frioux / dbix-class-helpers Goto Github PK

From 3502789f7aaabf4490fb060e238d69258704ee26 Mon Sep 17 00:00:00 2001
From: Niko Tyni <[email protected]>
Date: Fri, 19 Sep 2014 12:54:58 +0300
Subject: [PATCH] Fix a strictness error in a deprecated module

The deprecation messages refer to $VERSION, which isn't
declared despite 'use strict'. This causes a compile
time error.

'+columns => { book_count => { '' => correlated_..., -as => 'book_count' } }

package MyApp::Schema::TestAccessor;
use warnings;
use strict;
use base qw( DBIx::Class::Core );

__PACKAGE__->load_components('Helper::Row::OnColumnMissing');
__PACKAGE__->table('test_accessor');
__PACKAGE__->add_columns(
  id => { },
  title => { },
  user_id => {},
);

sub on_column_missing {'nothing'}

1;

my $obj = $schema->resultset("TestAccessor")->create({id=>2,user_id=>1});
$obj->title; # undef
my @answers = $obj->title; # ()

my $obj = $schema->resultset("TestAccessor")->find(2);
$obj->title; # undef
my @answers = $obj->title; # (undef)

+A::ResultSet::DateMethods1            undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+A::Role::TestConnect                  undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+A::Util                               undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+Lolbot                                undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+ParentRS                              undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+ParentSchema                          undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+ParentSchema::Result                  undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+ParentSchema::Result::Bar             undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+ParentSchema::Result::Foo             undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+RS                                    undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema                            0.001  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::Bar               undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::Bloaty            undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::Foo               undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::Foo_Bar           undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::Gnarly            undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::Gnarly_Station      undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::HasAccessor       undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::HasDateOps        undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::Result::Station           undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::ResultSet                 undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::ResultSet::Bloaty         undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::ResultSet::Foo            undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::ResultSet::Gnarly         undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::ResultSet::HasAccessor      undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::ResultSet::HasDateOps      undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz
+TestSchema::ResultSet::Station        undef  F/FR/FREW/DBIx-Class-Helpers-2.023003.tar.gz

use parent qw{
   DBIx::Class::Helper::ResultSet::IgnoreWantarray
   DBIx::Class::Helper::ResultSet::Random
   DBIx::Class::Helper::ResultSet::Union
   DBIx::Class::Helper::ResultSet::VirtualView
};

08:46:05 ribasushi | <ribasushi> frew: ^^ (last ~10ish lines) perhaps a C::C3::C post-apply hook check that
                   | someone tried to add a resultset helper to a result, or something....?
08:46:29      frew | right I saw that but I don't see why you think that?
08:47:13      frew | like, he posted a thing where he did something stupid, and it has a stacktrace that is a
                   | little confusing but it's hard when someone does something crazy like use an RS component in
                   | a result
08:47:26 ribasushi | because when people end up doing this, the failure ( site / exception text ) is completely
                   | impenetrable unless you know what to look for
08:47:32      frew | ooohhh
08:47:41 ribasushi | mainly because add_columns clashes (both in resultsource proxy and in your helper )
08:47:42      frew | you are recommending a possible additional helper base to check
08:47:44      frew | I hear you
08:48:03 ribasushi | so the very next thing after load_components is CXSA barfing because it is called on a class
08:48:10      frew | right I see
08:48:14 ribasushi | yes!
08:48:14 ribasushi | :)
08:48:33      frew | I'll add it to the list

21:41:38 aran | looking good.  althought I wouldn’t call it more correct than first.  first treats the result*set* like a set, so from  
              | an guts standpoint its as correct as it gets
21:43:00 aran | I’d also personally note that this should, for the sake of portability, just always be used when you want just the first
              | row available and no more no matter what, whether or not it improves anything.  Give it a best-practices bent.
21:44:01 frew | yeah I should document that; how often do people $rs->first and then $rs->next through the rest of the list? (I guess
              | almost neveR)
21:44:14 aran | aye

$result->result_source->resultset->search( $result->ident_condition )

02:05:20 ribasushi | one_row can be made much more efficient, if you aren't worried about calling things that can go away at a later date
02:06:09 ribasushi | calling single is vastly cheaper than next (there is no intermediate DBIx::Class::Storage::Cursor instantiation)
02:06:46 ribasushi | you would need to call ->_resolved_attrs, and then look whether collapse is set and call one or the other
02:07:18 ribasushi | this will also bypass ::Cursor::Cached if one is using it so keeping things calling ->next may be ok
02:07:42 ribasushi | just laying out musings, nothing catastrophically wrong with what you have now

~dbic_repo$ perl -Ilib -It/lib -MDBICTest -MDevel::Dwarn -e '
  Dwarn [ DBICTest->init_schema
                   ->resultset("Artist")
                    ->search({}, { group_by => "name" })
                     ->count_rs
                      ->as_query ]
'

~dbic_repo$ perl -Ilib -It/lib -MDBICTest -MDevel::Dwarn -e '
  require DBICTest::BaseResultSet;
  DBICTest::BaseResultSet->load_components("Helper::ResultSet::AutoRemoveColumns");
  Dwarn [ DBICTest->init_schema
                   ->resultset("Artist")
                    ->search({}, { group_by => "name" })
                     ->count_rs
                      ->as_query ]
'

[ahartmai@ahartmai-nb:~/perl5/perlbrew/perls/22$ (master)]$ cpanm --look DBIx::Class::Helpers
[ahartmai@ahartmai-nb:~/.cpanm/work/1448628618.8509/DBIx-Class-Helpers-2.032000$ ]$ export DBIC_TRACE=1
[ahartmai@ahartmai-nb:~/.cpanm/work/1448628618.8509/DBIx-Class-Helpers-2.032000$ ]$ perl -MCarp::Always t/ResultSet/DateMethods1/oracle.t
    # Subtest: Oracle
        # Subtest: basic
alter session 
  SET nls_date_format = 'YYYY-MM-DD HH24:MI:SS'
alter session 
  SET nls_timestamp_format = 'YYYY-MM-DD HH24:MI:SS.FF'
alter session 
  SET nls_timestamp_tz_format = 'YYYY-MM-DD HH24:MI:SS.FF TZHTZM'

DROP TABLE "Gnarly" CASCADE CONSTRAINTS

CREATE TABLE "Gnarly"( "id" number, NOT NULL, "name" varchar2( 4000 ) NOT NULL, "literature" clob, "your_mom" blob, PRIMARY KEY( "id" ) )

