16 years agofixed failing test for sqlt
John Napiorkowski [Fri, 30 May 2008 14:32:24 +0000]
fixed failing test for sqlt

16 years agoMerge 'trunk' into 'replication_dedux'
John Napiorkowski [Fri, 30 May 2008 13:55:35 +0000]
Merge 'trunk' into 'replication_dedux'

r13471@dev (orig r4428):  matthewt | 2008-05-30 07:36:19 -0500
fix SQL::Translator parser to add $table_idx_ on the front of index names to prevent clashes
r13472@dev (orig r4429):  castaway | 2008-05-30 08:20:26 -0500
Started doc standards doc in Manual::Reading.
Fixed up belongs_to example to put more emphasis on the "column" part of "foreign_key_column" and corrected first paragraph.

r13473@dev (orig r4430):  ash | 2008-05-30 08:29:32 -0500
Make ingore_{constraint,index}_names default
r13474@dev (orig r4431):  ash | 2008-05-30 08:34:39 -0500
Update docs (and code changes that were missed from last commit, oops):
Set ignore_contraint_names to default on

16 years agoUpdate docs (and code changes that were missed from last commit, oops):
Ash Berlin [Fri, 30 May 2008 13:34:39 +0000]
Update docs (and code changes that were missed from last commit, oops):
Set ignore_contraint_names to default on

16 years agoMake ingore_{constraint,index}_names default
Ash Berlin [Fri, 30 May 2008 13:29:32 +0000]
Make ingore_{constraint,index}_names default

16 years agoStarted doc standards doc in Manual::Reading.
Jess Robinson [Fri, 30 May 2008 13:20:26 +0000]
Started doc standards doc in Manual::Reading.
Fixed up belongs_to example to put more emphasis on the "column" part of "foreign_key_column" and corrected first paragraph.

16 years agofix SQL::Translator parser to add $table_idx_ on the front of index names to prevent...
Matt S Trout [Fri, 30 May 2008 12:36:19 +0000]
fix SQL::Translator parser to add $table_idx_ on the front of index names to prevent clashes

16 years agofixed failing test in podcoverage, fixed regression in replication test that caused...
John Napiorkowski [Fri, 30 May 2008 00:18:39 +0000]
fixed failing test in podcoverage, fixed regression in replication test that caused the default fake sqlite replication to fail.  not so important since sqlite doesnt replicate but we do not like to see failing tests.

16 years agochanged Storage->reload_row to do less, reverted some behavior to PK->discard_changes...
John Napiorkowski [Thu, 29 May 2008 23:03:15 +0000]
changed Storage->reload_row to do less, reverted some behavior to PK->discard_changes.  Did this to solve some capatibility issues with partitioning.  updated docs to reflect this.

16 years agodiscard changes now is forced to use master for replication. changed discard_changes...
John Napiorkowski [Thu, 29 May 2008 18:30:39 +0000]
discard changes now is forced to use master for replication.  changed discard_changes guts to point to new method called reload_row in storage.  fixed problem with replicated transactionws not returning the right thing.  added tests to all the above

16 years agocreated storage method to execute a coderef using master storage only, changed tnx_do...
John Napiorkowski [Wed, 28 May 2008 19:15:34 +0000]
created storage method to execute a coderef using master storage only, changed tnx_do to only use the master, wrote tests for both the above, wrote docs for both the above

16 years agoMerge 'trunk' into 'replication_dedux'
John Napiorkowski [Wed, 28 May 2008 17:29:00 +0000]
Merge 'trunk' into 'replication_dedux'

r13166@dev (orig r4392):  castaway | 2008-05-19 16:57:25 -0500
Added doc for "for => update" attribute, thanks StuartL.

r13167@dev (orig r4393):  castaway | 2008-05-19 16:59:16 -0500
Oops, =cut after the pod not in the middle

r13272@dev (orig r4419):  ash | 2008-05-27 05:41:44 -0500
Update FAQ to mention behavour of scalar refs w.r.t. update
r13273@dev (orig r4420):  ash | 2008-05-27 07:49:09 -0500
Add set_cache example to cookbook
r13274@dev (orig r4421):  ash | 2008-05-27 07:50:10 -0500
Fix var name typo in cookbook
r13376@dev (orig r4422):  matthewt | 2008-05-27 08:31:58 -0500
version bump, deprecated ResultSetManager

16 years agoversion bump, deprecated ResultSetManager
Matt S Trout [Tue, 27 May 2008 13:31:58 +0000]
version bump, deprecated ResultSetManager

