Commits on Jun 11, 2009
  1. @ribasushi

    Release 0.08106

    ribasushi committed Jun 11, 2009
  2. @ribasushi

    Factor out as_query properly

    ribasushi committed Jun 11, 2009
  3. @arcanez
  4. @ribasushi
  5. @frioux
  6. @ribasushi

    Release 0.08105

    ribasushi committed Jun 11, 2009
  7. @ribasushi

    Changes

    ribasushi committed Jun 11, 2009
  8. @ribasushi

    Merge 'type_aware_update' into 'trunk'

    r5713@Thesaurus (orig r5712):  ribasushi | 2009-03-08 23:53:28 +0100
    Branch for datatype-aware updates
    r6604@Thesaurus (orig r6603):  ribasushi | 2009-06-10 18:08:25 +0200
    Test for type-aware update
    r6607@Thesaurus (orig r6606):  ribasushi | 2009-06-10 19:57:04 +0200
    Datatype aware update works
    r6609@Thesaurus (orig r6608):  ribasushi | 2009-06-10 20:06:40 +0200
    Whoops
    r6614@Thesaurus (orig r6613):  ribasushi | 2009-06-11 09:23:54 +0200
    Add attribute doc
    r6620@Thesaurus (orig r6619):  ribasushi | 2009-06-11 16:15:53 +0200
    Use equality, not comparison
    ribasushi committed Jun 11, 2009
  9. @ribasushi

    Use equality, not comparison

    ribasushi committed Jun 11, 2009
  10. @ribasushi
  11. @ribasushi

    IC::DT changes:

    Switch SQLite storage to DT::F::SQLite
    Fix exception when undef_if_invalid and timezone are both set on a column
    Split t/89inflate_datetime into separate tests
    Adjust makefile author dependencies
    ribasushi committed Jun 11, 2009
  12. @ribasushi
  13. @ribasushi

    Move around inflation tests

    ribasushi committed Jun 11, 2009
  14. @ribasushi

    Add attribute doc

    ribasushi committed Jun 11, 2009
Commits on Jun 10, 2009
  1. @ribasushi

    Whoops

    ribasushi committed Jun 10, 2009
  2. @ribasushi

    Datatype aware update works

    ribasushi committed Jun 10, 2009
  3. @ribasushi

    Test for type-aware update

    ribasushi committed Jun 10, 2009
  4. @ribasushi

    Merge 'trunk' into 'type_aware_update'

    r5719@Thesaurus (orig r5718):  ribasushi | 2009-03-09 13:22:54 +0100
    Disable reconnect test except fpor when AUTHORS is set
    r5720@Thesaurus (orig r5719):  ribasushi | 2009-03-09 13:23:12 +0100
    One more cdbi-compat author req
    r5721@Thesaurus (orig r5720):  ribasushi | 2009-03-09 13:42:12 +0100
    Use homegrown AUTHOR detection
    r5722@Thesaurus (orig r5721):  ribasushi | 2009-03-09 17:29:30 +0100
    Strip bogus Data::Dump requires
    r5728@Thesaurus (orig r5727):  groditi | 2009-03-09 23:33:50 +0100
    remove (groditi) instances
    r5733@Thesaurus (orig r5732):  wreis | 2009-03-11 01:56:06 +0100
    fixed View's doc | added me to contributors list
    r5734@Thesaurus (orig r5733):  ribasushi | 2009-03-11 07:57:05 +0100
    Moose-related test hierarchy not removed at r4326
    r5735@Thesaurus (orig r5734):  ribasushi | 2009-03-11 10:43:40 +0100
    Allow accessors type single to return defined but false objects (needed for null-pattern style components)
    r5736@Thesaurus (orig r5735):  ribasushi | 2009-03-11 11:03:06 +0100
    Replace the fallback _dbh_last_insert_id with an explicit exception.
    Currently all known Storage::DBI engines provide their own _dbh_last_insert_id, except for Storage::DBI::Sybase, for which the old sqlite-ish fallback doesn't work anyway: http://rt.cpan.org/Public/Bug/Display.html?id=40265
    r5737@Thesaurus (orig r5736):  wreis | 2009-03-11 20:25:25 +0100
    removed configure_sqlt method
    r5738@Thesaurus (orig r5737):  ribasushi | 2009-03-12 00:23:18 +0100
    Properly returb undef/null on single rewlationship accessor
    r5744@Thesaurus (orig r5743):  matthewt | 2009-03-12 19:35:16 +0100
    warning in author mode to avoid confusing the easily confused
    r5745@Thesaurus (orig r5744):  castaway | 2009-03-12 21:24:55 +0100
    Document how to overload connect/connection
    
    r5746@Thesaurus (orig r5745):  castaway | 2009-03-12 21:36:41 +0100
    Add docs in various places to point out how to get the total count of rows for a paged resultset
    
    r5747@Thesaurus (orig r5746):  castaway | 2009-03-12 23:33:42 +0100
    Add doc on how conditions and attrs are merged when chaining.
    
    r5748@Thesaurus (orig r5747):  ribasushi | 2009-03-13 00:30:26 +0100
    Move author warning in front of auto_install
    r5754@Thesaurus (orig r5753):  jmmills | 2009-03-13 13:32:35 +0100
    Added test for source sub-class register_extra_sources warning.
    Sub-classes should not generate this warning - bug via non-inhertiance sorting via register_source.
    
    r5759@Thesaurus (orig r5758):  arcanez | 2009-03-15 17:14:56 +0100
    update Makefile.PL to require latest C3, C3::Componentised, MRO::Compat
    r5761@Thesaurus (orig r5760):  arcanez | 2009-03-16 20:53:34 +0100
     * change search_literal to use \[] when passing into search (help with binding order)
     * provide documentation on how to use search instead of search_literal
    
    
    r5762@Thesaurus (orig r5761):  plu | 2009-03-17 16:31:08 +0100
    putting IC::DateTime locale, timezone or floating_tz_ok attributes into extra => {} has been deprecated. The new way is to put these things directly into the columns definition
    r5764@Thesaurus (orig r5763):  ribasushi | 2009-03-18 00:57:49 +0100
    Basic binary op logic FAIL
    r5769@Thesaurus (orig r5768):  rafl | 2009-03-18 20:02:07 +0100
    Use MRO::Compat instead of Class::C3 directly.
    r5770@Thesaurus (orig r5769):  rafl | 2009-03-18 20:02:23 +0100
    Depend on the latest C3::Componentised so we use mro:: on 5.10 and Class::C3 on 5.8.
    r5773@Thesaurus (orig r5772):  robkinyon | 2009-03-19 18:05:00 +0100
    Special-case handling for (undef,undef) passed to search_rs()
    r5777@Thesaurus (orig r5776):  robkinyon | 2009-03-19 21:35:37 +0100
    Moved some SQL parsing of ORDER BY from _select_args() to retrieve_from_sql() where it belongs
    r5779@Thesaurus (orig r5778):  robkinyon | 2009-03-19 21:52:54 +0100
    Commit another fix for arcanez to test
    r5780@Thesaurus (orig r5779):  robkinyon | 2009-03-19 22:07:06 +0100
    Commit another fix for arcanez to test
    r5785@Thesaurus (orig r5784):  wdh | 2009-03-20 16:23:40 +0100
    clarify need for scalarref when using db functions with 'default_value' attribute
    r5787@Thesaurus (orig r5786):  solomon | 2009-03-20 22:41:49 +0100
    Add cookbook entry for dealing with runaway prepared statement cache
    r5792@Thesaurus (orig r5791):  groditi | 2009-03-22 02:10:30 +0100
    adding test for exception if nullable rel call is not explicitly defined as undef on insert
    r5793@Thesaurus (orig r5792):  ribasushi | 2009-03-22 22:09:22 +0100
    New SQLA::Test handles () differences *much* better, thus fixing the tests
    r5804@Thesaurus (orig r5803):  groditi | 2009-03-23 19:16:41 +0100
    skip test doc issue relating to db-side defaults
    r5809@Thesaurus (orig r5808):  jmmills | 2009-03-24 08:49:54 +0100
    Fixed _register_source to not produce 'register_extra_source' warnings
    when "registering class" isa (or is a sub-class) source class.
    This is not exactly an elegant fix as I had to pass a hash key (as _to_register via $params) with the original source class instance; I was not able to determine inheritance via the result_source_instance (ResultSource::Table) object.
     
    r5816@Thesaurus (orig r5815):  robkinyon | 2009-03-26 02:09:55 +0100
    Deprecated search_like()
    r5817@Thesaurus (orig r5816):  lukes | 2009-03-26 10:13:18 +0100
    changed my nick in contribs section
    r5820@Thesaurus (orig r5819):  castaway | 2009-03-26 14:54:09 +0100
    Improved  metapod docs
    
    r5823@Thesaurus (orig r5822):  nigel | 2009-03-27 13:21:14 +0100
     r11142@hex:  nigel | 2009-03-27 12:20:55 +0000
     Removed Cookbook section on pagers that stated page attribute was not required.
     [Since the pager throws an exception if the pager attribute is not specified, 
     this was completely wrong]
    
    r5824@Thesaurus (orig r5823):  norbi | 2009-03-27 13:30:00 +0100
     r5969@vger:  mendel | 2009-03-27 13:29:24 +0100
      * Fixed the alphabetical order of the contributors list (that I screwed up, sorry).
    
    r5831@Thesaurus (orig r5830):  ribasushi | 2009-03-28 22:46:34 +0100
    Bump SQLA and CAG dependencies
    r5832@Thesaurus (orig r5831):  ribasushi | 2009-03-28 23:05:35 +0100
    Clarify DBIC behavior wrt RDBMS default values
    r5833@Thesaurus (orig r5832):  ribasushi | 2009-03-29 10:31:47 +0200
    detabify
    r5836@Thesaurus (orig r5835):  ribasushi | 2009-03-29 10:49:32 +0200
    detabify
    r5837@Thesaurus (orig r5836):  ribasushi | 2009-03-29 12:06:02 +0200
    Revert r5808 - this is NOT a bug, the warning IS correct
    r5838@Thesaurus (orig r5837):  ribasushi | 2009-03-29 12:18:19 +0200
    Adjust parenthesis to pass with latest SQLA
    r5839@Thesaurus (orig r5838):  ribasushi | 2009-03-29 12:22:49 +0200
    Remove failing test before branching
    r5847@Thesaurus (orig r5846):  ribasushi | 2009-03-30 01:05:37 +0200
    Take experimental off TxnScopeGuard
    r5848@Thesaurus (orig r5847):  ribasushi | 2009-03-30 01:44:40 +0200
    Dev-Release 0.08099_08
    r5852@Thesaurus (orig r5851):  castaway | 2009-03-31 08:41:42 +0200
    Fix my email address in AUTHOR
    
    r5859@Thesaurus (orig r5858):  plu | 2009-04-07 16:15:32 +0200
    Added caveat notes to dual trick in cookbook
    r5864@Thesaurus (orig r5863):  jmmills | 2009-04-08 21:28:16 +0200
     r81350@ofelia (orig r5840):  ribasushi | 2009-03-29 03:30:21 -0700
     Re-add failing test
    
    r5875@Thesaurus (orig r5874):  arcanez | 2009-04-09 15:35:46 +0200
    require DBD::SQLite 1.21+
    r5885@Thesaurus (orig r5884):  caelum | 2009-04-13 14:47:47 +0200
    add test for connect_info hashref
    r5887@Thesaurus (orig r5886):  pcc | 2009-04-13 20:35:51 +0200
    Correctly propagate forced left joins through arrayrefs and hashrefs
    r5889@Thesaurus (orig r5888):  caelum | 2009-04-15 15:47:53 +0200
    duplicate
    r5891@Thesaurus (orig r5890):  plu | 2009-04-17 10:10:19 +0200
    Added update_or_new to ResultSet.pm
    r5903@Thesaurus (orig r5902):  ilmari | 2009-04-18 20:38:44 +0200
    Don't warn when reregistering a source under the same name
    r5913@Thesaurus (orig r5912):  matthewt | 2009-04-19 16:52:15 +0200
    0.08100 release
    r5916@Thesaurus (orig r5915):  arcanez | 2009-04-21 00:07:07 +0200
    move load_optional_class into Class::C3::Componentised
    
    r5923@Thesaurus (orig r5922):  ribasushi | 2009-04-21 10:37:11 +0200
    New SQLite kicks ass - remove some warning silencers and reinstantiate tests
    r5924@Thesaurus (orig r5923):  ribasushi | 2009-04-21 10:51:03 +0200
    Multiple cdbi-compat test cleanups
    Clarify search_like deprecation
    r5925@Thesaurus (orig r5924):  ribasushi | 2009-04-21 10:51:59 +0200
    More author test-requires
    r5926@Thesaurus (orig r5925):  ribasushi | 2009-04-21 10:53:35 +0200
    t/ janitorial work
    r5936@Thesaurus (orig r5935):  ribasushi | 2009-04-21 16:17:04 +0200
    Fix dumb omission in t/89dbicadmin.t
    Adjust svn:ignore
    r5937@Thesaurus (orig r5936):  peter | 2009-04-21 17:52:58 +0200
    added IRC nick to contributor list
    r5944@Thesaurus (orig r5943):  arcanez | 2009-04-22 05:21:28 +0200
    un-todo these tests, load_optional_class is fixed
    r5945@Thesaurus (orig r5944):  ribasushi | 2009-04-22 10:14:13 +0200
    RT45195 various indexer fixes
    r5946@Thesaurus (orig r5945):  ribasushi | 2009-04-22 10:18:14 +0200
    Another indexer trick
    r5949@Thesaurus (orig r5948):  ribasushi | 2009-04-23 00:25:32 +0200
    This code belongs in t/
    r5951@Thesaurus (orig r5950):  ribasushi | 2009-04-23 00:28:26 +0200
     r5846@Thesaurus (orig r5845):  ribasushi | 2009-03-29 16:55:03 +0200
     Fix test - inane assumption on my part
    
    r5952@Thesaurus (orig r5951):  ribasushi | 2009-04-23 00:33:42 +0200
     r5863@Thesaurus (orig r5862):  jmmills | 2009-04-08 21:26:57 +0200
     patch for no-op, further testing required.
     
    
    r5953@Thesaurus (orig r5952):  ribasushi | 2009-04-23 00:34:11 +0200
     r5927@Thesaurus (orig r5926):  ribasushi | 2009-04-21 11:57:03 +0200
     Fix number of tests
    
    r5954@Thesaurus (orig r5953):  ribasushi | 2009-04-23 00:35:52 +0200
     r5928@Thesaurus (orig r5927):  ribasushi | 2009-04-21 12:42:14 +0200
     Clarify _register_source code, no functional changes
    
    r5955@Thesaurus (orig r5954):  ribasushi | 2009-04-23 00:35:59 +0200
     r5929@Thesaurus (orig r5928):  ribasushi | 2009-04-21 12:53:25 +0200
     Back out my bogus test from r4902 - this warning was never supposed to be here in the first place - we are essentially dynamically subclassing Artist (although in a very sloppy way)
    
    r5956@Thesaurus (orig r5955):  ribasushi | 2009-04-23 00:37:43 +0200
    
    r5957@Thesaurus (orig r5956):  ribasushi | 2009-04-23 00:37:49 +0200
     r5931@Thesaurus (orig r5930):  ribasushi | 2009-04-21 13:06:25 +0200
     Sloppy merge - adding imlari's fix from r5902
    
    r5959@Thesaurus (orig r5958):  ribasushi | 2009-04-23 00:48:08 +0200
    Fix botched merge again - sorry r5959-5956 are actually a single commit
    r5967@Thesaurus (orig r5966):  gphat | 2009-04-23 22:49:04 +0200
    Fix busted get_column when using +select (and friends)
    r5968@Thesaurus (orig r5967):  arcanez | 2009-04-23 22:58:59 +0200
    update Changes for moving load_optional_class
    r5976@Thesaurus (orig r5975):  ribasushi | 2009-04-24 01:47:04 +0200
    Whops
    r5978@Thesaurus (orig r5977):  mo | 2009-04-24 16:24:26 +0200
    patch for DBI.pm so store_column is called only once on create() and tests for that
    r5979@Thesaurus (orig r5978):  mo | 2009-04-24 16:27:20 +0200
    cleanup 72pg.t
    r5980@Thesaurus (orig r5979):  ribasushi | 2009-04-24 17:22:05 +0200
    Cosmetic change
    r5987@Thesaurus (orig r5986):  castaway | 2009-04-25 11:59:28 +0200
    Docs on using multiple db schemas (from abraxxa maybe, I forget)
    Docs on setting result_class from agaran
    
