Em 06-09-2016 00:53, Kent Fredric escreveu:
On 5 September 2016 at 04:10, Slaven Rezic <sla...@rezic.de> wrote:
Of course I can try to compile the interpreter without perlbrew, but that leads
to the questions: are we looking for to validate CPAN modules against the
interpreter shipped with OpenBSD or the standard perl?
There's no "or". Both types of reports are valuable.
+1
Because we should expect software to run on system perls wherever
system perls exist. Things that do not run are bugs, and must be fixed
by avoiding the problem case in the code, or where that is impossible,
resolving the breakage on the system perl.
I'll try to implement that with two different users on the same OpenBSD
VM: one with the "system wide" perl and the other with the same
(hopefully) version, but only for this particular user. perlbrew does
fail one test during setup, but that can be safely ignored (at least AFAIK).