16 years agoFix var name typo in cookbook
Ash Berlin [Tue, 27 May 2008 12:50:10 +0000]
Fix var name typo in cookbook

16 years agoAdd set_cache example to cookbook
Ash Berlin [Tue, 27 May 2008 12:49:09 +0000]
Add set_cache example to cookbook

16 years agoUpdate FAQ to mention behavour of scalar refs w.r.t. update
Ash Berlin [Tue, 27 May 2008 10:41:44 +0000]
Update FAQ to mention behavour of scalar refs w.r.t. update

16 years agoOops, =cut after the pod not in the middle
Jess Robinson [Mon, 19 May 2008 21:59:16 +0000]
Oops, =cut after the pod not in the middle

16 years agoAdded doc for "for => update" attribute, thanks StuartL.
Jess Robinson [Mon, 19 May 2008 21:57:25 +0000]
Added doc for "for => update" attribute, thanks StuartL.

16 years agomake sure that the Pool validates the replicants on the first query
John Napiorkowski [Fri, 16 May 2008 17:31:16 +0000]
make sure that the Pool validates the replicants on the first query

16 years agodocumentation updates
John Napiorkowski [Wed, 14 May 2008 17:57:57 +0000]
documentation updates

16 years agochanged SQLT::Parser::DBIC so that in FK constraints, both the field order for the...
John Napiorkowski [Wed, 14 May 2008 16:05:22 +0000]
changed SQLT::Parser::DBIC so that in FK constraints, both the field order for the source and target tables are ensured to follow the same ordering rule

16 years agoMerge 'trunk' into 'replication_dedux'
John Napiorkowski [Wed, 14 May 2008 13:40:07 +0000]
Merge 'trunk' into 'replication_dedux'

r12028@dev (orig r4378):  captainL | 2008-05-11 13:43:58 -0500
added test for find or create related functionality in nested inserts
r12029@dev (orig r4379):  captainL | 2008-05-11 14:03:27 -0500
sanified new multi_create test
r12030@dev (orig r4381):  ribasushi | 2008-05-12 06:09:20 -0500
Add failing tests for missing safeguards for multilevel prefetch (don't know the internals well enough to fix the issue itself)

16 years agoAdd failing tests for missing safeguards for multilevel prefetch (don't know the...
Peter Rabbitson [Mon, 12 May 2008 11:09:20 +0000]
Add failing tests for missing safeguards for multilevel prefetch (don't know the internals well enough to fix the issue itself)

16 years agosanified new multi_create test
Luke Saunders [Sun, 11 May 2008 19:03:27 +0000]
sanified new multi_create test

16 years agoadded test for find or create related functionality in nested inserts
Luke Saunders [Sun, 11 May 2008 18:43:58 +0000]
added test for find or create related functionality in nested inserts

16 years agoremoved bad tabbing
John Napiorkowski [Fri, 9 May 2008 19:34:17 +0000]
removed bad tabbing

16 years agodoh, replaced ability to configure pool args via the storage args
John Napiorkowski [Fri, 9 May 2008 18:32:58 +0000]
doh, replaced ability to configure pool args via the storage args

16 years agochanged the way args are passed to a storage type that needs args so they can be...
John Napiorkowski [Fri, 9 May 2008 17:00:46 +0000]
changed the way args are passed to a storage type that needs args so they can be in the form of a hash or array ref.  This plays nicer with Config::General for loading

16 years agochanged the balancer to a role, created a new class to define the default balancer...
John Napiorkowski [Fri, 9 May 2008 15:53:45 +0000]
changed the balancer to a role, created a new class to define the default balancer behavior, finished the autovalidate code and added tests for all the above

16 years agoMerge 'trunk' into 'replication_dedux'
John Napiorkowski [Fri, 9 May 2008 13:39:51 +0000]
Merge 'trunk' into 'replication_dedux'

r8571@dev (orig r4363):  bricas | 2008-05-08 10:44:24 -0500
add Test::Exception to deps (RT #34256). move testing modules to test_requires().
r8572@dev (orig r4364):  bricas | 2008-05-08 15:08:18 -0500
add RT ticket number to changes
r8579@dev (orig r4371):  bricas | 2008-05-08 21:55:30 -0500
pod fix (RT #32988)

16 years agopod fix (RT #32988)
Brian Cassidy [Fri, 9 May 2008 02:55:30 +0000]
pod fix (RT #32988)

16 years agogood start on the validation of replicants and a system to automatically validate...
John Napiorkowski [Fri, 9 May 2008 01:40:03 +0000]
good start on the validation of replicants and a system to automatically validate them (although that might be a better role than inside a class, for someday!)

