11 years agoStop needlesly invoking the $rs bool overload
Peter Rabbitson [Mon, 18 Mar 2013 06:27:51 +0000]
Stop needlesly invoking the $rs bool overload

11 years agoReorder prefetch logic for optimal execution (no functional changes)
Peter Rabbitson [Sat, 16 Mar 2013 10:22:12 +0000]
Reorder prefetch logic for optimal execution (no functional changes)

Avoid yet another expensive try call

11 years agoNormalize varnames (zero functional changes)
Peter Rabbitson [Sun, 10 Mar 2013 18:51:21 +0000]
Normalize varnames (zero functional changes)

11 years agoFix test description (DrHyde++)
Peter Rabbitson [Mon, 11 Mar 2013 14:36:29 +0000]
Fix test description (DrHyde++)

11 years agoMerge branch 'master' into topic/constructor_rewrite
Peter Rabbitson [Wed, 17 Apr 2013 07:34:50 +0000]
Merge branch 'master' into topic/constructor_rewrite

Add some extra code to enforce the assumption that any bind type constant
is accessible in _dbi_attrs_for_bind, or in other words that all necessary
DBDs are already loaded (concept originally introduced in ad7c50fc)

Without this the combination of 9930caaf7e (do not recalculate bind attrs
on dbh_do retry) and a2f228547 (do not wrap iterators in dbh_do) can result
in _dbi_attrs_for_bind being called before DBI/DBD::* has been loaded at all

11 years agoRemove idiotic RowCountOrGenericSubQ - it will never work as part of as_query
Peter Rabbitson [Thu, 21 Mar 2013 08:07:32 +0000]
Remove idiotic RowCountOrGenericSubQ - it will never work as part of as_query

I am not sure how I overlooked this, blame is on me :(

Replacing with a ROWCOUNT fallback at the appropriate place

11 years agoChanges pod references to deprecated SQLT methods
Matt Phillips [Mon, 15 Apr 2013 15:55:30 +0000]
Changes pod references to deprecated SQLT methods

quote_table_names/quote_field_names is deprecated, refer to quote_identifiers
instead.

11 years agoAdd a TODO to catch when Text::Balanced RT#74994 is finally fixed
Peter Rabbitson [Thu, 11 Apr 2013 05:57:56 +0000]
Add a TODO to catch when Text::Balanced RT#74994 is finally fixed

11 years agoIn example code, placing a necessary semicolon after a try/catch code-block, and...
Jason McIntosh [Wed, 10 Apr 2013 20:23:25 +0000]
In example code, placing a necessary semicolon after a try/catch code-block, and correcting a nearby variable name.

11 years agoFix oversight in subqueried MySQL update/delete
Peter Rabbitson [Tue, 9 Apr 2013 13:15:12 +0000]
Fix oversight in subqueried MySQL update/delete

11 years agoRelease v0.08210 v0.08210
Peter Rabbitson [Thu, 4 Apr 2013 07:22:25 +0000]
Release v0.08210

11 years agoTrap and display STDERR during travis test runs
Peter Rabbitson [Thu, 4 Apr 2013 03:20:43 +0000]
Trap and display STDERR during travis test runs

11 years agoMore shuffling of travis testdeps
Peter Rabbitson [Thu, 4 Apr 2013 03:50:17 +0000]
More shuffling of travis testdeps

11 years agoMake sure external DBIC envvars do not cause tests to fail
Peter Rabbitson [Wed, 3 Apr 2013 13:59:16 +0000]
Make sure external DBIC envvars do not cause tests to fail

Run the appropriate tests on travis to detect future problems

11 years agoFix incorrect warning/exception originator
Peter Rabbitson [Thu, 24 Nov 2011 09:01:59 +0000]
Fix incorrect warning/exception originator

Until now we would report the second to last function as the
exception originator, whereas we need to report the first after
skipping the usual suspects of throw_exception etc

11 years agoDo not repeatedly trace test schema populate under TravisCI
Peter Rabbitson [Thu, 4 Apr 2013 02:23:23 +0000]
Do not repeatedly trace test schema populate under TravisCI

11 years agoExtra TODO tests for a txn_guard silencing problem
Peter Rabbitson [Thu, 4 Apr 2013 02:20:39 +0000]
Extra TODO tests for a txn_guard silencing problem

11 years agoWork around SQLite's RT#79576
Peter Rabbitson [Mon, 1 Apr 2013 10:07:51 +0000]
Work around SQLite's RT#79576

Includes a warning that data may have been corrupted in previous
DBIC versions, with resources on how to deal with this.

