DBIx-ThinSQL
view release on metacpan or search on metacpan
lib/DBIx/ThinSQL.pm view on Meta::CPAN
# A "where" with an ARRAYref concatenates items together. Note the
# string that is quoted according to the database type.
my $count = $db->xdo(
delete_from => 'actors',
where => [
'actor_id = 1', ' OR ',
'last_name != ', qv("Jones", DBI::SQL_VARCHAR ),
],
);
# Methods for reading from the database depend on the type of
# structure you want back: arrayref or hashref references.
my $ref = $db->xhashref(
select => [ 'id', 'name', qv("Some string") ],
from => 'actors',
where => [
'id = ', qv( 1, DBI::SQL_INTEGER ),
' AND photo IS NOT NULL',
],
limit => 1,
);
$db->xdo(
insert_into => [ 'table', 'col1', 'col2', 'col3' ],
select => [ 't1.col3', 't3.col4', bv( 'value', DBI::SQL_VARCHAR ) ],
from => 'table AS t1',
inner_join => 'other_table AS t2',
on => 't1.something = t2.else',
left_join => 'third_table AS t3',
on => [ 't3.dont = t1.care AND t1.fob = ', qv( 1, DBI::SQL_INT ) ],
where => [],
order_by => [ 't3.dont', 't1.col4' ],
limit => 2,
);
$db->txn( sub {
# Anything you like, done inside a BEGIN/COMMIT pair, with
# nested calls to txn() done inside a SAVEPOINT/RELEASE pair.
})
=head1 DESCRIPTION
Sorry, this documentation is invalid or out of date.
B<DBIx::ThinSQL> is an extension to the Perl Database Interface
(L<DBI>). It is designed for complicated queries and efficient access
to results. With an API that lets you easily write almost-raw SQL,
DBIx::ThinSQL gives you unfettered access to the power and flexibility
of your underlying database. It aims to be a tool for programmers who
want their databases to work just as hard as their Perl scripts.
DBIx::ThinSQL gives you access to aggregate expressions, joins, nested
selects, unions and database-side operator invocations. Transactional
support is provided via L<DBIx::Connector>. Security conscious coders
will be pleased to know that all user-supplied values are bound
properly using L<DBI> "bind_param()". Binding binary data is handled
transparently across different database types.
DBIx::ThinSQL offers a couple of very simple Create, Retrieve, Update
and Delete (CRUD) action methods. These are designed to get you up and
running quickly when your query data is already inside a hashref. The
methods are abstractions of the real API, but should still read as much
as possible like SQL.
Although rows can be retrieved from the database as simple objects,
DBIx::ThinSQL does not attempt to be an Object-Relational-Mapper (ORM).
There are no auto-inflating columns or automatic joins and the code
size and speed reflect the lack of complexity.
DBIx::ThinSQL uses the light-weight L<Log::Any> for logging.
=head1 CONSTRUCTOR
Works like a normal DBI. Can be used with things like
L<DBIx::Connector> to get nice transaction support.
=head1 DBH METHODS
=over
=item share_dir -> Path::Tiny
Returns the path to the distribution share directory. If
C<$DBIx::ThinSQL::SHARE_DIR> is set then that value will be returned
instead of the default method which uses L<File::ShareDir>.
=item throw_error
If B<DBIX::ThinSQL> or a statement raises an exception then the
C<throw_error()> method will be called. By default it just croaks but
classes that inherit from B<DBIx::ThinSQL> can override it. The
original use case was to turn database error text into blessed objects.
=item xprepare
Does a prepare but knows about bind values and quoted values.
=item xprepare_cached
Does a prepare_cached but knows about bind values and quoted values.
=item xval
Creates a statement handle using xprepare(), executes it, and returns
the result of the val() method.
=item xlist
Creates a statement handle using xprepare(), executes it, and returns
the result of the list() method.
=item xarrayref
Does a prepare but knows about bind values and quoted values.
=item xarrayrefs
( run in 1.932 second using v1.01-cache-2.11-cpan-39bf76dae61 )