NAME
Test::DBUnit - Database test framework.
SYNOPSIS
use DBIx::Connection;
use Test::DBUnit connection_name => 'test';
use Test::More tests => $tests;
DBIx::Connection->new(
name => 'test',
dsn => $ENV{DB_TEST_CONNECTION},
username => $ENV{DB_TEST_USERNAME},
password => $ENV{DB_TEST_PASSWORD},
);
#or
use Test::More;
use Test::DBUnit dsn => 'dbi:Oracle:localhost:1521/ORACLE_INSTANCE', username => 'user', password => 'password';
plan tests => $tests;
my $connection = test_connection();
my $dbh = test_dbh();
reset_schema_ok('t/sql/create_schema.sql');
populate_schema_ok('t/sql/create_schema.sql');
xml_dataset_ok('test1');
#you database operations here
$connection->execute_statement("UPDATE ....");
expected_xml_dataset_ok('test1');
#or
reset_sequence_ok('table1_seq1');
dataset_ok(
table1 => [column1 => 'x', column2 => 'y'],
table1 => [column1 => 'x1_X', column2 => 'y1_X'],
...
table2 => [column1 => 'x2, column2 => 'y2'],
table2 => [column1 => 'x1_N', column2 => 'y1_N'],
);
#you database operations here
$connection->execute_statement("UPDATE ....");
expected_dataset_ok(
table1 => [column1 => 'z', column2 => 'y'],
)
DESCRIPTION
Database test framework to verify that your database data match expected set of values.
Managing test data
Database tests should giving you complete and fine grained control over the test data that is used.
use Test::DBUnit dsn => $dsn, username => $username, password => $password;
reset_schema_ok('t/sql/create_schema.sql');
populate_schema_ok('t/sql/create_schema.sql');
reset_sequence_ok('emp_seq');
Loading test data sets
Before you want to test your business logic it is essential to have repeatable snapshot of your tables you want to test, so this module allows you populate/synchronize your database with the passed in data structure or with the xml content.
dataset_ok(
emp => [ename => "john", deptno => "10", job => "project manager"],
emp => [ename => "scott", deptno => "10", job => "project manager"],
bonus => [ename => "scott", job => "project manager", sal => "20"],
);
or
xml_dataset_ok('test1');
t/test_unit.test1.xml #given that you testing module is t/test_unit.t
<?xml version='1.0' encoding='UTF-8'?>
<dataset load_strategy="INSERT_LOAD_STRATEGY">
<emp empno="1" ename="scott" deptno="10" job="project manager" />
<emp empno="2" ename="john" deptno="10" job="engineer" />
<bonus ename="scott" job="project manager" sal="20" />
</dataset>
Getting connection to test database
my $connection = test_connection();
#business logic that change tested data comes here
....
Verifying test results
It can be useful to use data sets for checking the contents of a database after is has been modified by a test. You may want to check the result of a update/insert/delete method or a stored procedure.
expected_dataset_ok(
emp => [empno => "1", ename => "Scott", deptno => "10", job => "project manager"],
emp => [empno => "2", ename => "John", deptno => "10", job => "engineer"],
emp => [empno => "3", ename => "Mark", deptno => "10", job => "sales assistant"],
bonus => [ename => "scott", job => "project manager", sal => "20"],
)
or
expected_xml_dataset_ok('test1');
t/test_unit.test1-result.xml #given that you testing module is t/test_unit.t
<?xml version='1.0' encoding='UTF-8'?>
<dataset>
<emp empno="1" ename="Scott" deptno="10" job="project manager" />
<emp empno="2" ename="John" deptno="10" job="engineer" />
<emp empno="3" ename="Mark" deptno="10" job="sales assistant" />
<bonus ename="scott" job="project manager" sal="20" />
</dataset>
Configuring the dataset load strategy
By default, datasets are loaded into the database using an insert load strategy. This means that all data in the tables that are present in the dataset is deleted, after which the test data records are inserted. Order in with all data is deleted depends on reverse table occurrence in the dataset, however you may force order of data by specifying empty table:
table1 => [], #this fore delete operation in occurrence order
table1 => [col1 => 1, col2 => 'some data'],
or in xml file
<table1 />
<table1 col1="1" col2="some data"/>
In this strategy number of rows will be validated against datasets in (xml_)expexted_dataset_ok method. Load strategy behavior is configurable, it can be modified by calling:
set_insert_load_strategy();
or in XML
<?xml version='1.0' encoding='UTF-8'?>
<dataset load_strategy="INSERT_LOAD_STRATEGY">
<emp empno="1" ename="Scott" deptno="10" job="project manager" />
....