11 years agoSaner check for non-integer values bound as integers in SQLite
Peter Rabbitson [Mon, 1 Apr 2013 09:05:05 +0000]
Saner check for non-integer values bound as integers in SQLite

11 years agoSimplify _select_args_to_query - contortions not needed since 0e773352
Peter Rabbitson [Mon, 1 Apr 2013 06:16:09 +0000]
Simplify _select_args_to_query - contortions not needed since 0e773352

11 years agoRefactor sth preparation/binding - no functional changes
Peter Rabbitson [Sun, 31 Mar 2013 13:06:59 +0000]
Refactor sth preparation/binding - no functional changes

Notable internal changes:

- _dbi_attrs_for_bind call is moved out of _dbh_execute - there is no point
  recalculating these on retry
- _dbh_execute changed signature: $ident => $bind_attrs
- sth, _sth and _dbh_sth are no more - instead we now have _prepare_sth and
  _bind_sth_params

The test in t/storage/base.t has not been actually working for years (it
did not register a spurious success with $count == 0). Removing it is safe
as t/storage/reconnect.t tests the same codepath more thoroughly

11 years agoDESCRIPTION typo fix
David Steinbrunner [Fri, 29 Mar 2013 14:52:01 +0000]
DESCRIPTION typo fix

11 years agoFix test warnings left in by 11343b34
Peter Rabbitson [Mon, 25 Mar 2013 10:21:20 +0000]
Fix test warnings left in by 11343b34

11 years agoimprove outline by introducing synopsis subheadings
Lars Dɪᴇᴄᴋᴏᴡ 迪拉斯 [Sun, 24 Mar 2013 11:24:57 +0000]
improve outline by introducing synopsis subheadings

11 years agoimprove mark-up
Lars Dɪᴇᴄᴋᴏᴡ 迪拉斯 [Sun, 24 Mar 2013 11:23:24 +0000]
improve mark-up

11 years agodevelopment info is less important than user info
Lars Dɪᴇᴄᴋᴏᴡ 迪拉斯 [Sun, 24 Mar 2013 11:05:11 +0000]
development info is less important than user info

11 years agoAdd daxim to contribs, specify main doc encoding
Lars Dɪᴇᴄᴋᴏᴡ 迪拉斯 [Sun, 24 Mar 2013 11:31:58 +0000]
Add daxim to contribs, specify main doc encoding

11 years ago9 years of perl and I somehow did not know that...
Peter Rabbitson [Sun, 24 Mar 2013 07:28:53 +0000]
9 years of perl and I somehow did not know that...

11 years agoFix copypasta - strictures test needs no global lock
Peter Rabbitson [Sat, 23 Mar 2013 12:42:44 +0000]
Fix copypasta - strictures test needs no global lock

11 years agoOnly display output of failed workers in parallel_installdeps_notest
Peter Rabbitson [Thu, 21 Mar 2013 08:23:41 +0000]
Only display output of failed workers in parallel_installdeps_notest

11 years agoFile::Path needs to be upgraded separately on smoking
Peter Rabbitson [Wed, 20 Mar 2013 11:09:51 +0000]
File::Path needs to be upgraded separately on smoking

11 years agoAdd deprecation warnings for cols/include_columns
Darius Jokilehto [Tue, 19 Mar 2013 12:18:09 +0000]
Add deprecation warnings for cols/include_columns

* Also update docs and include test

11 years agoFix test failing with some tracing envvars (and smoke on travis too)
Peter Rabbitson [Tue, 19 Mar 2013 08:02:05 +0000]
Fix test failing with some tracing envvars (and smoke on travis too)

