Skip Menu |

This queue is for tickets about the DateManip CPAN distribution.

Report information
The Basics
Id: 3896
Status: resolved
Priority: 0/
Queue: DateManip

People
Owner: Nobody in particular
Requestors: mschwendt [...] users.sf.net
Cc:
AdminCc:

Bug Information
Severity: Unimportant
Broken in: 5.42
Fixed in: (no value)



Subject: t/events test fails randomly
This is on "Fedora Core release 0.94 (Severn)", which is the second test release of what will become the successor of Red Hat Linux 9. Summary: In "make test", t/events fails randomly because &Events_Calc returns an unsorted list of events and the test expects constant order. Perl's "keys" function returns keys in random order. Details and example error output at: https://bugzilla.fedora.us/show_bug.cgi?id=714#c7 Proposed fix (see patch) would be to sort the returned list of events in &Events_Calc, so it would be safe to expect a constant order in t/events. Alternatively, I could think of doing the sorting in the test.pl code.
diff -Nur DateManip-5.42a-orig/Manip.pm DateManip-5.42a/Manip.pm --- DateManip-5.42a-orig/Manip.pm 2003-07-02 18:42:47.000000000 +0200 +++ DateManip-5.42a/Manip.pm 2003-09-28 13:11:46.000000000 +0200 @@ -5117,7 +5117,7 @@ $tmp =~ s/^!//; delete $tmp{$tmp}; } - push(@ret,$date,[ keys %tmp ]); + push(@ret,$date,[ reverse sort keys %tmp ]); } return \@ret;