I don't have time to track this down, but I've noticed the following when I run
my Test::Class tests through prove:
shared_cp $ prove prove -l t/test_class_tests.t
prove.................prove does not exist
t/test_class_tests....Tests::****::ControlPanel::Remote
Tests::****::DB::Version
Tests::****::RPC::Action
Tests::****::RPC::Action::db_version
Tests::****::Net::Interfaces
t/test_class_tests....ok 27/0
# Failed (TODO) test 'Trying to finish an upgrade with bad SQL should
fail'
# in t/tests//Tests/****/DB/Version.pm at line 295.
# undef
# doesn't match '(?-xism:You have an error in your SQL syntax)'
# Failed (TODO) test '... even if it has multiple statements'
# in t/tests//Tests/****/DB/Version.pm at line 308.
# didn't found a warning
# expected to find warning: (?-xism:You have an error in your SQL syntax)
# Failed (TODO) test '... or a quoted statement separator'
# in t/tests//Tests/****/DB/Version.pm at line 318.
# didn't found a warning
# expected to find warning: (?-xism:Table '[^.]*\.foo' doesn't exist)
t/test_class_tests....ok
All tests successful.
Files=2, Tests=123, 2 wallclock secs ( 1.00 cusr + 0.70 csys = 1.70 CPU)
As you may notice, you see the diagnostics from the TODO failures. As Schwern
has noted repeatedly:
The user should not see the diagnostics from that TODO failure.
Oddly enough, while I find the output annoying, my tests aren't broken. Fancy
that.
In any event, when I have more time, I'll try and find out what's broken and
file a bug report. However, I'm off to Scotland this weekend and won't be back
for a few days (yeah, you'll get some peace and quiet!) I'd dig into this
further if it was really causing problems, but for now, I'll ignore it other
than to ask, has anyone else seen this behavior? And is there any way that
displaying the diagnostics from TODO failures can break things (such as making
it difficult to install modules)?
Cheers,
Ovid
--
Buy the book -- http://www.oreilly.com/catalog/perlhks/
Perl and CGI -- http://users.easystreet.com/ovid/cgi_course/