11 years agoEXPERIMENTAL Release v0.08242 v0.08242
Peter Rabbitson [Sun, 10 Mar 2013 14:45:57 +0000]
EXPERIMENTAL Release v0.08242

11 years agoProperly wrap a subquery around grouped resultsets in case of $rscol->func($func)
Peter Rabbitson [Sat, 23 Feb 2013 15:44:13 +0000]
Properly wrap a subquery around grouped resultsets in case of $rscol->func($func)

11 years agoRadically rethink complex prefetch - make most useful cases just work (tm)
Peter Rabbitson [Thu, 28 Feb 2013 08:31:13 +0000]
Radically rethink complex prefetch - make most useful cases just work (tm)

TL;DR: mst - I AM SORRY!!! I will rebase the dq branch for you when this
pile of eyebleed goes stable.

The long version - since we now allow arbitrary prefetch, the old
_prefetch_selector_range mechanism doesn't cut it anymore. Instead we
recognize prefetch solely based on _related_results_construction.
Furthermore group_by/limits do not play well with right-side order_by
(which we now also support, by transforming foreign order criteria into
aggregates).

Thus a much more powerful introspection is needed to decide what goes on
the inside and outside of the prefetch subquery. This is mostly done now
by the augmented _resolve_aliastypes_from_select_args to track
identifiers it saw (97e130fa48), and by extra logic considering what
exactly are we grouping by.

Everything is done while observing the "group over selection +
aggregates only" rule, which sould allow us to remain RDBMS agnostic
(even for pathological cases of "MySQL-ish aggregates").

As a bonus more cases of "the user knows what they are doing" are now
correctly recognized and left alone. See a t/prefetch/with_limit.t diff
for a general idea of the scope of improvements.

Yes - there is more regexing crap in the codebase now, and it is
possible we will call _resolve_aliastypes_from_select_args up to 4(!!!)
times per statement preparation. However this allows us to establish a
set of test cases towards which to write optimizations/flog the dq
framework.

11 years agoThere is no point of ordering the insides of a complex prefetch without a limit
Peter Rabbitson [Sat, 9 Mar 2013 18:41:41 +0000]
There is no point of ordering the insides of a complex prefetch without a limit

11 years agoConsider unselected order_by during complex subqueried prefetch
Peter Rabbitson [Sat, 9 Mar 2013 10:57:39 +0000]
Consider unselected order_by during complex subqueried prefetch

Augment _resolve_aliastypes_from_select_args to collect the column names
it sees, allowing it to replace _extract_condition_columns() entirely.

In the process fix a number of *incorrect* limit_dialect tests

11 years agoWarn in case of iterative collapse being upgraded to an eager cursor slurp
Peter Rabbitson [Thu, 7 Mar 2013 12:12:58 +0000]
Warn in case of iterative collapse being upgraded to an eager cursor slurp

11 years agoMove CDBI code out of the main ResultSet implementation
Peter Rabbitson [Tue, 5 Mar 2013 06:01:33 +0000]
Move CDBI code out of the main ResultSet implementation

11 years agoStill do null-branch pruning when we are using our own inflate_result()
Peter Rabbitson [Tue, 5 Mar 2013 05:49:53 +0000]
Still do null-branch pruning when we are using our own inflate_result()

52864fbd was too aggressive in removing stuff - inheritance no longer
matters, but which method we will be calling does. If we are still using
our own inflate_result() there is no point of doing the double work of
blessing and then skipping null branches.

11 years agoForgotten cruft after 52864fbd5
Peter Rabbitson [Tue, 5 Mar 2013 01:55:51 +0000]
Forgotten cruft after 52864fbd5

11 years agoRename some private methods/attributes to align with what they do
Peter Rabbitson [Thu, 28 Feb 2013 17:12:26 +0000]
Rename some private methods/attributes to align with what they do

No functional changes, just a s///

11 years agoRemove some superfluous expensive try blocks
Peter Rabbitson [Thu, 28 Feb 2013 07:44:08 +0000]
Remove some superfluous expensive try blocks

We already know that the relationship exists, and ->related_source without
a schema returns undef anyway

11 years agoSimplify more - remove loop and cleanup the while cond
Peter Rabbitson [Thu, 28 Feb 2013 07:48:11 +0000]
Simplify more - remove loop and cleanup the while cond

11 years agoAn underdefined root has to use the %cur_row_ids container
Peter Rabbitson [Wed, 27 Feb 2013 10:19:43 +0000]
An underdefined root has to use the %cur_row_ids container

Otherwise we are polluting $cur_row_data with the extra "virtual id column"
value, and may end up stashing it back on iterative cursor parsing

11 years agoMOAR optimization - tracking is_new_res is double work
Peter Rabbitson [Wed, 27 Feb 2013 10:01:12 +0000]
MOAR optimization - tracking is_new_res is double work

We can get the root node assembler to do that for us instead

Also remove more of the superfluous ||= assignments

11 years agoBAIL_OUT stops the entire test run - do not want that
Peter Rabbitson [Wed, 27 Feb 2013 09:35:17 +0000]
BAIL_OUT stops the entire test run - do not want that

11 years agoScale back validation of the 'as' attribute (revert 95e41036)
Peter Rabbitson [Sun, 24 Feb 2013 17:45:23 +0000]
Scale back validation of the 'as' attribute (revert 95e41036)

It turned out that users employ invalid-but-sensical inflation maps in
production. Remove the early sanity check of the infmap and move the code
throwing an exception all the way back to ::Row

Follow through this thread for more details:
http://lists.scsys.co.uk/pipermail/dbix-class/2013-February/011115.html

11 years agoClarify result_class tweaking on active/cached cursors
Peter Rabbitson [Sun, 24 Feb 2013 16:56:53 +0000]
Clarify result_class tweaking on active/cached cursors

We need this to simplify logic of result_class type caching

11 years agoRemove test cruft forgotten after 52864fbd
Peter Rabbitson [Sun, 24 Feb 2013 10:49:58 +0000]
Remove test cruft forgotten after 52864fbd

11 years agoDo not rely on Data::Dumper to produce rolled out hashrefs
Peter Rabbitson [Sun, 24 Feb 2013 10:41:58 +0000]
Do not rely on Data::Dumper to produce rolled out hashrefs

11 years agoMerge branch 'master' into topic/constructor_rewrite
Peter Rabbitson [Sun, 10 Mar 2013 11:10:06 +0000]
Merge branch 'master' into topic/constructor_rewrite

11 years agoAbraxxa-proof the distbuild (disallow `make manifest dist`)
Peter Rabbitson [Sun, 10 Mar 2013 10:21:30 +0000]
Abraxxa-proof the distbuild (disallow `make manifest dist`)

This does not work under dmake (there is no $(MAKECMDGOALS)), but it does
not seem to have adverse effects either

11 years agoExclude DBICTest namespace from stacktraces
Peter Rabbitson [Sun, 10 Mar 2013 10:05:06 +0000]
Exclude DBICTest namespace from stacktraces

11 years agoAugment 1b6cbedd (reverting my moronic use of preamble in EU::MM)
Peter Rabbitson [Fri, 8 Mar 2013 17:00:57 +0000]
Augment 1b6cbedd (reverting my moronic use of preamble in EU::MM)

Instead of whacking things during the MY::distdir override, properly
override MY::postamble for upload prereq check injection

11 years agomove the location of the extra upload checks so it isn't the default target
Graham Knop [Fri, 8 Mar 2013 16:56:39 +0000]
move the location of the extra upload checks so it isn't the default target

This should result in equivalent behavior since it is still before the
main upload rule.

11 years agoMore comprehensive test of qualifier stripping for update/delete
Peter Rabbitson [Fri, 8 Mar 2013 10:06:33 +0000]
More comprehensive test of qualifier stripping for update/delete

11 years agoChange the rt link from one requiring a login to a publicly accessible one
Christian Walde [Tue, 5 Mar 2013 16:24:28 +0000]
Change the rt link from one requiring a login to a publicly accessible one

Also switch to https:// whatever can be reasonably switched

11 years agoReorganize and freshen up the DocMap
Peter Rabbitson [Tue, 5 Mar 2013 05:22:24 +0000]
Reorganize and freshen up the DocMap

It accumulated lots of stuff that is confusing / not relevant for beginners

11 years agoMove DocMap link to top of DBIx::Class to prevent it being missed
Christian Walde [Mon, 4 Mar 2013 13:28:33 +0000]
Move DocMap link to top of DBIx::Class to prevent it being missed

Reorganize the section to be a bit more comprehensive

11 years agoRemove outdated (and kinda offensive) CONTRIBUTING text
Peter Rabbitson [Tue, 5 Mar 2013 03:47:07 +0000]
Remove outdated (and kinda offensive) CONTRIBUTING text

A replacement should come shortly, courtesy of Getty

11 years agoAnd more fulltest dep conflicts
Peter Rabbitson [Tue, 5 Mar 2013 05:52:21 +0000]
And more fulltest dep conflicts

11 years agoClarify wtf do we have a _bool
Peter Rabbitson [Tue, 5 Mar 2013 03:25:11 +0000]
Clarify wtf do we have a _bool

11 years agoRequire latest EU::MM at dist-building time to avoid $-escaping issues
Christian Walde [Fri, 1 Mar 2013 17:08:19 +0000]
Require latest EU::MM at dist-building time to avoid $-escaping issues

11 years agoDo not auto-install dist-building dependencies in author mode
Peter Rabbitson [Sat, 2 Mar 2013 06:16:46 +0000]
Do not auto-install dist-building dependencies in author mode

Instead only require them for actual make dist/distdir/upload runs

11 years agoSmoke all branches on travis except for blocked* and wip/*
Peter Rabbitson [Fri, 1 Mar 2013 16:02:03 +0000]
Smoke all branches on travis except for blocked* and wip/*

11 years agoAnother adjustment of travis depinstall order
Peter Rabbitson [Fri, 1 Mar 2013 15:41:20 +0000]
Another adjustment of travis depinstall order

11 years agoPunt TODO until the dateops work merges
Peter Rabbitson [Fri, 1 Mar 2013 09:05:38 +0000]
Punt TODO until the dateops work merges

11 years agoThis infrastructure was removed ages ago by 263e41be
Peter Rabbitson [Fri, 1 Mar 2013 08:46:54 +0000]
This infrastructure was removed ages ago by 263e41be

11 years agoRemove TODO that could never be made to work (revert c49fcf72)
Peter Rabbitson [Fri, 1 Mar 2013 08:42:15 +0000]
Remove TODO that could never be made to work (revert c49fcf72)

I do not know what groditi was thinking - the behavior codified makes 0 sense

11 years agoRelease v0.08209 v0.08209
Alexander Hartmaier [Fri, 1 Mar 2013 13:09:42 +0000]
Release v0.08209

11 years agoFulltests deps clarification
Peter Rabbitson [Thu, 28 Feb 2013 14:46:48 +0000]
Fulltests deps clarification

- CDBI::Plugin::DeepAbstractSearch already depends on CDBI
- Install DateTime earlier in the CLEANTEST=false cycle

11 years agoGuard against exceptions in _describe_connection
Peter Rabbitson [Thu, 28 Feb 2013 09:02:56 +0000]
Guard against exceptions in _describe_connection

Stuff like this happens otherwise:
http://lists.scsys.co.uk/pipermail/dbix-class/2013-January/011085.html

11 years agoWhitespace-check the travis scripts as well
Peter Rabbitson [Wed, 27 Feb 2013 12:32:00 +0000]
Whitespace-check the travis scripts as well

11 years agoShuffle travis CLEANTEST=false depchain again
Peter Rabbitson [Wed, 27 Feb 2013 10:33:13 +0000]
Shuffle travis CLEANTEST=false depchain again

11 years agoTest::Strict is smarter now
Peter Rabbitson [Wed, 27 Feb 2013 10:27:50 +0000]
Test::Strict is smarter now

11 years agoRegex /o is *EVIL* - no idea why I even thought of using that
Peter Rabbitson [Mon, 18 Feb 2013 07:37:54 +0000]
Regex /o is *EVIL* - no idea why I even thought of using that

11 years agoWarn on signs of Moose-borkage (in case it gets to be the first new())
Peter Rabbitson [Tue, 26 Feb 2013 08:08:44 +0000]
Warn on signs of Moose-borkage (in case it gets to be the first new())

11 years agoAdjust CLEANTEST=false depinstall a bit more
Peter Rabbitson [Sat, 23 Feb 2013 15:16:45 +0000]
Adjust CLEANTEST=false depinstall a bit more

11 years agoFix cookbook - missing cond in 'Equivalent SQL'
David Schmidt [Wed, 20 Feb 2013 13:46:08 +0000]
Fix cookbook - missing cond in 'Equivalent SQL'

11 years agoInstrument the lockfile creation failure
Peter Rabbitson [Sat, 23 Feb 2013 14:52:50 +0000]
Instrument the lockfile creation failure

Attempting to get to the bottom of http://www.cpantesters.org/cpan/report/27f16417-89ea-1014-a33e-b361b1e9163f

11 years agoFix regression breaking search on prefetched rel (broken by 5e2a0518)
Peter Rabbitson [Sat, 23 Feb 2013 14:17:47 +0000]
Fix regression breaking search on prefetched rel (broken by 5e2a0518)

Boy 0.08205 was a bad release, so much crap went in undetected by
any of our tests :( Maybe it's time for Devel::Cover...?

11 years agoUncomment disabled test (not sure why it was)
Peter Rabbitson [Sat, 23 Feb 2013 14:13:32 +0000]
Uncomment disabled test (not sure why it was)

11 years agoEXPERIMENTAL Release v0.08241 v0.08241
Peter Rabbitson [Wed, 20 Feb 2013 10:56:29 +0000]
EXPERIMENTAL Release v0.08241

11 years agoClarify prefetch documentation a bit, remove the nonsensical 0.09 warning
Peter Rabbitson [Wed, 20 Feb 2013 11:05:31 +0000]
Clarify prefetch documentation a bit, remove the nonsensical 0.09 warning

11 years agoSanity-check columns added as primary key
Peter Rabbitson [Wed, 20 Feb 2013 10:49:45 +0000]
Sanity-check columns added as primary key

11 years agoOptimize the HRI-direct collapser even more
Peter Rabbitson [Wed, 20 Feb 2013 10:12:35 +0000]
Optimize the HRI-direct collapser even more

Since we know that we will never output null-branches in the case of
HRI, we can wrap the related assemblers in a definedness condition to
never fire on a null branch, and skip entirely the calculation of
non-NULL %cur_row_ids.

11 years agoFatalize warnings within the compiled row parsers
Peter Rabbitson [Tue, 19 Feb 2013 20:57:13 +0000]
Fatalize warnings within the compiled row parsers

Encountering an undef value anywhere in this code is a sign of some glaring
omission. Perhaps a nullable column is annotated as is_nullable => 0 or
something similar. When this is the case we need the collapsing to stop
right there, as it is no longer safe to proceed.

11 years agoModify the null-branch pruning introduced in ce556881, restore compat
Peter Rabbitson [Tue, 19 Feb 2013 20:54:16 +0000]
Modify the null-branch pruning introduced in ce556881, restore compat

In the case of "direct to HRI" we want to throw stuff away as before. In
other cases however we no longer take an educated guess when to prune.
Instead we mark each null-branch by simply blessing its containing
arrayref. This allows us to keep the arguments to inflate_result 100%
backwards compatible, while at the same time allowing "enlightened"
inflate_result implementations to skip over the dead branches with
minimal effort.

Both ::Row::inflate_result and ::HRI::inflate_result were adjusted to
react correctly to these marks. As a result the HRI folder gained
another 5% speedup (unless sidestepped by the direct-to-HRI code, which
is naturally much much more efficient).

The current name of the "null class" is stored in the global
$DBIx::Class::ResultSource::RowParser::Util::null_branch_class. Given
that the entire codebase begs for standalone CPAN extraction this should
do for the time being.

While implementing the test changes it became clear that
Test::More::is_deeply does not cut it (it ignores the blessed-ness of
structures by design: https://github.com/schwern/test-more/issues/347).
As a result a dep on Test::Deep was added, and several of the tests
looking at inflate_result RVs were converted.

11 years agoAdd optdep on CPAN::Uploader to get an effortless `make upload`
Peter Rabbitson [Wed, 20 Feb 2013 09:57:03 +0000]
Add optdep on CPAN::Uploader to get an effortless `make upload`

11 years agoEven more distbuilding checks
Peter Rabbitson [Wed, 20 Feb 2013 09:35:00 +0000]
Even more distbuilding checks

Frew's environment (and *only* his environment) keeps producing weirdly
broken dists. Add an extra check for this until we figure out how exactly
his environment is broken.

Extra check tested on cygwin and strawberry for good measure.

11 years agoRelease v0.08208 v0.08208
Peter Rabbitson [Wed, 20 Feb 2013 09:58:31 +0000]
Release v0.08208

11 years agoAdd optdep on CPAN::Uploader to get an effortless `make upload`
Peter Rabbitson [Wed, 20 Feb 2013 09:57:03 +0000]
Add optdep on CPAN::Uploader to get an effortless `make upload`

11 years agoEven more distbuilding checks
Peter Rabbitson [Wed, 20 Feb 2013 09:35:00 +0000]
Even more distbuilding checks

Frew's environment (and *only* his environment) keeps producing weirdly
broken dists. Add an extra check for this until we figure out how exactly
his environment is broken.

Extra check tested on cygwin and strawberry for good measure.

11 years agoNew Test::Strict no longer detects Moose::Role (RT#83433) - workaround
Peter Rabbitson [Mon, 18 Feb 2013 20:22:26 +0000]
New Test::Strict no longer detects Moose::Role (RT#83433) - workaround

11 years agoReadjust the parallel no-test dependency install order
Peter Rabbitson [Sat, 16 Feb 2013 16:44:54 +0000]
Readjust the parallel no-test dependency install order

Better utilize the massive amount of CPUs we have available

11 years agoIt appears that travis drastically increased the amount of available CPUs
Peter Rabbitson [Sat, 16 Feb 2013 16:05:12 +0000]
It appears that travis drastically increased the amount of available CPUs

Well, this is awesome - we just happen to be equipped to take advantage
of this ;)

Remove the branch-name limitations - now a smoke takes fraction of the time
it used to, no point to limit what we smoke anymore.

At the same time reduce the amount of threads building the perls - older
binaries do not build very well on -j32. This is a temporary thing - the
perl-building will soon be exported away anyway.

11 years agoNew Test::Strict no longer detects Moose::Role (RT#83433) - workaround
Peter Rabbitson [Mon, 18 Feb 2013 20:22:26 +0000]
New Test::Strict no longer detects Moose::Role (RT#83433) - workaround

11 years agoPrevent SQL::Translator::Producer::YAML from seeing our potential $dbh
Peter Rabbitson [Sat, 16 Feb 2013 15:27:23 +0000]
Prevent SQL::Translator::Producer::YAML from seeing our potential $dbh

The failure mode here is extremely convoluted, full of unimaginable fail
and makes baby jesus cry :(

The long story: there are two ways to supply a $schema to the SQLT producer:
via an argument to translate() (sane) or at the time of translator instance
creation (less sane). In the second case this becomes a problem because some
serializing producers (e.g. SQLT::Producer::YAML) take the *entire*
SQLT instance and dump it to YAML. This includes the arguments which include
our schema with its storage and potentially its $dbh. This causes the
weird warnings described in e.g. RT#75394, because YAML creates a dead
DBI object upon de-serialization of said arguments. What's even more sad is
that while the YAML Producer dumps the entire SQLT instance, the YAML Parser
simply takes the ->{schema} part of this structure and throws everything
else away.

Of course we can not just fix the ::Producer - folks may very well be relying
on the structure it spits out (nobody knows if the *only* consumer of the
Producer::YAML output is in fact Parser::YAML). Really the only thing we can
know (well not know, but reasonably assume) is that there is no way that any
user relies on the accessing the storage passed in with a schema instance to
be parsed.

So what we do is the least invasive thing of all - at translate() time we
check the state of the passed in $schema and if it does have a storage
instance we simply detach it (by cloning the $schema instance) and as a
result everything remains happy.