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
such that stack build (after stack build) could appear be an 'no op'. Such a user could command stack build --run-tests, if they wished for test suites of project packages to be run.
However, consider a simple two-project package (packageA, packageB) each with two test suites (1 and 2) and two benchmarks (1 and 2). The second stack build yields:
Completed 4 action(s).
The four actions in question are the running of test suites and benchmarks for each of two project packages; these actions do nothing because --no-run-tests and --no-run-benchmarks.
With notification restored, the output is:
packageA> Test running disabled by --no-run-tests flag.
packageB> Test running disabled by --no-run-tests flag.
packageA> Benchmark running disabled by --no-run-benchmarks flag.
packageB> Benchmark running disabled by --no-run-benchmarks flag.
Completed 4 action(s).
I propose that --no-run-tests and --no-run-benchmarks should be respected when the list of actions is created, as opposed to when the list of actions that has been created is subsequently executed.
After the change, with notification restored, the output is (but in colour):
All test running disabled by --no-run-tests flag. To mute this message in future, set
notify-if-no-run-tests: false in Stack's configuration.
All benchmark running disabled by --no-run-benchmarks flag. To mute this message in future, set
notify-if-no-run-benchmarks: false in Stack's configuration.
After the change, with notification disabled, the second stack build is the desired 'no op'.
The text was updated successfully, but these errors were encountered:
The motivation of:
--run-tests
and--run-benchmarks
#6637notify-if-no-run-tests
andnotify-if-no-run-benchmarks
#6639taken together, was to allow a user to specify configuration:
such that
stack build
(afterstack build
) could appear be an 'no op'. Such a user could commandstack build --run-tests
, if they wished for test suites of project packages to be run.However, consider a simple two-project package (
packageA
,packageB
) each with two test suites (1 and 2) and two benchmarks (1 and 2). The secondstack build
yields:The four actions in question are the running of test suites and benchmarks for each of two project packages; these actions do nothing because
--no-run-tests
and--no-run-benchmarks
.With notification restored, the output is:
I propose that
--no-run-tests
and--no-run-benchmarks
should be respected when the list of actions is created, as opposed to when the list of actions that has been created is subsequently executed.After the change, with notification restored, the output is (but in colour):
After the change, with notification disabled, the second
stack build
is the desired 'no op'.The text was updated successfully, but these errors were encountered: