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
Currently, to specify an expected nonzero exit status for a test command, you have to configure it manually: either in turnt.toml or in an embedded marker in the test. This works fine when, for example, handcrafting test programs with expected failures.
But for differential testing (a newish use case for Turnt that is still not completely seamless), this is annoying. We instead want to save the expected exit status from one command run, in a file, and then compare it against the exit status from another command—just like we do with textual output streams.
Maybe we can hack this in with another special output value in the config, just like - and 2 indicate the stdout and stderr streams. Like this:
output.status = '?'
That would tell Turnt to save the exit status in a file called <testname>.status.
The text was updated successfully, but these errors were encountered:
Currently, to specify an expected nonzero exit status for a test command, you have to configure it manually: either in
turnt.toml
or in an embedded marker in the test. This works fine when, for example, handcrafting test programs with expected failures.But for differential testing (a newish use case for Turnt that is still not completely seamless), this is annoying. We instead want to save the expected exit status from one command run, in a file, and then compare it against the exit status from another command—just like we do with textual output streams.
Maybe we can hack this in with another special
output
value in the config, just like-
and2
indicate the stdout and stderr streams. Like this:That would tell Turnt to save the exit status in a file called
<testname>.status
.The text was updated successfully, but these errors were encountered: