=head1 NAME

DBIx::Simple::Examples - Examples of how to use DBIx::Simple

=head1 DESCRIPTION

DBIx::Simple provides a simplified interface to DBI, Perl's powerful database
module.

=head1 EXAMPLES

=head2 General

    #!/usr/bin/perl -w
    use strict;
    use DBIx::Simple;

    # Instant database with DBD::SQLite
    my $db = DBIx::Simple->connect('dbi:SQLite:dbname=file.dat')
        or die DBIx::Simple->error;

    # Connecting to a MySQL database
    my $db = DBIx::Simple->connect(
	'DBI:mysql:database=test',     # DBI source specification
	'test', 'test',                # Username and password
	{ RaiseError => 1 }            # Additional options
    );

    # Using an existing database handle
    my $db = DBIx::Simple->connect($dbh);

    # Abstracted example: $db->query($query, @variables)->what_you_want;

    $db->commit or die $db->error;

=head2 Simple Queries

    $db->query('DELETE FROM foo WHERE id = ?', $id) or die $db->error;

    for (1..100) {
	$db->query(
	    'INSERT INTO randomvalues VALUES (?, ?)',
	    int rand(10),
	    int rand(10)
	) or die $db->error;
    }

    $db->query(
	'INSERT INTO sometable VALUES (??)',
	$first, $second, $third, $fourth, $fifth, $sixth
    );
    # (??) is expanded to (?, ?, ?, ?, ?, ?) automatically

=head2 Single row queries

    my ($two)          = $db->query('SELECT 1 + 1')->list;
    my ($three, $four) = $db->query('SELECT 3, 2 + 2')->list;

    my ($name, $email) = $db->query(
	'SELECT name, email FROM people WHERE email = ? LIMIT 1',
	$mail
    )->list;

Or, more efficiently:

    $db->query('SELECT 1 + 1')->into(my $two);
    $db->query('SELECT 3, 2 + 2')->into(my ($three, $four));

    $db->query(
	'SELECT name, email FROM people WHERE email = ? LIMIT 1',
	$mail
    )->into(my ($name, $email));

=head2 Fetching all rows in one go

=head3 One big flattened list (primarily for single column queries)

    my @names = $db->query('SELECT name FROM people WHERE id > 5')->flat;

=head3 Rows as array references

    for my $row ($db->query('SELECT name, email FROM people')->arrays) {
	print "Name: $row->[0], Email: $row->[1]\n";
    }

=head3 Rows as hash references

    for my $row ($db->query('SELECT name, email FROM people')->hashes) {
	print "Name: $row->{name}, Email: $row->{email}\n";
    }

=head2 Fetching one row at a time

=head3 Rows into separate variables

    {
        my $result = $db->query('SELECT name, email FROM people');
        $result->bind(my ($name, $email));
        while ($result->fetch) {
            print "Name: $name, Email: $email\n";
        }
    }

or:

    {
        my $result = $db->query('SELECT name, email FROM people');
        while ($result->into(my ($name, $email))) {
            print "Name: $name, Email: $email\n";
        }
    }

=head3 Rows as lists

    {
	my $result = $db->query('SELECT name, email FROM people');
	while (my @row = $result->list) {
	    print "Name: $row[0], Email: $row[1]\n";
	}
    }

=head3 Rows as array references

    {
	my $result = $db->query('SELECT name, email FROM people');
	while (my $row = $result->array) {
	    print "Name: $row->[0], Email: $row->[1]\n";
	}
    }

=head3 Rows as hash references

    {
	my $result = $db->query('SELECT name, email FROM people');
	while (my $row = $result->hash) {
	    print "Name: $row->{name}, Email: $row->{email}\n";
	}
    }

=head2 Building maps (also fetching all rows in one go)

=head3 A hash of hashes

    my $customers =
	$db
	-> query('SELECT id, name, location FROM people')
	-> map_hashes('id');

    # $customers = { $id => { name => $name, location => $location } }

=head3 A hash of arrays

    my $customers =
	$db
	-> query('SELECT id, name, location FROM people')
	-> map_arrays(0);

    # $customers = { $id => [ $name, $location ] }

=head3 A hash of values (two-column queries)

    my $names =
	$db
	-> query('SELECT id, name FROM people')
	-> map;

    # $names = { $id => $name }

=head1 EXAMPLES WITH SQL::Abstract

If you have SQL::Abstract installed, you can use the abstracting methods
C<select>, C<insert>, C<update>, C<delete>. These work like C<query>, but
instead of a query and bind arguments, use abstracted arguments.

You should read L<SQL::Abstract>. These examples are not enough to fully
understand what is going on.

The SQL::Abstract object is available (writable) through the C<abstract>
property.

The following examples are based on the documentation of SQL::Abstract.

=head2 Overview

If you don't like the defaults, just assign a new object:

    $db->abstract = SQL::Abstract->new(
        case    => 'lower',
        cmp     => 'like', 
        logic   => 'and', 
        convert => 'upper'
    );

If you don't assign any object, one will be created automatically using the
default options. The SQL::Abstract module is loaded on demand.

    my $result = $db->select($table, \@fields, \%where, \@order);
    my $result = $db->insert($table, \%fieldvals || \@values);
    my $result = $db->update($table, \%fieldvals, \%where);
    my $result = $db->delete($table, \%where);

    for ($result->hashes) { ... }

=head2 Complete examples

=head3 select

    my @tickets = $db->select(
        'tickets', '*', {
            requestor => 'inna',
            worker    => ['nwiger', 'rcwe', 'sfz'],
            status    => { '!=', 'completed' }
        }
    )->hashes;

=head3 insert

If you already have your data as a hash, inserting becomes much easier:

    $db->insert('people', \%data);

Instead of:

    $db->query(
        q[
            INSERT 
            INTO people (name, phone, address, ...)
            VALUES (??)
        ],
        @data{'name', 'phone', 'address', ... }
    );

=head3 update, delete

    $db->update(
        'tickets', {
            worker    => 'juerd',
            status    => 'completed'
        },
        { id => $id }
    )

    $db->delete('tickets', { id => $id });

=head3 where

The C<where> method is not wrapped directly, because it doesn't generate a
query and thus doesn't really have anything to do with the database module.

But using the C<abstract> property, you can still easily access it:

    my $where = $db->abstract->where({ foo => $foo });

=head1 EXAMPLES WITH DBIx::XHTML_Table

If you have DBIx::XHTML_Table installed, you can use the result methods
C<xto> and C<html>. 

You should read L<DBIx::XHTML_Table>. These examples are not enough to fully
understand what is going on. When reading that documentation, note that you
don't have to pass hash references to DBIx::Simple's methods. It is supported,
though.

DBIx::XHTML_Table is loaded on demand.

=head2 Overview

To print a simple table, all you have to do is:

    print $db->query('SELECT * FROM foo')->html;

Of course, anything that produces a result object can be used. The same thing
using the abstraction method C<select> would be:

    print $db->select('foo', '*')->html;

A DBIx::XHTML_Table object can be generated with the C<xto> (B<X>HTML_B<T>able
B<O>bject) method:

    my $table = $db->query($query)->xto;

=head2 Passing attributes

DBIx::Simple sends the attributes you pass to C<html> both to the constructor
and the output method. This allows you to specify both HTML attributes (like 
C<bgcolor>) and options for XHTML_Table (like C<no_ucfirst> and C<no_indent>)
all at once:

    print $result->html(
        tr         => { bgcolor => [ qw/silver white/ ] },
        no_ucfirst => 1
    );

=head2 Using an XHTML_Table object

Not everything can be controlled by passing attributes. For full flexibility,
the XHTML_Table object can be used directly:

    my $table = $db->query($query)->xto(
        tr => { bgcolor => [ qw/silver white/ ] }
    );

    $table->set_group('client', 1);
    $table->calc_totals('credit', '%.2f');

    print $table->output({ no_ucfirst => 1 });  # note the {}!

=head1 EXAMPLES WITH Text::Table

=over 8

=item C<< $result->text("neat") >>

Neither neat nor pretty, but useful for debugging. Uses DBI's C<neat_list>
method. Doesn't display column names.

    '1', 'Camel', 'mammal'
    '2', 'Llama', 'mammal'
    '3', 'Owl', 'bird'
    '4', 'Juerd', undef

=item C<< $result->text("table") >>

Displays a simple table using ASCII lines.

    id | animal |  type
    ---+--------+-------
     1 |  Camel | mammal
     2 |  Llama | mammal
     3 |  Owl   | bird
     4 |  Juerd |

=item C<< $result->text("box") >>

Displays a simple table using ASCII lines, with an outside border.

    +----+--------+--------+
    | id | animal |  type  |
    +----+--------+--------+
    |  1 |  Camel | mammal |
    |  2 |  Llama | mammal |
    |  3 |  Owl   | bird   |
    |  4 |  Juerd |        |
    +----+--------+--------+

=back

For C<table> and C<box>, you need Anno Siegel's Text::Table module installed.

=head1 LICENSE

There is no license. This software was released into the public domain. Do with
it what you want, but on your own risk. The author disclaims any
responsibility.

=head1 AUTHOR

Juerd Waalboer <juerd@cpan.org> <http://juerd.nl/>

=head1 SEE ALSO

L<DBIx::Simple>, L<SQL::Abstract>

=cut