DROP TABLE "HasDateOps" CASCADE CONSTRAINTS

CREATE TABLE "HasDateOps"( "id" number, NOT NULL, "a_date" date NOT NULL, "b_date" date, PRIMARY KEY( "id" ) )
        ok 1 - utc_now
        ok 2 - no floating dates
        ok 3 - datetime correctly UTC and stringified
DELETE FROM "HasDateOps"
INSERT INTO "HasDateOps"( "a_date", "id" ) 
  VALUES( sys_extract_utc( SYSTIMESTAMP ), '1' )
SELECT "me"."id", "me"."a_date", "me"."b_date" 
  FROM "HasDateOps" "me"
        ok 4 - UTC works! (and clock is correct)
        1..4
    ok 1 - basic
        # Subtest: comparisons
DELETE FROM "HasDateOps"
BEGIN WORK
INSERT INTO "HasDateOps"( "a_date", "id" ) 
  VALUES( ?, ? ) : '__BULK_INSERT__'
COMMIT
            # Subtest: datetime object
                # Subtest: no prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" < '2012-12-13 00:00:00'
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" <= '2012-12-13 00:00:00'
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" >= '2012-12-13 00:00:00'
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" > '2012-12-13 00:00:00'
                ok 4 - after
                1..4
            ok 1 - no prefix
                # Subtest: auto prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < '2012-12-13 00:00:00'
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= '2012-12-13 00:00:00'
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= '2012-12-13 00:00:00'
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > '2012-12-13 00:00:00'
                ok 4 - after
                1..4
            ok 2 - auto prefix
                # Subtest: manual prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < '2012-12-13 00:00:00'
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= '2012-12-13 00:00:00'
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= '2012-12-13 00:00:00'
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > '2012-12-13 00:00:00'
                ok 4 - after
                1..4
            ok 3 - manual prefix
            1..3
        ok 1 - datetime object
            # Subtest: datetime literal
                # Subtest: no prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" < '2012-12-13 00:00:00'
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" <= '2012-12-13 00:00:00'
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" >= '2012-12-13 00:00:00'
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" > '2012-12-13 00:00:00'
                ok 4 - after
                1..4
            ok 1 - no prefix
                # Subtest: auto prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < '2012-12-13 00:00:00'
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= '2012-12-13 00:00:00'
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= '2012-12-13 00:00:00'
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > '2012-12-13 00:00:00'
                ok 4 - after
                1..4
            ok 2 - auto prefix
                # Subtest: manual prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < '2012-12-13 00:00:00'
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= '2012-12-13 00:00:00'
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= '2012-12-13 00:00:00'
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > '2012-12-13 00:00:00'
                ok 4 - after
                1..4
            ok 3 - manual prefix
            1..3
        ok 2 - datetime literal
            # Subtest: subquery
                # Subtest: no prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" < ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" <= ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" >= ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" > ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 4 - after
                1..4
            ok 1 - no prefix
                # Subtest: auto prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 4 - after
                1..4
            ok 2 - auto prefix
                # Subtest: manual prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > ( 
    SELECT "me"."a_date" 
      FROM "HasDateOps" "me" 
    WHERE "id" = '2'
   )
                ok 4 - after
                1..4
            ok 3 - manual prefix
            1..3
        ok 3 - subquery
            # Subtest: both columns
DELETE FROM "HasDateOps"
BEGIN WORK
INSERT INTO "HasDateOps"( "a_date", "b_date", "id" ) 
  VALUES( ?, ?, ? ) : '__BULK_INSERT__'
COMMIT
                # Subtest: no prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" < "me"."b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" <= "me"."b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" >= "me"."b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" > "me"."b_date"
                ok 4 - after
                1..4
            ok 1 - no prefix
                # Subtest: auto prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < "me"."b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= "me"."b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= "me"."b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > "me"."b_date"
                ok 4 - after
                1..4
            ok 2 - auto prefix
                # Subtest: manual prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < "me"."b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= "me"."b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= "me"."b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > "me"."b_date"
                ok 4 - after
                1..4
            ok 3 - manual prefix
                # Subtest: no prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" < "b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" <= "b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" >= "b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" > "b_date"
                ok 4 - after
                1..4
            ok 4 - no prefix
                # Subtest: auto prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < "b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= "b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= "b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > "b_date"
                ok 4 - after
                1..4
            ok 5 - auto prefix
                # Subtest: manual prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < "b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= "b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= "b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > "b_date"
                ok 4 - after
                1..4
            ok 6 - manual prefix
                # Subtest: no prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" < "me"."b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" <= "me"."b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" >= "me"."b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "a_date" > "me"."b_date"
                ok 4 - after
                1..4
            ok 7 - no prefix
                # Subtest: auto prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < "me"."b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= "me"."b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= "me"."b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > "me"."b_date"
                ok 4 - after
                1..4
            ok 8 - auto prefix
                # Subtest: manual prefix
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" < "me"."b_date"
                ok 1 - before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" <= "me"."b_date"
                ok 2 - on_or_before
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" >= "me"."b_date"
                ok 3 - on_or_after
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."a_date" > "me"."b_date"
                ok 4 - after
                1..4
            ok 9 - manual prefix
            1..9
        ok 4 - both columns
            # Subtest: literal SQL
SELECT "me"."id" 
  FROM "HasDateOps" "me" 
WHERE "me"."b_date" < sys_extract_utc( SYSTIMESTAMP )
            ok 1 - literal SQL compared (and db clock correct)
            1..1
        ok 5 - literal SQL
        1..5
    ok 2 - comparisons
        # Subtest: add
DELETE FROM "HasDateOps"
BEGIN WORK
INSERT INTO "HasDateOps"( "a_date", "id" ) 
  VALUES( ?, ? ) : '__BULK_INSERT__'
COMMIT
SELECT( TO_TIMESTAMP( "a_date" ) + NUMTODSINTERVAL( '1', 'DAY' ) ) 
  FROM "HasDateOps" "me" 
WHERE "id" = '1'
        ok 1 - live day
        ok 2 - suspected day
        ok 3 - unit: day
SELECT( TO_TIMESTAMP( "a_date" ) + NUMTODSINTERVAL( '2', 'HOUR' ) ) 
  FROM "HasDateOps" "me" 
WHERE "id" = '1'
        ok 4 - live hour
        ok 5 - suspected hour
        ok 6 - unit: hour
SELECT( TO_TIMESTAMP( "a_date" ) + NUMTODSINTERVAL( '3', 'MINUTE' ) ) 
  FROM "HasDateOps" "me" 
WHERE "id" = '1'
        ok 7 - live minute
        ok 8 - suspected minute
        ok 9 - unit: minute
SELECT( TO_TIMESTAMP( "a_date" ) + NUMTODSINTERVAL( '5', 'SECOND' ) ) 
  FROM "HasDateOps" "me" 
WHERE "id" = '1'
        ok 10 - live second
        ok 11 - suspected second
        ok 12 - unit: second
        ok 13 - vanilla add
DELETE FROM "HasDateOps"
INSERT INTO "HasDateOps"( "a_date", "id" ) 
  VALUES( '2013-12-11 10:09:08', '1' )
            # Subtest: column
SELECT "foo" 
  FROM ( 
    SELECT ( TO_TIMESTAMP( TO_TIMESTAMP( TO_TIMESTAMP( "me"."a_date" ) + NUMTODSINTERVAL( '2', 'MINUTE' ) ) + NUMTODSINTERVAL( '4', 'SECOND' ) ) + NUMTODSINTERVAL( '1', 'HOUR' ) ) AS "foo" 
      FROM "HasDateOps" "me"
   ) "me" 
WHERE ROWNUM <= '1'
            # Child (column) exited without calling finalize()
        not ok 14 - column
        #   Failed test 'column'
        #   at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Carp/Always.pm line 28.
        # Child (add) exited without calling finalize()
    not ok 3 - add
    #   Failed test 'add'
    #   at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Carp/Always.pm line 28.
Invalid date format: 2013-12-11 11:11:12.000000000 at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/DateTime/Format/Oracle.pm line 140.
    DateTime::Format::Oracle::parse_datetime("DateTime::Format::Oracle", "2013-12-11 11:11:12.000000000") called at t/lib/A/ResultSet/DateMethods1.pm line 130
    A::ResultSet::DateMethods1::parse_datetime(A::ResultSet::DateMethods1=HASH(0x2992c68), "2013-12-11 11:11:12.000000000") called at t/lib/A/ResultSet/DateMethods1.pm line 343
    A::ResultSet::DateMethods1::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "column", CODE(0x3a04138)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("column", CODE(0x3a04138)) called at t/lib/A/ResultSet/DateMethods1.pm line 351
    A::ResultSet::DateMethods1::__ANON__(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 118
    Test::Roo::Class::each_test(A::ResultSet::DateMethods1=HASH(0x2992c68), CODE(0x19051d8)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo.pm line 38
    Test::Roo::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "add", CODE(0x2993388)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("add", CODE(0x2993388)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo.pm line 39
    Test::Roo::__ANON__(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at (eval 85) line 16
    A::ResultSet::DateMethods1::_do_tests(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 87
    Test::Roo::Class::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 89
    Test::Roo::Class::run_tests("A::ResultSet::DateMethods1", "Oracle") called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/DateTime/Format/Oracle.pm line 140.
    DateTime::Format::Oracle::parse_datetime("DateTime::Format::Oracle", "2013-12-11 11:11:12.000000000") called at t/lib/A/ResultSet/DateMethods1.pm line 130
    A::ResultSet::DateMethods1::parse_datetime(A::ResultSet::DateMethods1=HASH(0x2992c68), "2013-12-11 11:11:12.000000000") called at t/lib/A/ResultSet/DateMethods1.pm line 343
    A::ResultSet::DateMethods1::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "column", CODE(0x3a04138)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("column", CODE(0x3a04138)) called at t/lib/A/ResultSet/DateMethods1.pm line 351
    A::ResultSet::DateMethods1::__ANON__(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 118
    Test::Roo::Class::each_test(A::ResultSet::DateMethods1=HASH(0x2992c68), CODE(0x19051d8)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo.pm line 38
    Test::Roo::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "add", CODE(0x2993388)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("add", CODE(0x2993388)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo.pm line 39
    Test::Roo::__ANON__(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at (eval 85) line 16
    A::ResultSet::DateMethods1::_do_tests(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 87
    Test::Roo::Class::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 89
    Test::Roo::Class::run_tests("A::ResultSet::DateMethods1", "Oracle") called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 279.
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "column", CODE(0x3a04138)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("column", CODE(0x3a04138)) called at t/lib/A/ResultSet/DateMethods1.pm line 351
    A::ResultSet::DateMethods1::__ANON__(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 118
    Test::Roo::Class::each_test(A::ResultSet::DateMethods1=HASH(0x2992c68), CODE(0x19051d8)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo.pm line 38
    Test::Roo::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "add", CODE(0x2993388)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("add", CODE(0x2993388)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo.pm line 39
    Test::Roo::__ANON__(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at (eval 85) line 16
    A::ResultSet::DateMethods1::_do_tests(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 87
    Test::Roo::Class::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 89
    Test::Roo::Class::run_tests("A::ResultSet::DateMethods1", "Oracle") called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 279.
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "add", CODE(0x2993388)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("add", CODE(0x2993388)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo.pm line 39
    Test::Roo::__ANON__(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at (eval 85) line 16
    A::ResultSet::DateMethods1::_do_tests(A::ResultSet::DateMethods1=HASH(0x2992c68)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 87
    Test::Roo::Class::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 261
    Test::Builder::__ANON__() called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    eval {...} called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 266
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 89
    Test::Roo::Class::run_tests("A::ResultSet::DateMethods1", "Oracle") called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/Builder.pm line 279.
    Test::Builder::subtest(Test::Builder=HASH(0xdee7a8), "Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/5.22.0/Test/More.pm line 771
    Test::More::subtest("Oracle", CODE(0x13331e0)) called at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Test/Roo/Class.pm line 89
    Test::Roo::Class::run_tests("A::ResultSet::DateMethods1", "Oracle") called at t/ResultSet/DateMethods1/oracle.t line 18
    # Child (Oracle) exited without calling finalize()
not ok 1 - Oracle
#   Failed test 'Oracle'
#   at /home/ahartmai/perl5/perlbrew/perls/22/lib/site_perl/5.22.0/Carp/Always.pm line 28.
# Tests were run but no plan was declared and done_testing() was not seen.
# Looks like your test exited with 255 just after 1.

15:52:26 ribasushi | so - first of all JSON _en_coding is rather well contained, best seen here: https://metacpan.org/source/SRI/Mojolicious-7.0/lib/Mojo/JSON.pm#L209-260
15:53:26      frew | ok
15:53:51 ribasushi | then if you teach DBIC to recognize a specifc HRI-like class on the lowest possible level like so:
                   | https://metacpan.org/source/RIBASUSHI/DBIx-Class-0.082840/lib/DBIx/Class/ResultSet.pm#L1382-1390
15:53:59      frew | ooooooooh
15:54:18 ribasushi | you can then assemble a parser like so: https://metacpan.org/source/RIBASUSHI/DBIx-Class-0.082840/lib/DBIx/Class/ResultSet.pm#L1432-1439
15:54:26 ribasushi | and DBIC will use it properly
15:55:17 ribasushi | https://metacpan.org/source/RIBASUSHI/DBIx-Class-0.082840/lib/DBIx/Class/ResultSource/RowParser/Util.pm
15:55:27 ribasushi | search for hri_style in this file
15:55:37 ribasushi | it will make it clear how to implement json_style as well
15:55:46 ribasushi | once all of this is assembled/compiled
15:55:53      frew | cool
15:55:56 ribasushi | you have completely functioncall-free JSON assembly
15:56:02      frew | right
15:56:06 ribasushi | just like you have the infres/hri one
15:56:28 ribasushi | which while still in perl space, is going to be a simple loop, and I am pretty sure will beat any transfer to C space

my $row = $c->model('Pool::Passholder')->find($passholder_no, { key => 'primary', columns => [ 'holder_photo' ] })

Index: libdbix-class-helpers-perl-2.019004/lib/DBIx/Class/Helper/ResultSet/RemoveColumns.pm
===================================================================
--- libdbix-class-helpers-perl-2.019004.orig/lib/DBIx/Class/Helper/ResultSet/RemoveColumns.pm   2014-02-14 08:53:27.0000
+++ libdbix-class-helpers-perl-2.019004/lib/DBIx/Class/Helper/ResultSet/RemoveColumns.pm        2014-04-05 13:36:38.2834
@@ -13,7 +13,7 @@

    my $attrs  = $self->{attrs}; # not copying on purpose...

-   if ( $attrs->{remove_columns} ) {
+   if ( !$attrs->{columns} && $attrs->{remove_columns} ) {
       my %rc = map { $_ => 1 } @{$attrs->{remove_columns}};
       $attrs->{columns} = [
          grep { !$rc{$_} } $self->result_source->columns

fix POD
- add whatis part
- drop second NAME
Origin: vendor
Author: gregor herrmann <[email protected]>
Last-Update: 2014-04-07

@frew | package MyApp::Schema::ResultClass::Tee; sub new { return [ ::HRI->new(@_), ::Foo->new(@_) ] }
@frew | the hard part is that you'll want to store the name of ::Foo (say, MyApp::Schema::Result::User) in an attr, so you'll have to make
      | an instantiation method other than new

Description: spelling fixes
Origin: vendor
Author: gregor herrmann <[email protected]>
Last-Update: 2016-07-24

 $self->dt_SQL_add($enddate, 'day', 1)

Description: add abstract to POD
Origin: vendor
Author: gregor herrmann <[email protected]>
Last-Update: 2015-06-28

11:24:33     Perlkonig | https://metacpan.org/source/FREW/DBIx-Class-Helpers-2.031000/lib/DBIx/Class/Helper/ResultSet/Shortcut/ResultsExist.pm <- heh, and
                       | there's ->first again :-)
11:25:30    @ribasushi | frew: ^^ that's indeed odd
11:25:47         @frew | not totally sure I wrote that
11:25:54         @frew | the Shortcuts were mostly contributed
11:26:01    @ribasushi | frew: I'd go for rows => 1 + next, so it implicitly exhausts
11:26:19    @ribasushi | ( in addition to the EXISTS )
11:26:20         @frew | but fwiw, there was a LONG time where I heard a lot of conflicting information about first, next, and single
11:27:23     Perlkonig | wait, what SQL does that even generate? "SELECT EXISTS (SELECT ...);"?
11:27:39     Perlkonig | no, that's way off
11:29:08     Perlkonig | "SELECT * FROM table WHERE EXISTS (SELECT ...);"?
11:29:23         @frew | something along those lins
11:29:29         @frew | you could run it and see :)
11:29:32     Perlkonig | so that would return either no rows, or all rows
11:29:35         @frew | no
11:29:50         @frew | the query would, but we don't fetch it all
11:30:21         @frew | Perlkonig: https://metacpan.org/pod/DBIx::Class::Helper::ResultSet::Shortcut#results_exist
11:30:24         @frew | the doc might help?
11:31:41     Perlkonig | oh, I thought it was undocumented
11:31:52     Perlkonig | I saw it in https://metacpan.org/release/DBIx-Class-Helpers but it had no POD
11:32:03         @frew | yeah it's sorta internal to ::Shortcut
11:32:12     Perlkonig | I did run it
11:32:14     Perlkonig | I was right
11:32:21         @frew | I figured ::Shortcut wouldb e easier if everything were documnted in one place
11:32:24     Perlkonig | it also runs out of memory because it fetches all rows
11:32:25         @frew | Perlkonig: do you use postgres?
11:32:43     Perlkonig | in general, yes, but this app also needs to access a mysql db
11:32:44         @frew | sounds like you are a pg user expecting all dbd's to be as silly as yours :)
11:32:51    @ribasushi | errrrr
11:32:52             * | ribasushi looks closer at the thing
11:33:03    @ribasushi | frew: this is actually entirely incorrect (putting it in where)
11:33:05     Perlkonig | so yeah, ResultsExist is completely broken
11:33:32             * | frew looks at test
11:33:36    @ribasushi | you want to SELECT EXISTS (....) FROM <something, doesn't matter what)
11:33:53    @ribasushi | which will always return 1 row, with a boolean in the 1st col
11:34:11     Perlkonig | or just do the { rows => 1 } thing again
11:34:20     Perlkonig | hmm, but that would still have to transfer all columns
11:34:49         @frew | patches welcome?
11:34:53         @frew | I haven't actually used this helper

11:34:53    @ribasushi | frew: sorry I didn't catch that back when you wrote it, I should have seen that :/
11:35:02         @frew | ribasushi: again: I didn't write it :)
11:35:06         @frew | it was contributed
11:35:09    @ribasushi | hm... was it oalders maybe?
11:35:12         @frew | yessir
11:35:12             * | ribasushi fuzzy
11:35:22         @frew | https://github.com/frioux/DBIx-Class-Helpers/blame/master/lib/DBIx/Class/Helper/ResultSet/Shortcut/ResultsExist.pm
11:35:32    @ribasushi | oalders: ^^ read past ~15 lines
11:35:37       oalders | ok
11:35:43     Perlkonig | - Add ::Shortcut::ResultsExist RS helper (Olaf Alders)
11:35:44         @frew | anyway I'm totally interested in fixing it, I just won't take the blame :)
11:35:46     Perlkonig | oh, too late
11:36:07       oalders | ribasushi: that's a copy/paste of some code you gave me
11:36:08         @frew | hm
11:36:11       oalders | but it seems to work for me
11:36:16         @frew | did I break it somehow? https://github.com/frioux/DBIx-Class-Helpers/commit/fd4024322fb7601352f2caf1ab6c5873f6ea6263
11:36:29         @frew | oalders: they are claiming that while it may work, it pulls in the whole table
11:36:36    @ribasushi | oalders: it works, it's just very inefficient
11:36:43       oalders | yeah, that's not at all what we wanted
11:36:44         @frew | I'm theorizing that it works for everyting but pg
11:36:48     Perlkonig | it works if your table is small enough to fit into memory
11:36:49         @frew | but I don't really know
11:37:28       oalders | i'm using it on Pg. obviously didn't realize there's an issue
11:37:38    @ribasushi | frew: yeah - the original https://github.com/frioux/DBIx-Class-Helpers/commit/fd402432 has the select => ...
11:37:46         @frew | iirc that actually failed tests
11:37:52         @frew | or I'dve left it alone
11:38:01     Perlkonig | I'm running out of memory on a mysql table with 6917243 rows :-)
11:38:10    @ribasushi | right - but the fix changed the concept entirely ;)
11:38:12         @frew | Perlkonig: at the client or the server?
11:38:40       oalders | https://github.com/frioux/DBIx-Class-Helpers/pull/25
11:38:51     Perlkonig | frew: client
11:38:54         @frew | huh
11:39:07     Perlkonig | perl gets oom-killed or runs into ulimits
11:39:11         @frew | sure sure
11:39:58         @frew | ribasushi: so yeah, with his original you get a SQL syntax error
11:40:02    @ribasushi | Perlkonig: problem has been identified, if frew and oalders doesn't have time to fixxxor it - perhaps you can lend a hand? :)
11:40:02         @frew | maybe a bug in core dbic?
11:40:12         @frew | DBIx::Class::Storage::DBI::_prepare_sth(): DBI Exception: DBD::SQLite::db prepare_cached failed: near "(": syntax error [for Statement
                       | "SELECT EXISTS( (SELECT "me"."id", "me"."bar_id" FROM "Foo" "me" WHERE ( "id" > ? )) ) FROM "Foo" "me""] at
                       | t/ResultSet/Shortcut/ResultsExist.t line 16
11:40:41     Perlkonig | that doesn't look right to me either
11:40:43    @ribasushi | frew: sqlite is choking on (( ))
11:40:48     Perlkonig | you don't want to select from Foo
11:40:55         @frew | oalders: fwiw, I could change it back to what you had and only test against pg
11:41:10         @frew | I have that capability in the helpers, and there are some helpers that don't support some engines
11:41:38    @ribasushi | Perlkonig: one needs to select from "somewhere" on most engines
11:41:50             * | frew could regex out the () but that seems perilous
11:41:57         @frew | ribasushi: thoughts re the parens?
11:42:19     Perlkonig | ribasushi: yeah, but then you need some kind of restriction (like rows => 1!), or you'll get as many results as there are rows in the
                       | table
11:42:47    @ribasushi | Perlkonig: yes that's a fair point too
11:43:01     Perlkonig | this is like  (map { exists $foo{key} } keys %foo)
11:43:02     Perlkonig | sort of
11:43:20    @ribasushi | frew: I'd say most expedient is 'EXISTS ' . ${$asq}->[0]
11:44:13         @frew | bummer
11:44:21     Perlkonig | ... in that case, why even bother with exists?
11:44:30     Perlkonig | select 1 from Foo limit 1  
11:44:49         @frew | Perlkonig: well the whole thing is if your db engine needs some other way of checking existence
11:45:00         @frew | I mean, for some people I think doing a limit at all can be pricy?
11:45:05         @frew | it varies from engine to engine
11:45:10    @ribasushi | Perlkonig: because something needs to apply all the $rs conditions
11:45:11         @frew | ribasushi might know better
11:45:33     Perlkonig | ah, true
11:46:04    @ribasushi | Perlkonig: basically the original thinking is to have a proper subquery and hint the optimizer "I just want to know if there's
                       | anything" for portaility
11:46:27    @ribasushi | instead of descending into the nightmares that are limits, and various codepaths depending on the $rs attributes
11:46:40    @ribasushi | nevertheless - given the amount of fail, it may be warranted
11:47:00    @ribasushi | s/warranted/necessary/
11:47:09     Perlkonig | I don't see how you can do this portably without a generic 1-row relation to select from
11:47:10         @frew | s/necesarry/wantarray/
11:47:29    @ribasushi | Perlkonig: in the current form - it won't work this is true  
11:47:56    @ribasushi | Perlkonig: but there may be a portable-ish FROM <> spec or somesuch
11:48:05    @ribasushi | someone needs to do the research basically
11:48:27             * | ribasushi is braincycle-booked for the next week or so, thus someone else stepping up would be awesome
11:50:34     Perlkonig | what do I need to do if I want to write my own dbic component? 
11:50:49         @frew | subclass something, maybe less
11:51:14     Perlkonig | ... huh?
11:51:22         @frew | Perlkonig: like, a component is just a perl package
11:51:44         @frew | you set your base class so that C3 knows what to do (likely DBIx::Class::Core or DBIx::Class::ResultSet)
11:51:53         @frew | and then load the component with load_components
11:52:33             * | frew adds DBICH issue re ResultsExist

diff -bu ./t/row/numifyget.t~ ./t/row/numifyget.t
--- ./t/row/numifyget.t~    2013-07-02 20:40:24.000000000 -0500
+++ ./t/row/numifyget.t 2013-07-25 10:01:23.852572001 -0500
@@ -19,9 +19,9 @@
    my $value = shift;
    my $b_obj = B::svref_2object(\$value);
    my $flags = $b_obj->FLAGS;
-   return ( $flags & B::SVf_IOK or $flags & B::SVp_IOK
+   return (( $flags & B::SVf_IOK or $flags & B::SVp_IOK
           or $flags & B::SVf_NOK or $flags & B::SVp_NOK
-        ) and !($flags & B::SVf_POK )
+           ) and !($flags & B::SVf_POK ))
 }