This is a report on differences between the output of
'make test' and 'prove' which are, well, different
from those reported by Stevan Little in a thread
beginning on Sept 4.
While teaching myself how to use 'prove' tonight, I
decided to try it out on a Perl core module which I
have also been using as a test case for my
understanding of Devel::Cover. The core module in
question is Time::Local, version 1.10. When I called
'make test' on this module, I got the following
results:
[Time-Local-1.10 562]$ make test
PERL_DL_NONLAZY=1 /usr/local/bin/perl
"-MExtUtils::Command::MM" "-e" "test_harness(0,
'blib/lib', 'blib/arch')" t/*.t
t/Local....ok
All tests successful.
Running this with TEST_VERBOSE=1 indicated that the
number of tests run was 102.
However, when I called 'prove t', I got different
results:
[Time-Local-1.10 563]$ prove t
t/Local....Cannot handle date (7, 14, 3, 19, 0,
2038) at t/Local.t line 105
t/Local....dubious
Test returned status 255 (wstat 65280,
0xff00)
DIED. FAILED tests 101-102
Failed 2/102 tests, 98.04% okay
Failed Test Stat Wstat Total Fail Failed List of
Failed
-------------------------------------------------------------------------------
t/Local.t 255 65280 102 4 3.92% 101-102
Failed 1/1 test scripts, 0.00% okay. 2/102 subtests
failed, 98.04% okay.
As best as I can tell, these are the tests in
t/Local.t which failed:
ok(sprintf('%x', timegm(gmtime(0x7fffffff))),
sprintf('%x', 0x7fffffff), # line 105
'0x7fffffff round trip through gmtime then
timegm');
ok(sprintf('%x', timelocal(localtime(0x7fffffff))),
sprintf('%x', 0x7fffffff),
'0x7fffffff round trip through localtime then
timelocal');
Any ideas?
Jim Keenan
=====
Affiliations: Perl Seminar NY / New York Perlmongers / Toronto Perlmongers
__________________________________
Do you Yahoo!?
Check out the new Yahoo! Front Page.
www.yahoo.com