</dataset>
set_refresh_load_strategy();
or in XML
<?xml version='1.0' encoding='UTF-8'?>
<dataset load_strategy="REFRESH_LOAD_STRATEGY">
<emp empno="1" ename="Scott" deptno="10" job="project manager" />
....
</dataset>
The alternative to the insert load strategy is refresh load strategy. In this case update on existing rows will take place or insert occurs if rows are missing.
head3 Tests with multiple database instances.
You may need to test data from more then one database instance, so that you have to specify connection againt which tests will be performed either by adding prefix to test methods, or by seting explicit test connection context.
use Test::DBUnit connection_names => ['my_connection_1', 'my_connection_2'];
my $dbh = DBI->connect($dsn_1, $username, $password);
add_test_connection('my_connection_1', dbh => $dbh);
my $connection = DBIx::Connection->new(
name => 'my_connection_2',
dsn => $dsn_2,
username => $username,
password => $password,
);
add_test_connection($connection);
#set connection context by prefix
my_connection_1_reset_schema_ok('t/sql/create_schema_1.sql');
my_connection_1_populate_schema_ok('t/sql/create_schema_1.sql');
#set connection context explicitly.
set_test_connection('my_connection_2');
reset_schema_ok('t/sql/create_schema_2.sql');
populate_schema_ok('t/sql/create_schema_2.sql');
my_connection_2_set_test_connection('my_connection_2');
Working with sequences
You may use sequences or auto generated features, so this module allows you handle that.
reset_sequence_ok('emp_seq');
or for MySQL
reset_sequence_ok('test_table_name')
The ALTER TABLE test_table_name AUTO_INCREMENT = 1 will be issued Note that for MySQL reset sequence the test_table_name must be empty.
or in XML
<?xml version='1.0' encoding='UTF-8'?>
<dataset reset_sequences="emp_seq, dept_seq">
<emp empno="1" ename="Scott" deptno="10" job="project manager" />
....
</dataset>
Sequence tests with Oracle
t/sql/create_schema.sql
CREATE SEQUENCE emp_seq;
CREATE TABLE emp(
empno NUMBER NOT NULL,
ename VARCHAR2(10),
job VARCHAR2(20),
mgr NUMBER(4),
hiredate DATE,
sal NUMBER(7,2),
comm NUMBER(7,2),
deptno NUMBER(2),
CONSTRAINT emp_pk PRIMARY KEY(empno),
FOREIGN KEY (deptno) REFERENCES dept (deptno)
);
CREATE OR REPLACE TRIGGER emp_autogen
BEFORE INSERT ON emp FOR EACH ROW
BEGIN
IF :new.empno is null then
SELECT emp_seq.nextval INTO :new.empno FROM dual;
END IF;
END;
#unit test
reset_sequence_ok('emp_seq');
dataset_ok(
emp => [ename => "John", deptno => "10", job => "project manager"],
emp => [ename => "Scott", deptno => "10", job => "project manager"]
);
....
expected_dataset_ok(
emp => [empno => 1, ename => "John", deptno => "10", job => "project manager"],
emp => [empno => 2, ename => "Scott", deptno => "10", job => "project manager"]
)
Sequence tests with PostgreSQL
t/sql/create_schema.sql
CREATE SEQUENCE emp_seq;
CREATE TABLE emp(
empno INT4 DEFAULT nextval('emp_seq') NOT NULL,
ename VARCHAR(10),
job VARCHAR(20),
mgr NUMERIC(4),
hiredate DATE,
sal NUMERIC(7,2),
comm NUMERIC(7,2),
deptno NUMERIC(2),
CONSTRAINT emp_pk PRIMARY KEY(empno),
FOREIGN KEY (deptno) REFERENCES dept (deptno)
);
#unit test
reset_sequence_ok('emp_seq');
....
Auto generated filed values tests with MySQL
t/sql/create_schema.sql
CREATE TABLE emp(
empno MEDIUMINT AUTO_INCREMENT,
ename VARCHAR(10),
job VARCHAR(20),
mgr NUMERIC(4),
hiredate DATE,
sal NUMERIC(7,2),
comm NUMERIC(7,2),
deptno NUMERIC(2),
CONSTRAINT emp_pk PRIMARY KEY(empno),
FOREIGN KEY (deptno) REFERENCES dept (empno)
);
#unit test
reset_sequence_ok('emp');
dataset_ok(
emp => [ename => "John", deptno => "10", job => "project manager"],
emp => [ename => "Scott", deptno => "10", job => "project manager"]
);
....
expected_dataset_ok(
emp => [empno => 1, ename => "John", deptno => "10", job => "project manager"],
emp => [empno => 2, ename => "Scott", deptno => "10", job => "project manager"]
)
Working with LOBs
For handling very large datasets, the DB vendors provide the LOB (large object) data types. You may use this features, and this module allows you test it.
LOBs tests with Oracle
Oracle BLOB data type that contains binary data with a maximum size of 4 gigabytes. It is advisable to store blob size in separate column to optimize fetch process.(doc_size)
CREATE TABLE image(id NUMBER, name VARCHAR2(100), doc_size NUMBER, blob_content BLOB);
dataset_ok(
emp => [empno => 1, ename => 'scott', deptno => 10],
image => [id => 1, name => 'Moon'
blob_content => {file => 'data/chart1.jpg', size_column => 'doc_size'}
]
);
.....
expected_dataset_ok(
emp => [empno => 1, ename => 'scott', deptno => 10],
image => [id => 1, name => 'Moon'
blob_content => {file => 'data/chart2.jpg', size_column => 'doc_size'}
]
);
LOBs tests with PostgreSQL
PostgreSQL has a large object facility, but in this case the tested table doesn't contain LOBs type but keeps reference to lob_id, created by lo_creat PostgreSQL function. It is required to store blob size in separate column to be able fetch blob.(doc_size)
CREATE TABLE image(id NUMERIC, name VARCHAR(100), doc_size NUMERIC, blob_content oid)
dataset_ok(
emp => [empno => 1, ename => 'scott', deptno => 10],
image => [id => 1, name => 'Moon'
blob_content => {file => 'data/chart1.jpg', size_column => 'doc_size'}
]
);
LOBs test with MySQL
In MySQL, binary LOBs are just fields in the table, so storing blob size is optional.
CREATE TABLE lob_test(id NUMERIC, name VARCHAR(100), doc_size NUMERIC, blob_content LONGBLOB)
dataset_ok(
emp => [empno => 1, ename => 'scott', deptno => 10],
image => [id => 1, name => 'Moon'
blob_content => {file => 'data/chart1.jpg', size_column => 'doc_size'}
]
);
EXPORT
expected_data_set_ok dataset_ok expected_xml_dataset_ok xml_dataset_ok reset_schema_ok populate_schema_ok reset_sequence_ok set_refresh_load_strategy set_insert_load_strategy add_test_connection set_test_connection test_connection test_dbh by default.
METHODS
- connection_name
- generate_connection_test_stubs
-
Generated test stubs on fly for passed in connection names.
- reset_schema_ok
-
Tests database schema reset using sql file. Takes file name as parameter.
use Test::More tests => $tests; use Test::DBUnit dsn => $dsn, username => $username, password => $password; ... reset_schema_ok('t/sql/create_schema.sql');
- populate_schema_ok
-
Tests database schema population using sql file. Takes file name as parameter.
use Test::More tests => $tests; use Test::DBUnit dsn => $dsn, username => $username, password => $password; ... populate_schema_ok('t/sql/populate_schema.sql');
- reset_sequence_ok
-
Resets database sequence. Takes sequence name as parameter.
use Test::More tests => $tests; use Test::DBUnit dsn => $dsn, username => $username, password => $password; reset_sequnce('table_seq1');
- xml_dataset_ok
-
Tests database schema population/synch to the content of the xml file. Takes test unit name, that is used to resolve xml file name. Xml file name that will be loaded is build as follow <test_file>.<unit_name>.xml for instance the following invocation xml_dataset_ok('test1') from t/sub_dir/001_test.t file will expect t/sub_dir/001_test.test1.xml file.
<dataset load_strategy="INSERT_LOAD_STRATEGY" reset_sequences="emp_seq"> <emp ename="scott" deptno="10" job="project manager" /> <emp ename="john" deptno="10" job="engineer" /> <emp ename="mark" deptno="10" job="sales assistant" /> <bonus ename="scott" job="project manager" sal="20" /> </dataset>
- expected_xml_dataset_ok
-
Validates expected database loaded from xml file against database schema. Takes test unit name, that is used to resolve xml file name. Xml file name that will be loaded is build as follow <test_file>.<unit_name>.xml unless you pass full xml file name. for instance the following invocation xml_dataset_ok('test1') from t/sub_dir/001_test.t file will expect t/sub_dir/001_test.test1.xml file.
<dataset load_strategy="INSERT_LOAD_STRATEGY" reset_sequences="emp_seq,dept_seq"> <emp ename="Scott" deptno="10" job="project manager" /> <emp ename="John" deptno="10" job="engineer" /> <emp ename="Mark" deptno="10" job="sales assistant" /> <bonus ename="Scott" job="project manager" sal="20" /> </dataset>
- dataset_ok
-
Tests database schema population/synch to the passed in dataset.
dataset_ok( table1 => [], #this deletes all data from table1 (DELETE FROM table1) table2 => [], #this deletes all data from table2 (DELETE FROM table2) table1 => [col1 => 'va1', col2 => 'val2'], #this insert or update depend on strategy table1 => [col1 => 'xval1', col2 => 'xval2'], )
- expected_dataset_ok
-
Validates database schema against passed in dataset.
expected_dataset_ok( table1 => [col1 => 'va1', col2 => 'val2'], )
- _initialise_connection
-
Initialises default test connection
- test_connection_context
-
Returns tested connection name,
- test_connection
-
Returns test connection object.
- add_test_connection
-
Adds tests connection
use Test::DBUnit; # or use Test::DBUnit connection_names => ['my_connection_name', 'my_connection_name1']; my $connection = DBIx::Connection->new(...); add_test_connection($connection); #or add_test_connection('my_connection_name', dsn => $dsn, username => $username, password => 'password'); #or add_test_connection('my_connection_name', dbh => $dbh);
- set_test_connection
-
Sets test connection that will be tested.
- test_dbh
-
Returns test database handler.
- set_insert_load_strategy
-
Sets insert as the load strategy
- set_refresh_load_strategy
-
Sets refresh as the load strategy
- _xml_test_file
-
Returns xml file prefix to test
COPYRIGHT AND LICENSE
The Test::DBUnit module is free software. You may distribute under the terms of either the GNU General Public License or the Artistic License, as specified in the Perl README file.
SEE ALSO
AUTHOR
Adrian Witas, adrian@webapp.strefa.pl