Closed aquanight closed 1 year ago
Can you show a before-and-after of a failed command, since there is no test coverage of this area to show what the changes would look like?
As far as I can tell:
before:
Can't exec "lss": No such file or directory at /Users/rjbs/.plenv/versions/5.37.2/lib/perl5/site_perl/5.37.2/Dist/Zilla/Dist/Builder.pm line 854.
[DZ] error while running: lss at /Users/rjbs/.plenv/versions/5.37.2/lib/perl5/si
te_perl/5.37.2/Dist/Zilla/Dist/Builder.pm line 854.
[DZ] left failed dist in place at .build/a9IrSHIhsi
after:
[DZ] "lss" failed to start: "No such file or directory" at /Users/rjbs/.plenv/ve
rsions/5.37.2/lib/perl5/site_perl/5.37.2/Dist/Zilla/Dist/Builder.pm line 856.
[DZ] left failed dist in place at .build/tgqXvZr2Xb
Is this better? I don't know, I'm not really seeing the specific way it's better. With no reply to the question already asked, I'm closing this unless there's more to it.
This improves the error message provided by 'dzil run' to indicate the resulting exit status of a failed command. An example of where this might help is during exploratory testing using 'dzil run perl'.
In particular I ran into this problem when a module I was doing such testing with crashed with SIGSEGV: with no output from perl itself and only a very basic message from dzil, I had to guess (or discover via strace) what had actually happened.