Merge 'trunk' into 'sqla_1.50_compat'

Merge 'trunk' into 'sqla_1.50_compat'

r5340@vger:  mendel | 2008-12-20 22:01:53 +0100
 r5134@vger (orig r5117):  ribasushi | 2008-11-13 08:34:26 +0100
 result_class goodies by Caelum:
 - result_class can now be specified as a search attribute, attaching it to the returned resultset
 - the specified class is now autoloaded via ensure_loaded()

 r5136@vger (orig r5119):  ribasushi | 2008-11-13 19:34:29 +0100
 resultset attribute from can now take a scalarref and use it verbatim
 r5145@vger (orig r5128):  ribasushi | 2008-11-14 10:46:00 +0100
 Document the new from usage
 r5152@vger (orig r5134):  ribasushi | 2008-11-14 18:44:25 +0100
 we use register_extra_source() now
 r5162@vger (orig r5140):  castaway | 2008-11-15 22:08:30 +0100
 Added failing create-multi test

 r5163@vger (orig r5141):  ribasushi | 2008-11-16 10:53:07 +0100
 whops
 r5166@vger (orig r5142):  norbi | 2008-11-16 20:12:42 +0100
  r5165@vger:  mendel | 2008-11-16 20:11:42 +0100
   * Merged in changes from 'resultsetcolumn_custom_columns' branch,

 r5179@vger (orig r5149):  ribasushi | 2008-11-16 23:14:13 +0100
 Revert castaway's test - mildly bogus
 r5180@vger (orig r5150):  ribasushi | 2008-11-16 23:16:31 +0100
 Add new test relationship - a track can have a relates single_cd
 r5181@vger (orig r5151):  ribasushi | 2008-11-16 23:23:38 +0100
 might_have test for castaway to break
 r5182@vger (orig r5152):  ribasushi | 2008-11-17 01:33:51 +0100
 rip away a horribly wrong create_via_update test (will pass when multicreate is merged)
 r5183@vger (orig r5153):  ribasushi | 2008-11-17 02:00:28 +0100
 One more (passing) multicreate test and a bit of cleanup
 r5184@vger (orig r5154):  ribasushi | 2008-11-17 02:59:53 +0100
 Two more sets of might_have - has_many relationships for extra tests
 r5185@vger (orig r5155):  ribasushi | 2008-11-17 03:09:18 +0100
 Two failing multicreate tests (the root cause of castaway's problem)
 r5186@vger (orig r5156):  ribasushi | 2008-11-17 03:48:57 +0100
 Silence cdbi tests like everything else
 r5187@vger (orig r5157):  ribasushi | 2008-11-17 04:01:39 +0100
 Failing tests by zby, showing that recursing insert() gets in the way of some exotic insert scenarious - waiting for mst to decide course of action
 r5188@vger (orig r5158):  ribasushi | 2008-11-17 12:04:54 +0100
 Extend might_have test with ideas from zby
 r5192@vger (orig r5161):  ribasushi | 2008-11-20 11:25:32 +0100
 Trunk passes tests again - todoify everything multicreate related to branch it out, as the task turned out to be more complex (no indentation to aid future merging)
 r5195@vger (orig r5164):  ribasushi | 2008-11-20 12:01:14 +0100
 CDBI-compat tests by Dave Horwoth:

     OK. I've attached a patch to t/cdbi-t/15-accessor.t that does several
     things:

     (1) Fixes the way the arguments are built so the sheep argument is
     correct in the tests.

     (2) Adds explicit tests for the number of sheep so we can be sure which
     database record is actually returned.

     (3) Adds tests for find_or_create() with modified accessor names.

     (4) Fixes the test for search() to report all errors

     I still don't fully understand what's going on in the output below but I
     think they are genuine test failures. I'd appreciate it if somebody else
     could run the tests to make sure the results are not some artefact of my
     test environment. I also attached a copy of the modified test script so
     you don't even have to apply the patch first :)

     I don't know what the DestroyWarning at the end is all about.

 r5196@vger (orig r5165):  ribasushi | 2008-11-20 12:09:11 +0100
 Minor doc patch by Caelum
 r5199@vger (orig r5168):  castaway | 2008-11-20 14:49:39 +0100
 Fix pod errors so that some storage subclasses show up in cpan properly

 r5291@vger (orig r5222):  ash | 2008-12-03 18:23:00 +0100
 Make the many-to-many warning use warnings::register;
 r5317@vger (orig r5239):  jnapiorkowski | 2008-12-15 23:02:19 +0100
 changes to replication so that if a replicant is offline when we do the initial connection (or if we need to globally reconnect later) the connection does not die.  This is different from when a replicant is available just not listening, or if the replicant is too laggy.  Thanks David Steinbrunner for the patch suggestions.
 r5323@vger (orig r5245):  groditi | 2008-12-16 23:56:07 +0100
  r24740@martha (orig r5230):  groditi | 2008-12-09 20:28:10 -0500
  fix for bug. all tests seem to pass, we still need a new test and more research

 r5324@vger (orig r5246):  groditi | 2008-12-16 23:56:29 +0100
  r24754@martha (orig r5241):  groditi | 2008-12-16 16:27:06 -0500
  Introduce 'any_null_means_no_value' option to eliminate wasteful queries. The option is off by default and must be explicitly turned on. Tests, + docs included

 r5325@vger (orig r5247):  groditi | 2008-12-16 23:57:11 +0100
  r24756@martha (orig r5243):  groditi | 2008-12-16 17:42:10 -0500
  rename option to undef_on_null_fk and make it default for belongs_to

 r5326@vger (orig r5248):  groditi | 2008-12-16 23:57:34 +0100
  r24757@martha (orig r5244):  groditi | 2008-12-16 17:52:12 -0500
  minor typo and style change

 r5328@vger (orig r5250):  groditi | 2008-12-17 00:27:57 +0100
 somehow i messed up the merge. this fixes it
 r5329@vger (orig r5251):  ash | 2008-12-17 00:45:51 +0100
 Try to fix test on 5.10
 r5332@vger (orig r5254):  ribasushi | 2008-12-18 12:28:38 +0100
 Some cleanups to the m2m warnings test
 r5336@vger (orig r5258):  jnapiorkowski | 2008-12-19 20:52:44 +0100
 more noise debugging messages if debug is on, minor doc tweaks, changes so that the fake sqlite tests will work and laid the groundwork for replication without dbatabase native replication support.
 r5337@vger (orig r5259):  jnapiorkowski | 2008-12-19 21:48:35 +0100
 altered schema->populate so that it is a very thin wrapper on top of resultset->populate and changed resultset_populate so that is accepted both the arrayref of hashes and the arrayref of arrayref style of args.  Documented this, updated the tests a bit to make sure it is all good.
 r5338@vger (orig r5260):  ribasushi | 2008-12-19 22:09:17 +0100
 Todoify cdbi failing tests, waiting for schwern