefix of error messages, part 2 - #44208
Conversation
Test results for "MCP"3 failed 5237 passed, 633 skipped Merge workflow run. |
🟢 All three MCP failures predate this PRHi, I'm the Playwright bot. I checked the failures in MCP run 44332477796 at Details and evidenceThe PR only relaxes error-message assertions in five library/page test files. It changes no MCP code, crash tests, or shared helpers. Pre-existing flake / infra All three tests also failed in the earlier Ubuntu Chrome job on main, at
These failures are confirmed to exist without this PR; this comparison does not establish whether they are intermittent. No change to this PR is indicated by the MCP failures. Checked the failures on Linux by the Playwright bot - agent run |
Test results for "tests 1"2 failed 1 flaky23641 passed, 716 skipped Merge workflow run. |
🟢 Both JUnit failures predate this PRHi, I'm the Playwright bot. I checked tests 1 run 44332477780 at Details and evidenceThe PR only relaxes assertions in five library/page test files. It changes no JUnit reporter code, reporter tests, or shared helpers. Pre-existing flake / infra Both All four failures are at line 374: expected The small correction is to expect The report's Selenium test passed on retry, so it is not counted as a failure. The MCP failures already have a separate verdict above. Checked the failures on Linux by the Playwright bot - agent run |
67909fe
into
microsoft:main
Summary
tests/pageandtests/librarychecked the api name prefix (elementHandle.$eval:,selectors.register:,jsHandle.getProperty), which the language ports print differently or not at all. They now check the text after the prefix.