rcaputo/poe

report generator did not know where to find it, so it silently
reported an empty-versioned POE.  This change moves the POE module
load before tests, and makes it a fatal error if POE can't be loaded.
It also includes the new library directory, so POE can be loaded.
This should resolve RT ticket 6143.

Suggested by Philip Gwyn on POE's mailing list.

environment variables.  This commit corrects the omission and revises
the introduction to POE's assertion and tracing flags.

broken HTTP::Request object.  This commit should fix it.

effect.  This is in response to a comment by Matt Cashner.

generator's harness.  I am hoping this will solve some of the
differences in behavior between "make test" and "make testreport".

Whereas they pass on Mac OS X, the calls will fail without these
parameters on less forgiving systems like FreeBSD.

POE::Preprocessor with PerlApp.  This commit notes the dynamically
loaded POE::Resource classes, and it relaxes the error checking for
missing modules in POE::Resources.

current standard getaddrinfo().  The new call is heavy on
context-sensitive features, but it combines gethostbyname2() and
pack_sockaddr_in6() so it's not all bad.  Also, Solaris should like
it, so this closes RT#6191.

Oh, and I made the t/29_sockfact6.t skip messages a little more
informative.

As usual, CVS is committing touched files along for the ride.  It
doesn't matter that their content hasn't changed.

ActivePerl.  Updated the dependencies versions to request the most
recent ActivePerl LWP modules.  Skip the failing test on ActivePerl
versions before 5.008003.  The things I do for a clean CPAN Testers
bill of health....

single select_read() was turned on and off.  This commit moves idle
detection in Loop::Event's dispatcher forward a little so it's sure to
run all the time.

locations.

more machine friendly.

…s like testreport work as expected

…s and warnings caused by use of the old State parameters in Wheels. all tests pass in 5.004_005, 5.005_004, 5.6.2, and 5.8.3

…n bug in perls older than 5.6. the workarounds are commented with nice blocks explaining the problem. the only real solution is to get a perl without the bug. sorry folks. there's only so much i can do.

… qr// on perls that support qr//. tested on 5.004 and 5.8.3

… "no warnings" calls that were breaking poe on perls older than 5.6. replace them with $^W invocations as suggested by perllexwarn. second, fix the TRACE_FILE magic that was causing older perls to lose their mind. the comments here explain this in detail so i wont bother explaining it here. its kind of magic but works well. i have tested all this on 5.004 and 5.8.3 and all is well there (at least in relation to this patch - theres more to come in 5.004 land)

… perls

sungo's suggestion.