NAME
RDBAL::Schema - RDBAL Schema information object
SYNOPSIS
use RDBAL;
use RDBAL::Schema;
$X = RDBAL::Connect('username', 'password', 'server');
$schema = new($X,$database);
$database = $schema->Database();
@user_tables = $schema->User_Tables();
@views = $schema->Views();
@procedures = $schema->Procedures();
@system_tables = $schema->System_Tables();
@fields = $schema->Table_Fields($table);
@fields = $schema->System_Table_Fields($table);
@fields = $schema->View_Fields($view);
@fields = $schema->Procedure_Parameters($procedure);
@fields = $schema->Fields($object,$object_type);
$field_info = $schema->Field_Info($object,$field,$object_type,$info_type);
$primary_key_number = $schema->Primary_Key($object,$field,$object_type);
$field_type = $schema->Field_Type($object,$field,$object_type);
$field_length = $schema->Field_Length($object,$field,$object_type);
$field_width = $schema->Field_Width($object,$field,$object_type);
$field_precision = $schema->Field_Precision($object,$field,$object_type);
$field_scale = $schema->Field_Scale($object,$field,$object_type);
$field_identity = $schema->Field_Identity($object,$field,$object_type);
$field_null = $schema->Field_Null($object,$field,$object_type);
@indexes = $schema->Indexes($table,$object_type);
@primary_keys = $schema->Primary_Keys($table,$object_type);
@keys = $schema->Keys($table,$object_type);
@children_tables = $schema->Children($table);
@parent_tables = $schema->Parents($table);
@field_equivalences = $schema->Relation($parent,$child);
# Get comments (views' and procedures' definitions)
$comments = $schema->Comments($object,$object_type)
# Get view's tables
@tables = $schema->View_Tables($view)
ABSTRACT
This perl library uses perl5 objects to make it easy to retrieve information about a particular PostgreSQL, Sybase, MS SQL, or Oracle databases's schema.
INSTALLATION:
If you wish to change the location of the schema cache directory from the default value of '/usr/local/schema_cache', edit Config.pm.
To install this package, just change to the directory in which this file is found and type the following:
perl Makefile.PL
make
make test
make install
and to create the schema cache directory:
make schema_cache
Installation of pg_schema.pl
An example script for converting a DDL in Transact SQL format to PostgreSQL format with foreign key relational integrity enforced and remembered by PostgreSQL rules is included and must be manually installed (change the first line from #!/usr/local/bin/perl if your perl lives elsewhere). See pg_schema.pl below for more.
DESCRIPTION
The schema information available includes:
- Objects: tables, views, and procedures
- Objects' fields (or parameters)
- Objects' fields' properties: type, length, precision, scale, identity column, nullable
- Tables' indexes and primary keys
- Parent => child relations between tables including primary key/foreign key equivalences.
- Views' and Procedures' definitions (Comments).
- A view's underlying tables.
The database connection is cached in the schema object. Objects and their fields properties and index information are retrieved when the schema object is created. Table relationship information is retrieved for all tables when the first relationship information is requested.
pg_schema.pl
The rules this script creates are necessary for RDBAL to have foreign key relation information. It translates a foreign key reference into PostgreSQL rules:
FROM:
alter table table2
add foreign key (field1, field2)
references table1 (field1, field2)
go
TO:
Insert RI:
create rule fk_table2_insert as on insert to table2
where new.field1 <> table1.field1 and new.field2 <> table1.field1
do instead nothing;
Update RI:
create rule fk_table2_update as on update to table2
where new.field1 <> table1.field1 and new.field2 <> table1.field2
do instead nothing;
create rule pk_table1_update as on update to table1
where old.field1 = table2.field1 and old.field2 = table2.field2
do instead nothing;
NOTE: the pk_update rule must list all children!
Delete RI:
create rule pk_table1_delete as on delete to table1
where old.field1 = table2.field1 and old.field2 = table2.field2
do instead nothing;
NOTE: the delete rule must list all children!
This script also maps datatypes:
pg_schema.pl reads the input DDL on standard input and writes the output DDL to standard output.
CREATING A NEW RDBAL::Schema OBJECT:
$query = new RDBAL::Schema($connection,$database);
OR
$query = new RDBAL::Schema($connection,$database, -option => value);
Options are passed as: -option => value, where -option is one of:
-server Database server name. This is used to differentiate
between databases when caching.
-server_type Database server type. This is used to differentiate
how to retrieve the schema. The default is Transact-SQL
or a hand-crafted schema cache file. Currently, the
only correct values for this are I<undef>, I<pg>,
I<oracle>, I<dbi:Sybase>, or I<dbi:Oracle>.
-get_system 1 or undef. A true value for this option causes retrieval
(and caching) of schema for system tables.
-nocache 1 or undef. A true value causes the cached schema to not
be used and a new cache to be written.
-username Database username. This is used to differentiate
between different users views of a database when caching.
This will create a new schema object for the database. This must be given an open connection to a RDBAL database server object:
use RDBAL;
$connection = RDBAL::Connect('username', 'password', 'server');
Fetching the database from the schema object:
$database = $schema->Database();
The database may be retrieved from the database schema object.
Fetching the user tables from the schema:
@user_tables = $schema->User_Tables();
The user tables may be retrieved from the database schema.
Fetching the views from the schema:
@views = $schema->Views();
The views may be retrieved from the database schema.
Fetching the procedures from the schema:
@procedures = $schema->Procedures();
The procedures may be retrieved from the database schema.
Fetching the system tables from the schema:
@system_tables = $schema->System_Tables();
The system tables may be retrieved from the database schema if the -get_system option was given when the schema object (or its cache) was created.
Fetching the fields from the schema for a user table:
@fields = $schema->Table_Fields($table);
A user table's fields may be retrieved from the database schema.
Fetching the fields from the schema for a system table:
@fields = $schema->System_Table_Fields($table);
A system table's fields may be retrieved from the database schema if the -get_system option was given when the schema object (or its cache) was created.
Fetching the fields from the schema for a view:
@fields = $schema->View_Fields($view);
A view's fields may be retrieved from the database schema.
Fetching the parameters from the schema for a procedure:
@parameters = $schema->Procedure_Parameters($procedure);
A procedure's parameters may be retrieved from the database schema.
Fetching the fields from the schema:
@fields = $schema->Fields($object,$object_type);
$object_type is optional and defaults to 'User Table'.
The fields may be retrieved from the database schema for $object_type's of:
Fetching the primary key number from the schema for a field:
$primary_key_number = $schema->Primary_Key($object,$field,$object_type);
$object_type is optional and defaults to 'User Table'.
The primary key number may be retrieved from the database schema for a field. undef is returned if the field is not a primary key.
Fetching the field type from the schema:
$field_type = $schema->Field_Type($object,$field,$object_type);
$object_type is optional and defaults to 'User Table'.
The field type may be retrieved from the database schema.
Fetching the field length from the schema:
$field_length = $schema->Field_Length($object,$field,$object_type);
$object_type is optional and defaults to 'User Table'.
The field length may be retrieved from the database schema.
Fetching the field width from the schema:
$field_width = $schema->Field_Width($object,$field,$object_type);
$object_type is optional and defaults to 'User Table'.
A value for the character string width of a field may be retrieved from the database schema.
Fetching the field precision from the schema:
$field_precision = $schema->Field_Precision($object,$field,$object_type);
$object_type is optional and defaults to 'User Table'.
The field precision may be retrieved from the database schema.
Fetching the field scale from the schema:
$field_scale = $schema->Field_Scale($object,$field,$object_type);
$object_type is optional and defaults to 'User Table'.
The field scale may be retrieved from the database schema. If the field datatype does not have a scale, the value is undef.
Fetching the field's identity column status from the schema:
$field_identity = $schema->Field_Identity($object,$field,$object_type);
$object_type is optional and defaults to 'User Table'.
The field's identity column status may be retrieved from the database schema. Nonzero implies the field is an identity column.
Fetching the field nullable from the schema:
$field_null = $schema->Field_Null($object,$field,$object_type);
$object_type is optional and defaults to 'User Table'.
Whether a field is nullable may be retrieved from the database schema. It is a 1 if the field is nullable.
Fetching a field's (or parameter's) information from the schema:
$field_info = $schema->Field_Info($object,$field,$object_type,$info_type);
$object_type is optional and defaults to 'User Table'. $info_type is optional and defaults to 'Type'.
A field's information may be retrieved from the database schema for $object_type's of:
and $info_type's of:
- The primary key number may be retrieved from the database schema for a field. undef is returned if the field is not a primary key.
- The field type may be retrieved from the database schema.
- The field length may be retrieved from the database schema.
- The field precision may be retrieved from the database schema.
- The field scale may be retrieved from the database schema.
- The field's identity column status may be retrieved from the database schema. Nonzero implies the field is an identity column.
- Whether a field is nullable may be retrieved from the database schema. It is a 1 if the field is nullable.
Fetching the indexes from the schema:
@indexes = $schema->Indexes($table,$object_type);
$object_type is optional and defaults to 'User Table'.
The indexes may be retrieved from the database schema. Each index is reported as (each item seperated by tabs):
index_name index_description comma_seperated_index_field_list
Example:
PK_STS clustered, unique located on default chromosome, arm, id_number
Fetching the primary key fields from the schema:
@primary_keys = $schema->Primary_Keys($table,$object_type);
$object_type is optional and defaults to 'User Table'.
The primary key fields may be retrieved from the database schema.
Fetching all index keys from the schema:
@keys = $schema->Keys($table,$object_type);
$object_type is optional and defaults to 'User Table'.
All of the index keys may be retrieved from the database schema. Each array element is a comma delimited list of the index's keys.
Fetching the children tables from the schema:
@children_tables = $schema->Children($table);
A table's children tables may be retrieved from the database schema.
Fetching the parent tables from the schema:
@parent_tables = $schema->Parents($table);
A table's parent tables may be retrieved from the database schema.
Fetching the field equivalences from the schema:
@field_equivalences = $schema->Relation($parent,$child);
The key field equivalences may be retrieved from the database schema. Each key field equivalence array element is given as: parent_key_field=child_key_field
Fetching the comments (views' and procedures' definitions) from the schema:
$comments = $schema->Comments($object,$object_type)
$object_type is optional and defaults to 'Procedure'.
The comments (views' and procedures' definitions) may be retrieved from the database schema.
Getting a view's underlying tables from the schema:
@tables = $schema->View_Tables($view)
A view's underlying tables may be retrieved from the database schema.
Quote a field's value if necessary
$quoted_field = $schema->Quote_Field($object,$field,$value,$object_type);
Appropriately put quote marks around a field's value. Single quote marks get doubled, example: dont't ==> "don''t".
Schema Cache File Format
The schema cache file contains individual lines which describe some part of a schema (a descriptor). Each descriptor is a tab delimited line with the first element on the line specifying what is being described (the type). Carriage returns (not newlines) may be embedded in the comment (definition) descriptor elements. Blank lines follow each group of index descriptors for a table. Line feeds seperate different types of descriptors.
Object Descriptor
An object descriptor describes the fields or parameters of a system or user table, a view, or a procedure. It contains the following (tab delimited) elements:
- Type = 'O'
- Object = owner.object (owner defaults to 'dbo')
- Object type:
-
S => 'System Table' U => 'User Table' V => 'View' P => 'Procedure' TABLE => 'User Table' VIEW => 'View' PROCEDURE => 'Procedure'
- Field name
- Field type
- Field length
- Field precision
- Field scale
- Identity field (1 if true)
- Nullable field (1 if true)
Primary Key Descriptor
A primary key descriptor lists the primary key for a table and the fields which comprise the primary key. It contains the following (tab delimited) elements:
- Type = 'PK'
- Name of object having primary key
- Object type ('User Table' or 'System Table')
- Database
- Owner (default is 'dbo')
- Object (table)
- Primary Key field
- Position in primary key
- Primary Key name (optional)
Foreign Key Descriptor
A foreign key descriptor lists the foreign key relationships for a table. Each descriptor pairs a primary key field name of one table with the corresponding foreign key field name of another table. It also gives the position each of those fields within the primary and foreign key tuples. It contains the following (tab delimited) elements:
- Type = 'FK'
- Object type ('User Table' or 'System Table')
- Database
- Owner (default is 'dbo')
- Primary Key Object (table)
- Primary Key field
- Database
- Owner (default is 'dbo')
- Foreign Key Object (table)
- Foreign Key field
- Foreign key field position in tuple
- Update rule number -- always 1, if present
- Delete rule number -- always 1, if present
- Foreign Key name (Optional)
- Primary Key name (Optional)
Index Descriptors
An index descriptor lists the indexes for a table and the fields which comprise that index. It contains the following (tab delimited) elements (blank lines follow each set of indexes for a table):
- Type = 'I'
- Index Object (table)
- Index Object type ('User Table' or 'System Table')
- Index Name
- Index description
- Comma seperated, ordered index field list
Comments (and stored procedure definitions) Descriptors
A comment (or definition) descriptor contains the following (tab delimited) elements:
- Type = 'C'
- Object
- Object Type is one of: 'User Table', 'System Table', 'View', 'Procedure'
- Comment or definition (carriage returns replace newlines and tabs are replaced by five spaces
Example Script
#!/usr/local/bin/perl
use RDBAL
use RDBAL::Schema;
$server = shift;
$database = shift;
$username = shift;
$password = shift;
