Skip Menu |

Preferred bug tracker

Please visit the preferred bug tracker to report your issue.

This queue is for tickets about the DBD-Oracle CPAN distribution.

Report information
The Basics
Id: 7821
Status: resolved
Priority: 0/
Queue: DBD-Oracle

People
Owner: Nobody in particular
Requestors: stevelihn [...] rocketmail.com
Cc:
AdminCc:

Bug Information
Severity: Normal
Broken in: 1.15
Fixed in: (no value)



Subject: test failure on 10g build
I just built DBD::Oracle against Oracle 10g on Windows XP. The build went fine. Looks like this module can wrok with 10g pretty easily. But I encountered a few things in test. First, line 41 of general.t gives me hard time to pass. I have to modify the code to the following to pass the test. It seems that the "unrecognised attribute" message is not in $@, but has to be caught by $SIG{__WARN__}. { my $rc; $SIG{__WARN__} = sub { $rc = $_[0] }; eval { $p1=$sth->{NUM_OFFIELDS_typo} }; ok(0, $rc =~ /attribute/ ? 1 : 0); } Second, line 112 of ph_type.t responded with "not strip trailing space". This is contrary to 8i and 9i behavior. This seems odd. But I need to do more investigation on Metalink. --Steve Lihn
From: stevelihn [...] rocketmail.com
Show quoted text
> $SIG{__WARN__} = sub { $rc = $_[0] };
This line should read, local $SIG{__WARN__} = sub { $rc = $_[0] }; --Steve Lihn
The NUM_OFFIELDS_typo test was affected by a recent change in the DBI. It has been removed from the current release candidate. The blank stripping behaviour change seems to be an undocumented behaviour change in Oracle. I've added the local to the __WARN__ now. Thanks.