    r5988@Thesaurus (orig r5987):  ribasushi | 2009-04-27 10:03:58 +0200
    Refactor the (almost obsolete) DBD::SQLite check not to fork() within win32
    r5989@Thesaurus (orig r5988):  ribasushi | 2009-04-27 10:05:43 +0200
    Whops, we don't exit under win32
    r5990@Thesaurus (orig r5989):  ribasushi | 2009-04-27 10:14:38 +0200
    More AUTHOR dependencies pulled from Replicated.pm
    Up MI version to 0.79 (not going further as auto_install is broken again, need a permanent solution)
    Reorder configure_requires() as per Alias
    Remove Meta_TupleKeys munging as per Alias
    r5991@Thesaurus (orig r5990):  ribasushi | 2009-04-27 11:11:15 +0200
    Release 0.08101
    r5995@Thesaurus (orig r5994):  arcanez | 2009-04-28 09:05:53 +0200
    change from DB::Schema to My::Schema
    add ::Result:: as necessary (load_namespaces instead of load_classes)
    
    r5996@Thesaurus (orig r5995):  ribasushi | 2009-04-28 11:12:33 +0200
    Fix a todo and some minor test adjustments
    r5997@Thesaurus (orig r5996):  ribasushi | 2009-04-28 11:35:31 +0200
    Bogus test dependency
    r6003@Thesaurus (orig r6002):  arcanez | 2009-04-28 22:20:42 +0200
    fix for columns => [ ... ] issue
    r6014@Thesaurus (orig r6013):  arcanez | 2009-04-29 01:16:48 +0200
    test for prefetch by columns
    r6016@Thesaurus (orig r6015):  ribasushi | 2009-04-29 10:22:16 +0200
    test and patch for failing mini-prefetch via columns (arcanez++)
    r6017@Thesaurus (orig r6016):  ribasushi | 2009-04-29 10:23:25 +0200
    Add two TODOs by arcanez (not sure about those)
    r6018@Thesaurus (orig r6017):  ribasushi | 2009-04-29 10:24:20 +0200
    Add TODOs illustrating a problem with as_query
    r6019@Thesaurus (orig r6018):  arcanez | 2009-04-29 14:47:03 +0200
    untodo/fix test
    r6020@Thesaurus (orig r6019):  ribasushi | 2009-04-29 15:39:27 +0200
    Strip out inane todo
    r6021@Thesaurus (orig r6020):  ribasushi | 2009-04-29 15:40:03 +0200
    Forgotten debugging
    r6022@Thesaurus (orig r6021):  ribasushi | 2009-04-29 15:41:37 +0200
    Fixes to massive breakage introduced by 5948  ribasushi--
    r6023@Thesaurus (orig r6022):  ribasushi | 2009-04-29 15:47:23 +0200
    Changes
    r6024@Thesaurus (orig r6023):  jasonmay | 2009-04-29 17:18:11 +0200
    allow multiple classless sources to be registered without dying
    
    r6025@Thesaurus (orig r6024):  jasonmay | 2009-04-29 18:47:26 +0200
    add jasonmay to list of contributors
    
    r6034@Thesaurus (orig r6033):  ribasushi | 2009-04-30 08:53:51 +0200
    Add tests for r6002
    r6036@Thesaurus (orig r6035):  ribasushi | 2009-04-30 10:12:23 +0200
    Revert addition of superficial relationship in r5886
    r6039@Thesaurus (orig r6038):  ribasushi | 2009-04-30 10:28:04 +0200
    Release 0.08102
    r6110@Thesaurus (orig r6109):  plu | 2009-05-02 21:11:40 +0200
    Failing testcase for using subselect (as_query)
    r6111@Thesaurus (orig r6110):  ribasushi | 2009-05-03 02:00:19 +0200
    Failing test without immediate fixes go to branches, not to trunk
    r6114@Thesaurus (orig r6113):  ribasushi | 2009-05-03 10:23:28 +0200
    Bump SQLA ependencies so parenthesis_significant is guaranteed to be there
    r6127@Thesaurus (orig r6126):  arcanez | 2009-05-05 09:27:56 +0200
     r6037@mullet (orig r6036):  arcanez | 2009-04-30 01:24:41 -0700
     branch to work on Sybase/MSSQL subtleties
     r6038@mullet (orig r6037):  arcanez | 2009-04-30 01:27:11 -0700
     jump to ::DBI::Sybase::MSSQL if we are using MSSQL through Sybase
     r6062@mullet (orig r6061):  arcanez | 2009-04-30 14:05:24 -0700
     fixes for MSSQL via Sybase
     r6126@mullet (orig r6125):  arcanez | 2009-05-05 00:27:13 -0700
     add a line to Changes and add a CAVEAT
    
    r6164@Thesaurus (orig r6163):  ribasushi | 2009-05-07 19:09:01 +0200
     r6115@Thesaurus (orig r6114):  plu | 2009-05-03 10:39:16 +0200
     new branch to fix $rs->update and $rs->delete using the new as_query method
     
     r6116@Thesaurus (orig r6115):  plu | 2009-05-03 10:52:07 +0200
     Methods update/delete on resultset use now new as_query method to updated/delete properly on joined/prefetched resultset using a subquery. Therefore some tests have been added and some have been changed as well as the warnings around $rs->update/delete have been removed. Cheers!
     r6117@Thesaurus (orig r6116):  plu | 2009-05-03 11:13:48 +0200
     Using "is" instead of "cmp_ok"
     r6160@Thesaurus (orig r6159):  ribasushi | 2009-05-07 11:58:14 +0200
     Back out skip_parens support in as_query
     r6161@Thesaurus (orig r6160):  ribasushi | 2009-05-07 19:00:48 +0200
     This test is completely borked, needs a rewrite
     r6162@Thesaurus (orig r6161):  ribasushi | 2009-05-07 19:07:19 +0200
     Temporary fix or the IN ( ( ... ) ) problem until we get proper SQLA AST (needs SQLA released with commit 6158 to work)
    
    r6165@Thesaurus (orig r6164):  ribasushi | 2009-05-07 19:11:46 +0200
    Changes, remove merged branch
    r6169@Thesaurus (orig r6168):  ribasushi | 2009-05-07 19:24:54 +0200
    Bump SQLA dependency so -in/-between workarounds overload properly
    r6172@Thesaurus (orig r6171):  ribasushi | 2009-05-07 20:49:26 +0200
    Cookbook cleanup
    r6174@Thesaurus (orig r6173):  ribasushi | 2009-05-08 10:13:30 +0200
    Throw away some debugging code
    r6175@Thesaurus (orig r6174):  ribasushi | 2009-05-08 10:21:53 +0200
    Documentation patch by nniuq
    r6176@Thesaurus (orig r6175):  plu | 2009-05-08 10:30:20 +0200
    Set NLS_LANG so we  have a predictable date format when using MON
    r6177@Thesaurus (orig r6176):  ribasushi | 2009-05-08 12:15:15 +0200
    Fix POD
    r6179@Thesaurus (orig r6178):  jgoulah | 2009-05-08 16:27:49 +0200
    renaming rh performance test so it will show up at the end of test output
    r6195@Thesaurus (orig r6194):  caelum | 2009-05-09 15:46:55 +0200
    added postgres default port stuff to FAQ
    r6199@Thesaurus (orig r6198):  mo | 2009-05-10 17:37:16 +0200
    set_$rel accepts now a $link_vals hashref like add_to_$rel does
    r6202@Thesaurus (orig r6201):  ribasushi | 2009-05-10 19:57:31 +0200
    Require DBICTest.pm in all tests even if it is not needed at all
    r6203@Thesaurus (orig r6202):  ribasushi | 2009-05-10 20:17:09 +0200
    No more 'I forgot to run perl Makefile.PL'
    r6204@Thesaurus (orig r6203):  ribasushi | 2009-05-10 21:35:03 +0200
    Switch the m2m method warnings from warnings::register to $ENV{DBIC_METHOD_CLOBBER_OK} = 1
    r6206@Thesaurus (orig r6205):  ribasushi | 2009-05-11 12:50:54 +0200
    Change the makefile to test r6202
    r6218@Thesaurus (orig r6217):  ribasushi | 2009-05-12 07:02:14 +0200
    Trying to untangle failing merge (whitespace change)
    r6220@Thesaurus (orig r6219):  ribasushi | 2009-05-12 07:02:56 +0200
     r5757@Thesaurus (orig r5756):  arcanez | 2009-03-14 14:55:55 +0100
     created count_distinct branch
     r5758@Thesaurus (orig r5757):  arcanez | 2009-03-14 15:33:03 +0100
      * change count with group_by (distinct) to use a subquery
      * rewrite of _bind_to_sql (uses placeholders and bindvars)
      * tests for count distinct
      * fixed tests for from subquery 
     
     r5760@Thesaurus (orig r5759):  arcanez | 2009-03-16 16:48:28 +0100
     don't remove the where clause unless we're doing distinct, it needs to be there
     r5850@Thesaurus (orig r5849):  arcanez | 2009-03-30 21:40:05 +0200
      * add more tests
      * remove old cruft
      * remove old note
     
     r6035@Thesaurus (orig r6034):  ribasushi | 2009-04-30 09:10:36 +0200
     Add joined count test
     r6079@Thesaurus (orig r6078):  arcanez | 2009-05-01 08:41:34 +0200
     cleanup/fix some broken tests
     r6097@Thesaurus (orig r6096):  arcanez | 2009-05-01 19:37:04 +0200
     make sure merge bind
     test for aformentioned
     TODO count_joined test for a little while
     
     r6105@Thesaurus (orig r6104):  arcanez | 2009-05-02 02:33:49 +0200
     remove hackish ways
     
     r6106@Thesaurus (orig r6105):  arcanez | 2009-05-02 03:20:04 +0200
     more fixes to tests
     r6107@Thesaurus (orig r6106):  arcanez | 2009-05-02 03:33:47 +0200
     remove DBIx::Class::Storage::DBI::MultiDistinctEmulation
     r6111@Thesaurus (orig r6110):  ribasushi | 2009-05-03 02:00:19 +0200
     Failing test without immediate fixes go to branches, not to trunk
     r6114@Thesaurus (orig r6113):  ribasushi | 2009-05-03 10:23:28 +0200
     Bump SQLA ependencies so parenthesis_significant is guaranteed to be there
     r6148@Thesaurus (orig r6147):  ribasushi | 2009-05-06 17:40:31 +0200
     Add subquery/from test by michaelr (copied from subquery branch r5742)
     r6151@Thesaurus (orig r6150):  ribasushi | 2009-05-06 17:56:07 +0200
     TODOify sqla-dependent tests
     r6152@Thesaurus (orig r6151):  arcanez | 2009-05-06 18:33:58 +0200
     fix old test using new bind vars (no more interpolating)
     r6154@Thesaurus (orig r6153):  ribasushi | 2009-05-07 00:56:40 +0200
     Rename internal function to clarify what it does
     r6171@Thesaurus (orig r6170):  ribasushi | 2009-05-07 19:53:30 +0200
     Adjust tests for the IN fixes
     r6187@Thesaurus (orig r6186):  ribasushi | 2009-05-08 23:11:43 +0200
     Final count tests
     r6188@Thesaurus (orig r6187):  arcanez | 2009-05-09 03:50:12 +0200
     rewrite DISTINCT/COUNT(DISTINCT) Cookbook entries
     r6213@Thesaurus (orig r6212):  arcanez | 2009-05-11 22:41:21 +0200
     warn/die based on { select => { distinct => { } } }
     r6214@Thesaurus (orig r6213):  arcanez | 2009-05-11 23:21:11 +0200
     use carp instead of warn
     r6216@Thesaurus (orig r6215):  arcanez | 2009-05-11 23:45:05 +0200
     make sure we get just a string
     r6217@Thesaurus (orig r6216):  arcanez | 2009-05-11 23:52:11 +0200
     oops
    
    r6222@Thesaurus (orig r6221):  ribasushi | 2009-05-12 08:11:27 +0200
     r6112@Thesaurus (orig r6111):  nniuq | 2009-05-03 02:36:33 +0200
     Initially, fixes to enable saving of LOB types in Oracle.  Possibly timestamp tweaks.
     
     r6113@Thesaurus (orig r6112):  nniuq | 2009-05-03 03:52:29 +0200
     Support for saving CLOB and BLOB types in Oracle.
     
     r6118@Thesaurus (orig r6117):  nniuq | 2009-05-04 03:58:03 +0200
     Proper support for timestamp inflation.  Added last_updated_at to DBICTest::Schema::Track as a date by default, initialized in sqlite loader, redefined to timestamp for Oracle tests.
     
     r6119@Thesaurus (orig r6118):  nniuq | 2009-05-04 04:03:28 +0200
     Re-added last_updated_at to create table statement.
     
     r6134@Thesaurus (orig r6133):  nniuq | 2009-05-05 15:11:49 +0200
     Added self to contributors; clarified comment on :ora_types imports.
     
     r6182@Thesaurus (orig r6181):  nniuq | 2009-05-08 17:10:58 +0200
     Refactored to call _{inflate_to,deflate_from}_datetime through a _flate_or_fallback wrapper handling a parser's lack of support for the requested type.
     
     r6183@Thesaurus (orig r6182):  ribasushi | 2009-05-08 17:24:54 +0200
     Remove redundant var
     r6189@Thesaurus (orig r6188):  nniuq | 2009-05-09 03:59:20 +0200
     Changed test of lob values from is to ok on an eq expr, to avoid a huge got/expected diagnosis.  In doing so, discovered it was testing undef vs undef!  Whoops.  Fixed.
     
     r6210@Thesaurus (orig r6209):  nniuq | 2009-05-11 16:24:27 +0200
     Moved DBIC::SQL::Abstract inner classes to DBIx::Class::SQLAHacks namespace to decouple 41orrible.t tests from use of DBD::Oracle in Oracle Generic driver.
    
    r6224@Thesaurus (orig r6223):  ribasushi | 2009-05-12 08:13:31 +0200
     r6223@Thesaurus (orig r6222):  ribasushi | 2009-05-12 08:12:50 +0200
     Whops, forgotten post-merge tweaks
    
    r6226@Thesaurus (orig r6225):  ribasushi | 2009-05-12 08:43:58 +0200
    Cleanup
    r6227@Thesaurus (orig r6226):  ribasushi | 2009-05-12 09:42:45 +0200
    Shuffle delete tests, and sanify the delete related ones
    r6228@Thesaurus (orig r6227):  ribasushi | 2009-05-12 09:54:37 +0200
    Rename m2m warn disable envvar
    r6229@Thesaurus (orig r6228):  ribasushi | 2009-05-12 10:20:08 +0200
    Switch warn to carp and die to throw_exception where possible
    r6230@Thesaurus (orig r6229):  bricas | 2009-05-12 13:31:35 +0200
    added info to Changes from my DateTime::InflateColumn modifications
    r6231@Thesaurus (orig r6230):  nniuq | 2009-05-12 15:08:14 +0200
    Added synopsis of merged oracle-tweaks branch to Changes
    r6233@Thesaurus (orig r6232):  debolaz | 2009-05-12 17:17:09 +0200
    Test for failing code related to many to many in MySQL
    
    r6234@Thesaurus (orig r6233):  arcanez | 2009-05-12 17:33:56 +0200
    Changes for count_distinct branch merge
    r6235@Thesaurus (orig r6234):  ribasushi | 2009-05-12 18:09:14 +0200
    Some more die/croak conversions
    r6236@Thesaurus (orig r6235):  ribasushi | 2009-05-12 18:51:38 +0200
    Fix shot test
    r6244@Thesaurus (orig r6243):  ribasushi | 2009-05-13 03:02:11 +0200
    fix bogus test
    r6245@Thesaurus (orig r6244):  ribasushi | 2009-05-13 03:06:42 +0200
    Fix _select_for_update/delete - bring back old code, use subqueries only when resultset attributes call for it
    r6252@Thesaurus (orig r6251):  ribasushi | 2009-05-14 01:41:31 +0200
    Better testing and heuristics for subqueried _cond_for_update_delete
    r6253@Thesaurus (orig r6252):  ribasushi | 2009-05-14 01:54:41 +0200
    join can be found without seen_join
    r6255@Thesaurus (orig r6254):  ribasushi | 2009-05-14 12:12:44 +0200
    Sometimes _prepare_for_execute may not return a set of bind vars - guard against that
    r6256@Thesaurus (orig r6255):  ribasushi | 2009-05-14 12:14:19 +0200
    Clarify test
    r6264@Thesaurus (orig r6263):  ribasushi | 2009-05-14 15:40:19 +0200
     r6121@Thesaurus (orig r6120):  caelum | 2009-05-04 17:13:22 +0200
     Making new branch for storage tweaking
     r6122@Thesaurus (orig r6121):  caelum | 2009-05-04 20:07:47 +0200
     support hashrefs for connect_replicants
     r6123@Thesaurus (orig r6122):  caelum | 2009-05-04 23:07:43 +0200
     ::Replicated - test hashref for connect_replicants and croak on coderef, switch to MX::Types, make test less noisy
     r6143@Thesaurus (orig r6142):  caelum | 2009-05-06 05:13:56 +0200
     fix ::DBI::Replicated::all_storages
     r6144@Thesaurus (orig r6143):  caelum | 2009-05-06 05:25:04 +0200
     Replicated - fixup types and namespace::clean
     r6147@Thesaurus (orig r6146):  caelum | 2009-05-06 15:29:39 +0200
     ::DBI:Replicated - merge connect_info from master to replicants
     r6184@Thesaurus (orig r6183):  caelum | 2009-05-08 18:08:29 +0200
     support ::DBI::Replicated opts in connect_info
     r6190@Thesaurus (orig r6189):  caelum | 2009-05-09 05:31:15 +0200
     ::DBI::Replicated - add master_read_weight to ::Random balancer_type
     r6191@Thesaurus (orig r6190):  caelum | 2009-05-09 12:50:25 +0200
     ::DBI::Replicated - fix fallback to master, test for the warning, other cleanups
     r6193@Thesaurus (orig r6192):  caelum | 2009-05-09 13:52:52 +0200
     updated Changes
     r6194@Thesaurus (orig r6193):  caelum | 2009-05-09 14:21:44 +0200
     ::DBI::Replicated - don't build pool/balancer from connect_info unless necessary
    
    r6268@Thesaurus (orig r6267):  caelum | 2009-05-15 04:04:12 +0200
    minor replication changes - use a real hash merge, clarify master_read_weight, really done with this now.
    r6272@Thesaurus (orig r6271):  abraxxa | 2009-05-15 13:45:54 +0200
    added Static sub-classing DBIx::Class result classes section to the cookbook
    
    r6277@Thesaurus (orig r6276):  ribasushi | 2009-05-16 00:46:00 +0200
    Optimize some Ordered.pm code
    r6278@Thesaurus (orig r6277):  ribasushi | 2009-05-16 08:02:18 +0200
    Cleanup tests
    r6280@Thesaurus (orig r6279):  ribasushi | 2009-05-16 09:30:05 +0200
    Not sure what this part of the test is for, but it breaks custom resultsets, and the test passes without it. Removing as a possible remnant of an ancient civilization
    r6281@Thesaurus (orig r6280):  ribasushi | 2009-05-16 09:33:24 +0200
    Add default resultclass/resultsetclass to the entire test schema, as I am tired of typing extra shit for debugging purposes
    r6282@Thesaurus (orig r6281):  ribasushi | 2009-05-16 09:34:21 +0200
    Now we can do diag $rs->hri_dump, ain't that nice
    r6290@Thesaurus (orig r6289):  ribasushi | 2009-05-17 00:06:40 +0200
    Rewrite SqlMakerTest to fully depend on SQLA::Test
    Allow is_same_sql_bind to accept an arrayrefref instead of any pair of $sql, \@bind
    r6299@Thesaurus (orig r6298):  frew | 2009-05-18 18:16:44 +0200
    Fix order by clauses for MSSQL
    r6304@Thesaurus (orig r6303):  frew | 2009-05-18 21:43:48 +0200
    Test to show problem
    r6305@Thesaurus (orig r6304):  ribasushi | 2009-05-18 22:12:35 +0200
    Failing tests go to branches
    r6306@Thesaurus (orig r6305):  ribasushi | 2009-05-18 22:14:26 +0200
    Require bugfixed sqla
    r6312@Thesaurus (orig r6311):  ribasushi | 2009-05-18 23:05:10 +0200
     r6157@Thesaurus (orig r6156):  ribasushi | 2009-05-07 10:28:14 +0200
     new branch for the diamond join/prefetch bug
     r6158@Thesaurus (orig r6157):  ribasushi | 2009-05-07 11:12:41 +0200
     Add failing test by gheift
     r6294@Thesaurus (orig r6293):  ribasushi | 2009-05-18 01:10:22 +0200
     Fixes for the diamond-relationship prefetch/join problem
     The core of the issue was that resolve_prefetch calculated duplicate join alias numbers separate from resolve_join
     In order to solve this, now the only join alias calculation happens in resolve_join (with prefetch being always merged as extra joins), and each join arrayref in from is labeled with the full relationship chain from me to the particular join. Then resolve_prefetch has to walk this chain and pull the necessary alias in order to generate the correct select
     r6296@Thesaurus (orig r6295):  ribasushi | 2009-05-18 14:05:31 +0200
     Add a comment about a small resolve_join omission
    
    r6314@Thesaurus (orig r6313):  frew | 2009-05-18 23:56:37 +0200
    cleanup for ribasushi
    r6316@Thesaurus (orig r6315):  frew | 2009-05-19 01:10:20 +0200
    I contributed!
    r6317@Thesaurus (orig r6316):  arcanez | 2009-05-19 05:30:18 +0200
    doc patch per amiri
    r6318@Thesaurus (orig r6317):  arcanez | 2009-05-19 05:33:47 +0200
    I think I got it right this time
    r6322@Thesaurus (orig r6321):  abraxxa | 2009-05-19 19:07:08 +0200
    added test for total_entries on a paged resultset
    
    r6326@Thesaurus (orig r6325):  ribasushi | 2009-05-19 19:34:45 +0200
     r6258@Thesaurus (orig r6257):  ribasushi | 2009-05-14 13:18:58 +0200
     Refactor count handling - use a different code path depending on the complexity of the resultset query, allows proper collapsing of count results via group_by subqueries
     
    
    r6327@Thesaurus (orig r6326):  ribasushi | 2009-05-19 19:34:53 +0200
     r6259@Thesaurus (orig r6258):  ribasushi | 2009-05-14 13:19:35 +0200
     Straightforward test changes
    
    r6328@Thesaurus (orig r6327):  ribasushi | 2009-05-19 19:35:00 +0200
     r6260@Thesaurus (orig r6259):  ribasushi | 2009-05-14 13:19:59 +0200
     Questionable test changes - need a review
    
    r6329@Thesaurus (orig r6328):  ribasushi | 2009-05-19 19:35:07 +0200
     r6269@Thesaurus (orig r6268):  ribasushi | 2009-05-15 10:43:04 +0200
     Add failing multikey rs delete (and by implication update) test
    
    r6330@Thesaurus (orig r6329):  ribasushi | 2009-05-19 19:35:14 +0200
     r6276@Thesaurus (orig r6275):  ribasushi | 2009-05-15 23:35:06 +0200
     A workable fix for the resultset multicol update/delete debacle - by default fallback to per-row deletions, with the ability to overide this behavior for various storage drivers
    
    r6331@Thesaurus (orig r6330):  ribasushi | 2009-05-19 19:35:20 +0200
     r6279@Thesaurus (orig r6278):  ribasushi | 2009-05-16 08:36:45 +0200
     Rename tests
    
    r6332@Thesaurus (orig r6331):  ribasushi | 2009-05-19 19:36:37 +0200
     r6284@Thesaurus (orig r6283):  ribasushi | 2009-05-16 10:40:44 +0200
     Test and fixed for paged grouped count
    
    r6333@Thesaurus (orig r6332):  ribasushi | 2009-05-19 19:38:00 +0200
     r6301@Thesaurus (orig r6300):  ribasushi | 2009-05-18 19:32:42 +0200
     pager optimization
    
    r6334@Thesaurus (orig r6333):  ribasushi | 2009-05-19 19:39:27 +0200
     r6321@Thesaurus (orig r6320):  ribasushi | 2009-05-19 18:11:00 +0200
     Make select always equal group by on count subqueries
     Make sure offset does not increment boundlessly on every _resolved_attrs run
    
    r6335@Thesaurus (orig r6334):  ribasushi | 2009-05-19 19:39:34 +0200
     r6323@Thesaurus (orig r6322):  ribasushi | 2009-05-19 19:26:21 +0200
     Make sure pg test can restart after half-way failures
    
    r6336@Thesaurus (orig r6335):  ribasushi | 2009-05-19 19:39:42 +0200
     r6324@Thesaurus (orig r6323):  ribasushi | 2009-05-19 19:34:19 +0200
     Hide resultsource methods that should have never been documented
    