# Check to see if we want to use a different name for the server
if ($RDBAL::Layer{'SybaseDBlib'} || $RDBAL::Layer{'ApacheSybaseDBlib'}) {
$server = 'sybase_sql';
} else {
$server = 'odbc_sql';
}
if (!defined($server) ||
!defined($database) ||
!defined($username) ||
!defined($password)) {
die "Usage is: get_schema.pl server database username password\n";
}
# Get connnection to database server
( $X = RDBAL::Connect($username,$password,$server)
or (die "Failed to connect to $server $username"));
$schema = new RDBAL::Schema($X,$database, -get_system => 1);
$, = "\t";
print "Info for database: " . $schema->Database() . "\n";
print "User Tables:\n";
map {
$table = $_;
print "\tTable: $table\n";
map {
print "\t\t". $_ . (($schema->Primary_Key($table,$_)) ? '*' : ''),
$schema->Field_Type($table,$_),
$schema->Field_Length($table,$_),
$schema->Field_Precision($table,$_),
$schema->Field_Scale($table,$_),
(($schema->Field_Identity($table,$_)) ? 'Identity' : ''),
(($schema->Field_Null($table,$_)) ? 'NULL' : 'NONNULL')
. "\n";
} $schema->Table_Fields($table);
print "\t\tPrimary keys:", $schema->Primary_Keys($table,'User Table'),"\n";
map {
@keys = split(',',$_);
print "\t\tIndex keys:", @keys ,"\n";
} $schema->Keys($table,'User Table');
map {
($index_name, $index_description, $keys) = split("\t",$_);
print "\t\tIndexes:\t$index_name\t$keys\t$index_description\n";
} $schema->Indexes($table,'User Table');
print "\t\tComments:", $schema->Comments($table,'User Table'), "\n";
map {
print "\t\tParents: $_ (Reverse)", $schema->Relation($_,$table),"\n";
} $schema->Parents($table);
map {
print "\t\tChildren: $_", $schema->Relation($table,$_),"\n";
} $schema->Children($table);
} $schema->User_Tables();
print "System Tables:\n";
map {
$table = $_;
print "\tTable: $table\n";
map {
print "\t\t". $_ . (($schema->Primary_Key($table,$_,'System Table')) ? '*' : ''),
$schema->Field_Type($table,$_,'System Table'),
$schema->Field_Length($table,$_,'System Table'),
$schema->Field_Precision($table,$_,'System Table'),
$schema->Field_Scale($table,$_,'System Table'),
(($schema->Field_Identity($table,$_,'System Table')) ? 'Identity' : ''),
(($schema->Field_Null($table,$_,'System Table')) ? 'NULL' : 'NONNULL')
. "\n";
} $schema->System_Table_Fields($table);
print "\t\tComments:", $schema->Comments($table,'System Table'), "\n";
} $schema->System_Tables();
print "Views:\n";
map {
$table = $_;
print "\tView: $table (Tables:", $schema->View_Tables($_), ")\n";
map {
print "\t\t". $_,
$schema->Field_Type($table,$_,'View'),
$schema->Field_Length($table,$_,'View'),
$schema->Field_Precision($table,$_,'View'),
$schema->Field_Scale($table,$_,'View'),
(($schema->Field_Null($table,$_,'View')) ? 'NULL' : 'NONNULL')
. "\n";
} $schema->View_Fields($table);
print "\tComments:", $schema->Comments($table,'View'), "\n";
} $schema->Views();
print "Procedures:\n";
map {
$table = $_;
print "\tProcedure: $table\n";
map {
print "\t\t". $_,
$schema->Field_Type($table,$_,'Procedure'),
$schema->Field_Length($table,$_,'Procedure'),
$schema->Field_Precision($table,$_,'Procedure'),
$schema->Field_Scale($table,$_,'Procedure'),
(($schema->Field_Null($table,$_,'Procedure')) ? 'NULL' : 'NONNULL')
. "\n";
} $schema->Procedure_Parameters($table);
print "\tComments:", $schema->Comments($table,'Procedure'), "\n";
} $schema->Procedures();
REPORTING BUGS
When reporting bugs/problems please include as much information as possible.
A small script which yields the problem will probably be of help. If you cannot include a small script then please include a Debug trace from a run of your program which does yield the problem.
AUTHOR INFORMATION
Brian H. Dunford-Shore brian@ibc.wustl.edu
Copyright 1998, Washington University School of Medicine, Institute for Biomedical Computing. All rights reserved.
This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.
Address bug reports and comments to: www@ibc.wustl.edu
TODO
These are features that would be nice to have and might even happen someday (especially if YOU write it).
SEE ALSO
RDBAL -- http://www.ibc.wustl.edu/perl5/other/RDBAL.html
Sybase::DBlib -- http://www.ibc.wustl.edu/perl5/other/sybperl.html
Win32::odbc -- http://www.ibc.wustl.edu/perl5/other/Win32/odbc.html
CREDITS
Thanks very much to:
David J. States (states@ibc.wustl.edu)
for suggestions and bug fixes.
BUGS
You really mean 'extra' features ;). None known.
COPYRIGHT
Copyright (c) 1997, 1998, 1999 Washington University, St. Louis, Missouri. All rights reserved. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.
1 POD Error
The following errors were encountered while parsing the POD:
- Around line 1867:
You forgot a '=back' before '=head2'