Hi Mark
On Sat, 2008-03-15 at 22:52 -0400, MARKSTOS via RT wrote:
Show quoted text> Queue: CGI-Session
> Ticket <URL:
http://rt.cpan.org/Ticket/Display.html?id=21981 >
>
> I see there's a fix for this in SVN and some passing tests. Would it be
> too much hassle to make this warning the test emits go away? Some users
> could take it to indicate a problem.
>
> I looked at distributing Test::Warn for this purpose, but it requires a
> string of dependencies itself.
>
> Can't locate package Encode::Encoding for @Encode::utf8::ISA at (eval
> 19) line 1.
> Can't locate package Encode::Encoding for @Encode::utf8::ISA at (eval
> 19) line 1.
> (in cleanup) Can't locate object method "renewed" via package
> "Encode::utf8" (perhaps you forgot to load "Encode::utf8"?) at (eval 19)
> line 1.
> (in cleanup) Can't locate object method "renewed" via package
> "Encode::utf8" (perhaps you forgot to load "Encode::utf8"?) at (eval 19)
> line 1.
My first thought was to patch the text associated with the test to add:
Warnings expected. Consult docs re 'utf8'
I'll do that now in SVN.
Also, let me copy here for the record the rationale I posted to the
mailing list for these warnings:
<==8><==>
The new test program I've written passes all tests, but still emits a
deliberate warning message. Here's my thinking:
o Tests and installation should not fail in all those cases people are
not using utf8
o References to utf8 are on RT, the docs (with my patch), and this email
list's archive, to help people diagnose problems
o My new test program passes all tests so automated testing and
installation can continue, assuming all other problems are fixed, of
course
o The new test emits a warning so anyone watching the tests' progress
will see a reference to utf8 being problematic
o Clearly I do not have a solution to this problem, but I have no
intention of diving into the innards of utf8-related code, so I see the
new test as simply giving people more warning about utf8 problems than
they've had up to now
<==8><==>
Sure, users can be confused, but to me that's better than having their
programs behave mysteriously over a known problem.
--
Ron Savage
ron@savage.net.au
http://savage.net.au/index.html