16 years agofixed up the relicant status tests to exclude them if the database is not a real...
John Napiorkowski [Fri, 9 May 2008 00:31:58 +0000]
fixed up the relicant status tests to exclude them if the database is not a real replicating setup, removed some debug code, got the lag_behind_master and is_replicating methods working properly.

16 years agofixed up the relicant status tests to exclude them if the database is not a real...
John Napiorkowski [Fri, 9 May 2008 00:06:42 +0000]
fixed up the relicant status tests to exclude them if the database is not a real replicating setup, removed some debug code, got the lag_behind_master and is_replicating methods working properly.

16 years agoconverted replicant to a role so that we can apply it after ensure_connected properly...
John Napiorkowski [Thu, 8 May 2008 23:34:55 +0000]
converted replicant to a role so that we can apply it after ensure_connected properly reblesses the storage into the correct driver specific storage

16 years agomore cleanup of the test suite so that we can run it against other databases. fixed...
John Napiorkowski [Thu, 8 May 2008 21:43:16 +0000]
more cleanup of the test suite so that we can run it against other databases.  fixed the problem with tests using self-referential constrainsts for dbs that have trouble handling that

16 years agolots of updates to make the test suite work with databases other than sqlite
John Napiorkowski [Thu, 8 May 2008 20:26:01 +0000]
lots of updates to make the test suite work with databases other than sqlite

16 years agoadd RT ticket number to changes
Brian Cassidy [Thu, 8 May 2008 20:08:18 +0000]
add RT ticket number to changes

16 years agoadd Test::Exception to deps (RT #34256). move testing modules to test_requires().
Brian Cassidy [Thu, 8 May 2008 15:44:24 +0000]
add Test::Exception to deps (RT #34256). move testing modules to test_requires().

16 years agoMerge 'trunk' into 'replication_dedux'
John Napiorkowski [Thu, 8 May 2008 14:00:50 +0000]
Merge 'trunk' into 'replication_dedux'

r8568@dev (orig r4360):  nigel | 2008-05-08 08:21:33 -0500
cascade_copy docs - takes from http://osdir.com/ml/lang.perl.modules.dbix-class/2007-03/msg00106.html
r8569@dev (orig r4361):  nigel | 2008-05-08 08:22:44 -0500
cascade_copy docs (format fixed)

16 years agocascade_copy docs (format fixed)
Nigel Metheringham [Thu, 8 May 2008 13:22:44 +0000]
cascade_copy docs (format fixed)

16 years agocascade_copy docs - takes from http://osdir.com/ml/lang.perl.modules.dbix-class/2007...
Nigel Metheringham [Thu, 8 May 2008 13:21:33 +0000]
cascade_copy docs - takes from osdir.com/ml/lang.perl.modules.dbix-class/2007-03/msg00106.html

16 years agochanged the way args are passed to a storage, should make it easier to use existing...
John Napiorkowski [Wed, 7 May 2008 22:40:30 +0000]
changed the way args are passed to a storage, should make it easier to use existing code using this, added the master as a fallback to the the replicants, lots of small documentation updates and test improvements.  all tests passing

16 years agoMerge 'trunk' into 'replication_dedux'
John Napiorkowski [Wed, 7 May 2008 21:16:22 +0000]
Merge 'trunk' into 'replication_dedux'

r8564@dev (orig r4356):  ribasushi | 2008-05-07 10:25:41 -0500
Correct order of BEGIN/use so that skip_all does not produce failures
r8565@dev (orig r4357):  ribasushi | 2008-05-07 11:37:56 -0500
Various fixes to make testing with DBICTEST_SQLT_DEPLOY=1 possible

16 years agoVarious fixes to make testing with DBICTEST_SQLT_DEPLOY=1 possible
Peter Rabbitson [Wed, 7 May 2008 16:37:56 +0000]
Various fixes to make testing with DBICTEST_SQLT_DEPLOY=1 possible

16 years agoCorrect order of BEGIN/use so that skip_all does not produce failures
Peter Rabbitson [Wed, 7 May 2008 15:25:41 +0000]
Correct order of BEGIN/use so that skip_all does not produce failures

