You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
1.70.0: ui test failures due to "the full type name has been written to" #112209
= note: the full type name has been written to '$TEST_BUILD_DIR/methods/method-not-found-generic-arg-elision/method-not-found-generic-arg-elision.long-type-9165007450593903927.txt'
Should we suppress the "full type name output" within the test suite, or am I doing something wrong?
i wonder if this should be an explicit config instead of being auto detected based on a CI environment variable. then we could enable it by default for the dev channel and disable it everywhere else.
Having an explicit config for this sounds great. It would also mean that people with custom buildbots no longer need to spoof the CI detection logic (like e.g. Chromium does right now: chromium/chromium@9fc3ea5).
Activity
TimNN commentedon Jun 3, 2023
Some background:
I've attempted to fix this in #105924 by enabling path remapping in UI tests by default.
However, that caused some usability regressions for
rustcdevelopers (#109725), so now this logic only triggers on CI: #110115jyn514 commentedon Jun 3, 2023
i wonder if this should be an explicit config instead of being auto detected based on a CI environment variable. then we could enable it by default for the dev channel and disable it everywhere else.
TimNN commentedon Jun 3, 2023
Having an explicit config for this sounds great. It would also mean that people with custom buildbots no longer need to spoof the CI detection logic (like e.g. Chromium does right now: chromium/chromium@9fc3ea5).
apiraino commentedon Jun 5, 2023
WG-prioritization assigning priority (Zulip discussion).
@rustbot label -I-prioritize +P-low
chenyukang commentedon Aug 23, 2023
The path remapping is removed in #114090, I verified the test issue in the stable channel is fixed, closing this issue now.