    r6337@Thesaurus (orig r6336):  ribasushi | 2009-05-19 19:42:59 +0200
    Add files dropped by the screwed up merge - I suck
    r6338@Thesaurus (orig r6337):  ribasushi | 2009-05-19 19:43:56 +0200
    Adjust dependencies - most of them pull in bugfixes
    r6339@Thesaurus (orig r6338):  ribasushi | 2009-05-19 19:52:39 +0200
    Make podcoverage happy (deprecated, undocumented methods)
    r6340@Thesaurus (orig r6339):  ribasushi | 2009-05-19 20:46:40 +0200
    Simplify __count() - explicitly specify which attributes to copy, instead of deleting a whole bunch of them
    r6342@Thesaurus (orig r6341):  dandv | 2009-05-20 02:56:57 +0200
    Tiny POD formatting fix
    r6343@Thesaurus (orig r6342):  caelum | 2009-05-20 05:11:12 +0200
    add mysql lost password faq
    r6344@Thesaurus (orig r6343):  caelum | 2009-05-20 05:58:36 +0200
    add test for distinct result of sql function
    r6345@Thesaurus (orig r6344):  ribasushi | 2009-05-20 09:03:08 +0200
    Make joined rs counts backwards compatible - we do not collapse a result exploded by a has_many join unless it is explicitly requested by distinct => 1, OR unless we have collapse set which implies prefetch
    r6347@Thesaurus (orig r6346):  ribasushi | 2009-05-20 12:44:54 +0200
    detect/purge having bind for compeleteness
    r6348@Thesaurus (orig r6347):  ribasushi | 2009-05-20 12:51:05 +0200
    Bring back the old code from resolve_prefetch so ash's code can work
    Add tests to make sure we throw it away in 09
    r6349@Thesaurus (orig r6348):  ribasushi | 2009-05-20 14:30:40 +0200
    Add explicit grouping for rs update/delete operations if the parameters warrant it (tests coming tonight)
    r6350@Thesaurus (orig r6349):  ribasushi | 2009-05-20 15:32:27 +0200
    Make sure we always work with resolved rs attributes while counting (and test)
    r6351@Thesaurus (orig r6350):  ribasushi | 2009-05-20 15:42:48 +0200
    More explicit attribute deletion/detection - a lof of this is probably redundant, as _resoved_attrs() should flatten them away (i.e. prefetch to collapse and distinct to group_by). Nevertheless keep them there for explcetness ( delete() is cheap )
    r6352@Thesaurus (orig r6351):  arcanez | 2009-05-20 17:04:17 +0200
    update to Changes
    r6353@Thesaurus (orig r6352):  arcanez | 2009-05-20 17:06:50 +0200
    fix for sql functions in group_by
    r6354@Thesaurus (orig r6353):  frew | 2009-05-20 18:03:55 +0200
    fix test for people with spaces in path
    r6355@Thesaurus (orig r6354):  ribasushi | 2009-05-20 22:59:10 +0200
    POD fixes
    r6356@Thesaurus (orig r6355):  ribasushi | 2009-05-20 23:00:13 +0200
    Switch as_query testing to direct specification in is_same_sql_bind
    r6357@Thesaurus (orig r6356):  ribasushi | 2009-05-20 23:47:05 +0200
    on_connect_do now accepts a single string like it does an arrayref (patch by prema)
    r6358@Thesaurus (orig r6357):  ribasushi | 2009-05-21 09:45:52 +0200
    Clarify usage of _resolved_attrs by adding the explicit _resolved_attrs_copy
    Clarify code in ResultSetColumn
    r6359@Thesaurus (orig r6358):  ribasushi | 2009-05-21 09:59:50 +0200
    Duh, no point of having ordered subqueries while counting, Debolaz++
    r6367@Thesaurus (orig r6366):  caelum | 2009-05-22 04:54:12 +0200
    add some support for trying to determine storage driver without being connected
    r6368@Thesaurus (orig r6367):  caelum | 2009-05-22 06:52:47 +0200
    comment addition
    r6369@Thesaurus (orig r6368):  caelum | 2009-05-22 08:09:05 +0200
    fix limit for DB2
    r6370@Thesaurus (orig r6369):  ribasushi | 2009-05-22 10:28:58 +0200
    Extend DB2 test
    r6371@Thesaurus (orig r6370):  ribasushi | 2009-05-22 10:31:25 +0200
    Some attributes require a grouped count subquery, some - just a subquery. Differentiate properly
    r6372@Thesaurus (orig r6371):  ribasushi | 2009-05-22 10:35:36 +0200
    duh
    r6374@Thesaurus (orig r6373):  caelum | 2009-05-22 15:50:45 +0200
    ::Replicated -- check for master rebless and reapply role if necessary
    r6376@Thesaurus (orig r6375):  ash | 2009-05-22 18:54:00 +0200
    Make distinct calculate columns *after* prefetch has been resolved. Tests to come
    
    r6380@Thesaurus (orig r6379):  ribasushi | 2009-05-23 10:09:46 +0200
    Refactor rs_update_delete (too many methods for no reason)
    Make mysql use _per_row_update_delete by default as this is the only thing it understands: http://dev.mysql.com/doc/refman/5.0/en/subquery-errors.html
    r6381@Thesaurus (orig r6380):  caelum | 2009-05-23 16:44:59 +0200
    Sybase autopk, and a test, no limit support yet
    r6382@Thesaurus (orig r6381):  ribasushi | 2009-05-23 21:33:28 +0200
    Tests and test schema adjustments for resultset update/delete
    r6383@Thesaurus (orig r6382):  ribasushi | 2009-05-23 21:35:59 +0200
    Add storage component of multipk resultset update/delete for multicolumn IN capable rdbms
    Switch Pg to that (tested via DBICTEST_DSN)
    r6384@Thesaurus (orig r6383):  caelum | 2009-05-23 21:46:11 +0200
    Sybase bindvar and IC::DT support
    r6385@Thesaurus (orig r6384):  caelum | 2009-05-23 21:57:08 +0200
    hide internal Sybase classes from PAUSE
    r6387@Thesaurus (orig r6386):  caelum | 2009-05-23 22:28:42 +0200
    revert last Sybase changes
    r6390@Thesaurus (orig r6389):  ribasushi | 2009-05-23 22:59:17 +0200
    Extra test and count fixes for prefetch + distinct
    r6391@Thesaurus (orig r6390):  caelum | 2009-05-23 23:04:01 +0200
    minor clean up
    r6392@Thesaurus (orig r6391):  ribasushi | 2009-05-23 23:37:19 +0200
    Make sure MultiColumnIn quotes column names while munging literal sql
    Tested by running t/resultset/update_delete against pg with quote char " and name sep .
    r6393@Thesaurus (orig r6392):  ribasushi | 2009-05-24 00:47:40 +0200
    update changes
    r6394@Thesaurus (orig r6393):  ribasushi | 2009-05-24 00:50:08 +0200
    Extend distinct deprecation tests and clarify warnings
    r6396@Thesaurus (orig r6395):  ribasushi | 2009-05-24 10:47:03 +0200
    MSSQL through ODBC does not like unfinished statements - make sure we finish the scope identity retrieval
    (This worked before because of the automatic retry on exception, essentially running any select after insert twice)
    r6397@Thesaurus (orig r6396):  ribasushi | 2009-05-24 11:42:01 +0200
    Make sure we do not clobber search attributes when using subqueries
    r6398@Thesaurus (orig r6397):  ribasushi | 2009-05-24 11:43:32 +0200
    SUPER is so last century
    r6399@Thesaurus (orig r6398):  ribasushi | 2009-05-24 14:12:39 +0200
    Deprecate ::DBI::Sybase::MSSQL
    r6400@Thesaurus (orig r6399):  ribasushi | 2009-05-24 15:00:50 +0200
    eol adjustments
    r6401@Thesaurus (orig r6400):  ribasushi | 2009-05-24 15:07:45 +0200
    Switch trunk to native eol-style
    r6402@Thesaurus (orig r6401):  ribasushi | 2009-05-24 15:35:07 +0200
    Failing test about warnings triggered in SQLA::Limit when using a subquery
    r6406@Thesaurus (orig r6405):  ribasushi | 2009-05-24 22:53:15 +0200
    Switch around inheritance of MSSQL drivers, remove some duplicate code
    r6407@Thesaurus (orig r6406):  caelum | 2009-05-25 01:49:17 +0200
    fix double connect for ODBC/MSSQL
    r6408@Thesaurus (orig r6407):  caelum | 2009-05-25 01:53:12 +0200
    added test to make sure only one connection to ODBC/MSSQL is made
    r6411@Thesaurus (orig r6410):  ribasushi | 2009-05-25 08:48:38 +0200
    Factor out the order_by sqlahacks resolver
    r6412@Thesaurus (orig r6411):  ribasushi | 2009-05-25 09:42:45 +0200
    Move the DB2 Limit syntax setting into the storage class
    r6413@Thesaurus (orig r6412):  ribasushi | 2009-05-25 09:47:01 +0200
    Forgotten podcoverage override
    r6414@Thesaurus (orig r6413):  ribasushi | 2009-05-25 11:11:01 +0200
    Define how Top limit emulation should behave
    r6419@Thesaurus (orig r6418):  ribasushi | 2009-05-25 16:53:13 +0200
    Actually don't need this anymore
    r6421@Thesaurus (orig r6420):  tomboh | 2009-05-25 18:33:46 +0200
    Small documentation improvement:  link to a documented method.
    r6423@Thesaurus (orig r6422):  ribasushi | 2009-05-25 19:42:02 +0200
    This method does not exist anymore
    r6427@Thesaurus (orig r6426):  ribasushi | 2009-05-26 08:15:34 +0200
    TODOified test for RT#40701
    r6429@Thesaurus (orig r6428):  ash | 2009-05-26 16:17:44 +0200
    Fix 'timestamp with time zone' columns for IC::DT inflation
    
    r6431@Thesaurus (orig r6430):  ribasushi | 2009-05-26 16:29:55 +0200
     r6415@Thesaurus (orig r6414):  ribasushi | 2009-05-25 11:18:05 +0200
     Yet another branch to attempt a top fix
     r6416@Thesaurus (orig r6415):  ribasushi | 2009-05-25 11:24:32 +0200
     The Top limit emulation bundled with SQLA::Limit assumes that the limited resultset will be _always_ sorted. In order to fix this, I reimplemented _Top in SQLAHacks with a slight modification. Now the original order_by is passed to the outside of the nested select block, while order_up/down are calculated either based on the original order_by, or if one is not present an order by all PKs is attempted.
     Since I do not have access to $rsrc in SQLA, I pass the list of PKs as an extra group_by hash entry. This appears to be rather safe, and besides we already pollute order_by with group_by and having (which seems to work rather well).
     The only thing I am unsure about is the need for _gen_virtual_order(). Initially I was going to generate the pk list, only if we use the Top limit. Then it turned out there is no limit dialect before we connect, so I commented it out. Now all it does is check for a limit condition and returns the PK list. Is this necessary at all?
     r6417@Thesaurus (orig r6416):  ribasushi | 2009-05-25 16:08:40 +0200
     Shoot another Top problem, move test from top_limit_tweaks branch and delete
     r6420@Thesaurus (orig r6419):  ribasushi | 2009-05-25 17:45:33 +0200
     Too much logic for no benefit - always populate _virtual_order_by
    
    r6433@Thesaurus (orig r6432):  ribasushi | 2009-05-26 16:36:55 +0200
    Fix test skip message
    r6434@Thesaurus (orig r6433):  ribasushi | 2009-05-26 17:20:57 +0200
    Minor fixes of the return value of rs->update/delete
    r6435@Thesaurus (orig r6434):  ribasushi | 2009-05-26 20:49:49 +0200
    fix comments
    r6436@Thesaurus (orig r6435):  ribasushi | 2009-05-26 21:28:49 +0200
    Attempt to reproduce reported mysql error (failed) - fixed another bug in ResultSetColumn along the way
    r6437@Thesaurus (orig r6436):  ribasushi | 2009-05-26 22:02:29 +0200
    Release 0.08103
    r6447@Thesaurus (orig r6446):  ribasushi | 2009-05-28 10:20:57 +0200
    Commit rather useless but already written mysql test extension
    r6448@Thesaurus (orig r6447):  ribasushi | 2009-05-28 13:02:22 +0200
    Fix multiprefetch warning - we can now count properly
    r6457@Thesaurus (orig r6456):  ribasushi | 2009-05-30 07:40:24 +0200
    Patch + test for more informative exceptions on load_namespace a non-rs class
    r6458@Thesaurus (orig r6457):  ribasushi | 2009-05-30 09:34:20 +0200
    Add better error reporting on bulk_insert (ash++)
    r6470@Thesaurus (orig r6469):  ribasushi | 2009-05-30 19:46:09 +0200
    populate() fix and Changes
    r6471@Thesaurus (orig r6470):  ribasushi | 2009-05-31 09:37:37 +0200
    M::I 0.89 finally resolves all provlems with auto_install
    r6472@Thesaurus (orig r6471):  ribasushi | 2009-05-31 09:42:35 +0200
    Throw away the makefile SQLite test - it served its purpose
    r6473@Thesaurus (orig r6472):  ribasushi | 2009-05-31 11:24:15 +0200
    There is a saner way to write out resources
    r6474@Thesaurus (orig r6473):  ribasushi | 2009-05-31 16:07:18 +0200
    Last set of Makefile.PL optimizations
    r6475@Thesaurus (orig r6474):  ribasushi | 2009-06-01 12:24:41 +0200
    deploy-related pod fixes
    r6477@Thesaurus (orig r6476):  ribasushi | 2009-06-01 16:40:22 +0200
     r6462@Thesaurus (orig r6461):  mo | 2009-05-30 11:06:54 +0200
     order_by tests
    
    r6478@Thesaurus (orig r6477):  ribasushi | 2009-06-01 16:40:30 +0200
     r6463@Thesaurus (orig r6462):  ribasushi | 2009-05-30 16:54:37 +0200
     TODOify some of the order with bind tests
    
    r6479@Thesaurus (orig r6478):  ribasushi | 2009-06-01 16:40:41 +0200
     r6464@Thesaurus (orig r6463):  ribasushi | 2009-05-30 16:55:37 +0200
     Restructure bind tests
    
    r6480@Thesaurus (orig r6479):  ribasushi | 2009-06-01 16:40:50 +0200
     r6465@Thesaurus (orig r6464):  ribasushi | 2009-05-30 17:15:57 +0200
     Greatly simplify _order_by override to fallback on new SQLA
    
    r6481@Thesaurus (orig r6480):  ribasushi | 2009-06-01 16:40:56 +0200
     r6467@Thesaurus (orig r6466):  ribasushi | 2009-05-30 19:13:23 +0200
     Evil hack to make Carp::Clan work throughout SQLA as well
    
    r6482@Thesaurus (orig r6481):  ribasushi | 2009-06-01 16:41:42 +0200
     r6468@Thesaurus (orig r6467):  ribasushi | 2009-05-30 19:17:02 +0200
     Add changes
    
    r6483@Thesaurus (orig r6482):  ribasushi | 2009-06-01 16:41:49 +0200
    
    r6485@Thesaurus (orig r6484):  ribasushi | 2009-06-01 16:49:09 +0200
    Fix fallout from another botched merge (I suck, part 2)
    r6487@Thesaurus (orig r6486):  ribasushi | 2009-06-03 10:14:51 +0200
    Require a recent version of Date::Simple during CDBI tests
    r6496@Thesaurus (orig r6495):  ribasushi | 2009-06-04 09:35:25 +0200
    Move relationship tests around
    r6499@Thesaurus (orig r6498):  ribasushi | 2009-06-04 11:52:18 +0200
    Delegate actual counting to the storage class
    r6502@Thesaurus (orig r6501):  ribasushi | 2009-06-04 15:38:21 +0200
    Really delegate counting to the storage class - now we have either a grouped count or a regular one
    r6509@Thesaurus (orig r6508):  ribasushi | 2009-06-04 22:54:28 +0200
    having without group_by is useless - thus do not supply a group_by on count, let things die
    r6511@Thesaurus (orig r6510):  ribasushi | 2009-06-04 23:51:13 +0200
    It seems that this localisation can lead to problems
    Unfortunately no test case as the trigger was buried deep within Reaction, I coulnd't figure out an isolation
    r6519@Thesaurus (orig r6518):  ribasushi | 2009-06-05 16:28:58 +0200
    This was dumb - of course I want a copy
    Otherwise the next serach_related will work with a contaminated seen_joins
    r6521@Thesaurus (orig r6520):  ribasushi | 2009-06-05 19:25:56 +0200
    Factor out the grouped count subquery SELECTor so specific storage classes can override it
    Trim attribute deletion lists - we work with _resolved_attrs - +X are already folded into X and the like
    r6529@Thesaurus (orig r6528):  ribasushi | 2009-06-06 10:45:10 +0200
    rename MC test dir
    r6530@Thesaurus (orig r6529):  ribasushi | 2009-06-06 10:47:47 +0200
    TODOified reentrancy counter - this shall be used to optimize MC some day
    r6531@Thesaurus (orig r6530):  ribasushi | 2009-06-06 10:50:56 +0200
    Port another forgotten MC fix
    r6538@Thesaurus (orig r6537):  ribasushi | 2009-06-07 23:07:55 +0200
    Fix for mysql subquery problem
    r6539@Thesaurus (orig r6538):  ribasushi | 2009-06-07 23:36:43 +0200
    Make empty/default inserts use standard SQL
    r6540@Thesaurus (orig r6539):  ribasushi | 2009-06-08 00:59:21 +0200
    Add mysql empty insert SQL override
    Make SQLAHacks parts loadable at runtime via ensure_class_loaded
    r6541@Thesaurus (orig r6540):  ribasushi | 2009-06-08 01:03:04 +0200
    Make podcoverage happy
    r6542@Thesaurus (orig r6541):  ribasushi | 2009-06-08 01:24:06 +0200
    Fix find_or_new/create to stop returning random rows when default value insert is requested
    r6543@Thesaurus (orig r6542):  ribasushi | 2009-06-08 11:36:56 +0200
    Simply order_by/_virtual_order_by handling
    r6553@Thesaurus (orig r6552):  ribasushi | 2009-06-08 23:56:41 +0200
    duh
    r6557@Thesaurus (orig r6556):  ash | 2009-06-09 12:20:34 +0200
    Addjust bug to show problem with rows => 1 + child rel
    
    r6558@Thesaurus (orig r6557):  ribasushi | 2009-06-09 13:12:46 +0200
    Require a recent bugfixed Devel::Cycle
    r6560@Thesaurus (orig r6559):  ash | 2009-06-09 15:07:30 +0200
    Make IC::DT extra warning state the column name too
    
    r6575@Thesaurus (orig r6574):  ribasushi | 2009-06-10 00:19:48 +0200
    AuthorCheck fixes
    r6579@Thesaurus (orig r6578):  ribasushi | 2009-06-10 00:52:17 +0200
     r6522@Thesaurus (orig r6521):  ribasushi | 2009-06-05 19:27:55 +0200
     New branch to try resultsource related stuff
     r6545@Thesaurus (orig r6544):  ribasushi | 2009-06-08 13:00:54 +0200
     First stab at adding resultsources to each join in select - works won-der-ful-ly
     r6546@Thesaurus (orig r6545):  ribasushi | 2009-06-08 13:14:08 +0200
     Commit failing test and thoughts on search arg deflation
     r6576@Thesaurus (orig r6575):  ribasushi | 2009-06-10 00:31:55 +0200
     Todoify DT in search deflation test until after 0.09
     r6577@Thesaurus (orig r6576):  ribasushi | 2009-06-10 00:48:07 +0200
     Factor out the $ident resolver
    
    r6581@Thesaurus (orig r6580):  ribasushi | 2009-06-10 01:21:50 +0200
    Move as_query out of the cursor
    r6582@Thesaurus (orig r6581):  ribasushi | 2009-06-10 01:27:19 +0200
    Think before commit
    r6583@Thesaurus (orig r6582):  ribasushi | 2009-06-10 09:37:19 +0200
    Clarify and disable rows/prefetch test - fix is easy, but architecturally unsound - need more time
    r6591@Thesaurus (orig r6590):  ribasushi | 2009-06-10 13:33:37 +0200
     r6544@Thesaurus (orig r6543):  ribasushi | 2009-06-08 11:44:59 +0200
     Attempt to figure out why do we repeat joins on complex search_related
     r6586@Thesaurus (orig r6585):  ribasushi | 2009-06-10 11:22:05 +0200
     Move the rs preservation test to a more suitable place
     r6589@Thesaurus (orig r6588):  ribasushi | 2009-06-10 13:15:48 +0200
     Finally commit trully failing test
     r6590@Thesaurus (orig r6589):  ribasushi | 2009-06-10 13:33:14 +0200
     Duh, this was a pretty simple bug
    
    r6593@Thesaurus (orig r6592):  ribasushi | 2009-06-10 13:43:31 +0200
    What was I thinking - resultsource does not have an ->alias
    r6598@Thesaurus (orig r6597):  ribasushi | 2009-06-10 14:48:39 +0200
    Adjust changelog
    r6601@Thesaurus (orig r6600):  ribasushi | 2009-06-10 15:50:43 +0200
    Release 0.08104
    ribasushi committed Jun 10, 2009
  5. @ribasushi

    Release 0.08104

    ribasushi committed Jun 10, 2009
  6. @ribasushi

    Adjust changelog

    ribasushi committed Jun 10, 2009
  7. @ribasushi
  8. @ribasushi

    Merge 'mystery_join' into 'trunk'

    r6544@Thesaurus (orig r6543):  ribasushi | 2009-06-08 11:44:59 +0200
    Attempt to figure out why do we repeat joins on complex search_related
    r6586@Thesaurus (orig r6585):  ribasushi | 2009-06-10 11:22:05 +0200
    Move the rs preservation test to a more suitable place
    r6589@Thesaurus (orig r6588):  ribasushi | 2009-06-10 13:15:48 +0200
    Finally commit trully failing test
    r6590@Thesaurus (orig r6589):  ribasushi | 2009-06-10 13:33:14 +0200
    Duh, this was a pretty simple bug
    ribasushi committed Jun 10, 2009
  9. @ribasushi
  10. @ribasushi
  11. @ribasushi

    Merge 'trunk' into 'mystery_join'

    r6553@Thesaurus (orig r6552):  ribasushi | 2009-06-08 23:56:41 +0200
    duh
    r6557@Thesaurus (orig r6556):  ash | 2009-06-09 12:20:34 +0200
    Addjust bug to show problem with rows => 1 + child rel
    
    r6558@Thesaurus (orig r6557):  ribasushi | 2009-06-09 13:12:46 +0200
    Require a recent bugfixed Devel::Cycle
    r6560@Thesaurus (orig r6559):  ash | 2009-06-09 15:07:30 +0200
    Make IC::DT extra warning state the column name too
    
    r6575@Thesaurus (orig r6574):  ribasushi | 2009-06-10 00:19:48 +0200
    AuthorCheck fixes
    r6579@Thesaurus (orig r6578):  ribasushi | 2009-06-10 00:52:17 +0200
     r6522@Thesaurus (orig r6521):  ribasushi | 2009-06-05 19:27:55 +0200
     New branch to try resultsource related stuff
     r6545@Thesaurus (orig r6544):  ribasushi | 2009-06-08 13:00:54 +0200
     First stab at adding resultsources to each join in select - works won-der-ful-ly
     r6546@Thesaurus (orig r6545):  ribasushi | 2009-06-08 13:14:08 +0200
     Commit failing test and thoughts on search arg deflation
     r6576@Thesaurus (orig r6575):  ribasushi | 2009-06-10 00:31:55 +0200
     Todoify DT in search deflation test until after 0.09
     r6577@Thesaurus (orig r6576):  ribasushi | 2009-06-10 00:48:07 +0200
     Factor out the $ident resolver
    
    r6581@Thesaurus (orig r6580):  ribasushi | 2009-06-10 01:21:50 +0200
    Move as_query out of the cursor
    r6582@Thesaurus (orig r6581):  ribasushi | 2009-06-10 01:27:19 +0200
    Think before commit
    r6583@Thesaurus (orig r6582):  ribasushi | 2009-06-10 09:37:19 +0200
    Clarify and disable rows/prefetch test - fix is easy, but architecturally unsound - need more time
    ribasushi committed Jun 10, 2009
  12. @ribasushi
  13. @ribasushi

    Clarify and disable rows/prefetch test - fix is easy, but architectur…

    …ally unsound - need more time
    ribasushi committed Jun 10, 2009
Commits on Jun 9, 2009
  1. @ribasushi

    Think before commit

    ribasushi committed Jun 9, 2009
  2. @ribasushi
  3. @ribasushi
  4. @ribasushi
  5. @ribasushi
  6. @ribasushi

    AuthorCheck fixes

    ribasushi committed Jun 9, 2009
  7. @ashb
  8. @ribasushi