16 years agoMerge 'trunk' into 'replication_dedux'
Peter Rabbitson [Wed, 7 May 2008 13:33:50 +0000]
Merge 'trunk' into 'replication_dedux'

r8495@dev (orig r4354):  ribasushi | 2008-05-07 08:33:50 -0500
Fix a forgotten CDBICompat component relocation

16 years agoFix a forgotten CDBICompat component relocation
Peter Rabbitson [Wed, 7 May 2008 13:33:50 +0000]
Fix a forgotten CDBICompat component relocation

16 years agodocumented methods for detecting replicant reliability, created stub methods
John Napiorkowski [Wed, 7 May 2008 00:43:52 +0000]
documented methods for detecting replicant reliability, created stub methods

16 years agorefactored the duties of the different balancer classes, added tests and docs
John Napiorkowski [Wed, 7 May 2008 00:23:09 +0000]
refactored the duties of the different balancer classes, added tests and docs

16 years agoMerge 'trunk' into 'replication_dedux'
John Napiorkowski [Tue, 6 May 2008 18:55:17 +0000]
Merge 'trunk' into 'replication_dedux'

16 years agoupdated Changes to document Replicated Storage type changes
John Napiorkowski [Tue, 6 May 2008 18:55:17 +0000]
updated Changes to document Replicated Storage type changes

16 years agochanged the warning on DBIC::Storage::DBI->select_single so that it wont call fetch_...
John Napiorkowski [Tue, 6 May 2008 18:47:00 +0000]
changed the warning on DBIC::Storage::DBI->select_single so that it wont call fetch_* on an empty sth, updated tests for the above and added a todo test for the wrong count problem

16 years agoNote that single expects a single row in the ResultSet docs
Daniel Westermann-Clark [Mon, 5 May 2008 22:21:36 +0000]
Note that single expects a single row in the ResultSet docs

16 years agoremoved code that tossed an error in select_single when more than a single row is...
John Napiorkowski [Mon, 5 May 2008 21:41:48 +0000]
removed code that tossed an error in select_single when more than a single row is returned and updated the tests to TODO the bad count issue

16 years agoMerge 'trunk' into 'replication_dedux'
John Napiorkowski [Mon, 5 May 2008 20:50:11 +0000]
Merge 'trunk' into 'replication_dedux'

r8434@dev (orig r4321):  matthewt | 2008-05-05 12:55:23 -0500
reverse r4290 since we -do- -not- currently want these namespaces indexed
r8435@dev (orig r4326):  jnapiorkowski | 2008-05-05 13:33:57 -0500
removed query count stuff from trunk so we can play with this on a branch instead
r8436@dev (orig r4330):  matthewt | 2008-05-05 14:59:31 -0500
this was meant to be deleted with the rest of the Moose stuff
r8437@dev (orig r4331):  matthewt | 2008-05-05 15:00:18 -0500
make override follow Replication -> Replicated rename
r8438@dev (orig r4332):  matthewt | 2008-05-05 15:00:49 -0500
restore debolaz's cleanup patch since it's not what I thought it was and probably the right thing to do; mea culpa
r8439@dev (orig r4333):  matthewt | 2008-05-05 15:22:17 -0500
missed another package line
r8440@dev (orig r4334):  matthewt | 2008-05-05 15:32:58 -0500
improvements to the META.yml magic
r8442@dev (orig r4336):  matthewt | 2008-05-05 15:46:21 -0500
no author notes in Changes if you're already in CONTRIBUTORS please, once you have that and a commit bit you're just another poor bastard stuck maintaining an ORM, just like I am

16 years agono author notes in Changes if you're already in CONTRIBUTORS please, once you have...
Matt S Trout [Mon, 5 May 2008 20:46:21 +0000]
no author notes in Changes if you're already in CONTRIBUTORS please, once you have that and a commit bit you're just another poor bastard stuck maintaining an ORM, just like I am

16 years agoupdated mysql test to reflect the fetch without execute error
John Napiorkowski [Mon, 5 May 2008 20:42:52 +0000]
updated mysql test to reflect the fetch without execute error

16 years agoimprovements to the META.yml magic
Matt S Trout [Mon, 5 May 2008 20:32:58 +0000]
improvements to the META.yml magic

16 years agomissed another package line
Matt S Trout [Mon, 5 May 2008 20:22:17 +0000]
missed another package line

16 years agorestore debolaz's cleanup patch since it's not what I thought it was and probably...
Matt S Trout [Mon, 5 May 2008 20:00:49 +0000]
restore debolaz's cleanup patch since it's not what I thought it was and probably the right thing to do; mea culpa

16 years agomake override follow Replication -> Replicated rename
Matt S Trout [Mon, 5 May 2008 20:00:18 +0000]
make override follow Replication -> Replicated rename

16 years agothis was meant to be deleted with the rest of the Moose stuff
Matt S Trout [Mon, 5 May 2008 19:59:31 +0000]
this was meant to be deleted with the rest of the Moose stuff

16 years agoremoved query count stuff from trunk so we can play with this on a branch instead
John Napiorkowski [Mon, 5 May 2008 18:33:57 +0000]
removed query count stuff from trunk so we can play with this on a branch instead

16 years agoreverse r4290 since we -do- -not- currently want these namespaces indexed
Matt S Trout [Mon, 5 May 2008 17:55:23 +0000]
reverse r4290 since we -do- -not- currently want these namespaces indexed

16 years agochanged replication test to support custom database connect info, added a little...
John Napiorkowski [Thu, 1 May 2008 15:56:10 +0000]
changed replication test to support custom database connect info, added a little code to DBICTest to support this

16 years agoadded test to check when nothing is found
John Napiorkowski [Wed, 30 Apr 2008 23:15:28 +0000]
added test to check when nothing is found

16 years agocleanup of some docs, got the default shuffling balancer to work properly. Don't...
John Napiorkowski [Wed, 30 Apr 2008 22:50:09 +0000]
cleanup of some docs, got the default shuffling balancer to work properly.  Don't fall in love with this behavior, since I'm probably going to change the default to balancer to just return the first replicant in the list, since this is optimised for the common case of a single replicant

16 years agoadded some advice to debugging replicants so that we can see a replicant dsn, got...
John Napiorkowski [Wed, 30 Apr 2008 22:30:47 +0000]
added some advice to debugging replicants so that we can see a replicant dsn, got this balancing between one master and slave, seems to run well

16 years agoadded some advice to debugging replicants so that we can see a replicant dsn, got...
John Napiorkowski [Wed, 30 Apr 2008 22:16:55 +0000]
added some advice to debugging replicants so that we can see a replicant dsn, got this balancing between one master and slave, seems to run well

16 years agogot first pass on the replication and balancer, passing all of the old test suite...
John Napiorkowski [Wed, 30 Apr 2008 20:26:26 +0000]
got first pass on the replication and balancer, passing all of the old test suite (which is not much, but it is a milestone of some sort)

16 years agonew config option to DBICTest to let you set an alternative storage type, start on...
John Napiorkowski [Wed, 30 Apr 2008 15:51:48 +0000]
new config option to DBICTest to let you set an alternative storage type, start on creating a DBIC based load balancer

16 years agominor formatting updates and typos fixes to the schema role tests
John Napiorkowski [Mon, 28 Apr 2008 18:36:58 +0000]
minor formatting updates and typos fixes to the schema role tests

16 years agofixed up the replication test, added some tests for the dbd::multi problem of null...
John Napiorkowski [Fri, 25 Apr 2008 22:33:29 +0000]
fixed up the replication test, added some tests for the dbd::multi problem of null results, fixed replicated.pm docs and a few typos

16 years agorenamed replication storage to replicated (since it's not really doing any replicatio...
John Napiorkowski [Fri, 25 Apr 2008 20:49:07 +0000]
renamed replication storage to replicated (since it's not really doing any replication itself)

16 years agoremoved auto-generated indexes from being explicitly created
Luke Saunders [Fri, 25 Apr 2008 17:35:24 +0000]
removed auto-generated indexes from being explicitly created

16 years agoFix PAUSE indexing of packages
Anders Nor Berle [Thu, 24 Apr 2008 09:04:15 +0000]
Fix PAUSE indexing of packages

16 years agotodo test from zby for set_from_related
Matt S Trout [Wed, 23 Apr 2008 19:47:31 +0000]
todo test from zby for set_from_related

16 years agoadded role for the at query interval run job system
John Napiorkowski [Tue, 22 Apr 2008 19:18:08 +0000]
added role for the at query interval run job system

16 years agoAdd a proxy deploy_statements() method executable directly on $schema instead of...
Peter Rabbitson [Sun, 20 Apr 2008 22:18:28 +0000]
Add a proxy deploy_statements() method executable directly on $schema instead of $schema->storage

16 years agoTODO tests for multiple same level has_many's
Peter Rabbitson [Sun, 20 Apr 2008 18:48:28 +0000]
TODO tests for multiple same level has_many's
Currenty both exception checks pass - should fail until the code is in place

16 years agoSplit prefetch related tests from 76joins.t
Peter Rabbitson [Sun, 20 Apr 2008 18:44:16 +0000]
Split prefetch related tests from 76joins.t

16 years agoadd TODO on constraint check
Matt S Trout [Sun, 20 Apr 2008 16:12:14 +0000]
add TODO on constraint check

16 years agofix loading checks
Matt S Trout [Sun, 20 Apr 2008 15:41:09 +0000]
fix loading checks

16 years agoquerycounter role, test for that and a new schema hierarchy for additional Moose...
John Napiorkowski [Fri, 18 Apr 2008 22:22:41 +0000]
querycounter role, test for that and a new schema hierarchy for additional Moose related development

16 years agoremoved pointless debugging module
John Napiorkowski [Thu, 17 Apr 2008 13:55:11 +0000]
removed pointless debugging module

16 years ago- added tests to replication to make sure dbd::multi fails over when a dtabase is...
John Napiorkowski [Thu, 17 Apr 2008 13:41:17 +0000]
- added tests to replication to make sure dbd::multi fails over when a dtabase is forcible disconnected
- updated DBIC::DBI::Replication so that in addition to a dsn string, you can pass a preexisting database handle.

16 years agoDocumentation updates from omega, carbon
Jess Robinson [Wed, 16 Apr 2008 20:57:22 +0000]
Documentation updates from omega, carbon

16 years agoAdd warnings for non-unique find usage and improve explicit key attr handling in...
Daniel Westermann-Clark [Fri, 11 Apr 2008 21:25:26 +0000]
Add warnings for non-unique find usage and improve explicit key attr handling in find (zby)

16 years agoFix a typo.
Tom Hukins [Wed, 9 Apr 2008 11:18:46 +0000]
Fix a typo.

16 years agoAs of revision 4232, DBIx::Class uses Path::Class. Add this module to
Tom Hukins [Wed, 9 Apr 2008 09:28:32 +0000]
As of revision 4232, DBIx::Class uses Path::Class.  Add this module to
DBIx::Class's dependencies.

16 years agoMissing file
Ash Berlin [Tue, 1 Apr 2008 14:10:31 +0000]
Missing file

16 years agoConstraint/index name fix from rdj
Ash Berlin [Tue, 1 Apr 2008 12:01:16 +0000]
Constraint/index name fix from rdj

16 years agoRemoving since Build.PL is not longer supported in modern M::I
Ash Berlin [Tue, 1 Apr 2008 11:22:15 +0000]
Removing since Build.PL is not longer supported in modern M::I

16 years agofixed typos
Wallace Reis [Thu, 27 Mar 2008 02:35:48 +0000]
fixed typos

16 years agoadded search_related_rs at resultset
Wallace Reis [Wed, 26 Mar 2008 22:23:39 +0000]
added search_related_rs at resultset

16 years agofix pod fail
Matt S Trout [Wed, 26 Mar 2008 21:23:16 +0000]
fix pod fail

16 years agoCDBICompat::Relationships uses Clone. Make CDBICompat not load without it
Michael G Schwern [Wed, 26 Mar 2008 00:38:28 +0000]
CDBICompat::Relationships uses Clone.  Make CDBICompat not load without it
so the tests skip if it's not there.

Also improve the message to only show what's missing.

16 years agoBack all savepoint tests out of (mysql|pg).t
Anders Nor Berle [Tue, 25 Mar 2008 04:03:32 +0000]
Back all savepoint tests out of (mysql|pg).t

There's not a dire need to have them there, the only thing they
would be testing are the ->_svp_* methods on Storage, which are
tested in 98savepoints.t anyway.

I kept the changes in 72pg.t that was neccesary for it to run
to completion though.

16 years agoAnd document the last changes
Anders Nor Berle [Tue, 25 Mar 2008 02:33:50 +0000]
And document the last changes

16 years agoVarious fun things.
Anders Nor Berle [Tue, 25 Mar 2008 02:19:26 +0000]
Various fun things.

DONE:
* We now maintain a stack for savepoints.
* Most savepoint tests are now in a separate test file.
* Arguments to svp_* methods are now optional.
* We throw exceptions instead of warn on many places.
* gphat++

TODO:
* Document all this.