Andy Duncan > Orac-alpha-1.2.6 > DDL::Oracle

Download:
Orac-alpha-1.2.6.tar.gz

Dependencies

Annotate this POD

Related Modules

DBD::Oracle
more...
By perlmonks.org

CPAN RT

Open  1
View/Report Bugs
Module Version: 1.07   Source  

NAME ^

DDL::Oracle - a DDL generator for Oracle databases

VERSION ^

VERSION = 1.07

SYNOPSIS ^

 use DBI;
 use DDL::Oracle;

 my $dbh = DBI->connect(
                         "dbi:Oracle:dbname",
                         "username",
                         "password",
                         {
                          PrintError => 0,
                          RaiseError => 1
                         }
     );

 # Use default resize and schema options.
 # query default DBA_xxx tables (could use USER_xxx for non-DBA types)
 DDL::Oracle->configure( 
                         dbh    => $dbh,
                       );

 # Create a list of one or more objects
 my $sth = $dbh->prepare(
        "SELECT
                owner
              , table_name
         FROM
                dba_tables
         WHERE
                tablespace_name = 'MY_TBLSP'    -- your mileage may vary
        "
     );

 $sth->execute;
 my $list = $sth->fetchall_arrayref;

 my $obj = DDL::Oracle->new(
                             type  => 'table',
                             list  => $list,                          );
                           );

 my $ddl = $obj->create;      # or $obj->resize;  or $obj->drop;  etc.

 print $ddl;    # Use STDOUT so user can redirect to desired file.

DESCRIPTION ^

Overview

Designed for Oracle DBA's and users. It reverse engineers database objects (tables, indexes, users, profiles, tablespaces, roles, constraints, etc.). It generates DDL to *resize* tables and indexes to the provided standard or to a user defined standard.

We originally wrote a script to defrag tablespaces, but as DBA's we regularly find a need for the DDL of a single object or a list of objects (such as all of the indexes for a certain table). So we took all of the DDL statement creation logic out of defrag.pl, and put it into the general purpose DDL::Oracle module, then expanded that to include tablespaces, users, roles, and all other dictionary objects.

Oracle tablespaces tend to become fragmented (now THAT's an understatement). Even when object sizing standards are adopted, it is difficult to get 100% compliance from users. And even if you get a high degree of compliance, objects turn out to be a different size than originally thought/planned -- small tables grow to become large (i.e., hundreds of extents), what was thought would be a large table ends up having only a few rows, etc. So the main driver for DDL::Oracle was the object management needs of Oracle DBA's. The "resize" method generates DDL for a list of tables or indexes. For partitioned objects, the "appropriate" size of EACH partition is calculated and supplied in the generated DDL. The original defrag.pl will be rewritten to use DDL::Oracle, and supplied with its distribution.

Initialization and Constructor

configure

The configure method is used to supply the DBI connection and to set several session level attributes. These are:

      dbh      A reference to a valid DBI connection (obtained via
               DBI->connect).  This is the only mandatory attribute.

               NOTE: The user connecting should have SELECT privileges
                     on the following views (in addition to the DBA or
                     USER views), but see attributes 'heading' for
                     exceptions:

                         V$DATABASE

                     And, in order to generate CREATE SNAPSHOT LOG
                     statements, you will also need to create a PUBLIC
                     SYNONYM for DBA_SNAPSHOT_LOG_FILTER_COLS.  In
                     order for non-DBA users to do the same, you will
                     need to grant SELECT on this view to them (e.g.,
                     to PUBLIC).  Why Oracle Corp. feels this view is
                     of no interest to non-replication users is a
                     mystery to the author.

                     And, in order to generate CREATE INDEX statements
                     for indexes which have DESCending column(s) and/or
                     include FUNCTION based column(s), you must have
                     select privileges on SYS.COL$, wherein the real
                     name of the column or function definition is held.

      schema   Defines whether and what to use as the schema for DDL
               on objects which use this syntax.  "1" means use the
               owner of the object as the schema; "0" or "" means
               omit the schema syntax; any other arbtrary string will
               be imbedded in the DDL as the schema.  The default is "1".

      resize   Defines whether and what to use in resizing segments.
               "1" means resize segments using the default algorithm;
               "0" or "" means keep the current INITIAL and NEXT
               values; any other string will be interpreted as a
               resize definition.  The default is "1".

               To establish a user defined algorithm, define this with
               a string consisting of n sets of LIMIT:INITIAL:NEXT.
               LIMIT is expressed in Database Blocks.  The highest LIMIT
               may contain the string 'UNLIMITED', and in any event will
               be forced to be so by DDL::Oracle.

      view     Defines which Dictionary views to query:  DBA or USER
               (e.g., DBA_TABLES or USER_TABLES).  The default is DBA.

      heading  Defines whether to include a Heading having Host, Instance,
               Date/Time, List of generated Objects, etc.  "1" means 
               include the heading; "0" or "" means to suppress the
               heading (and eliminate the query against V$DATABASE).
               The default is "1".

      prompt   Defines whether to include a PROMPT statement along
               with the DDL.  If the output is intended for use in
               SQL*Plus, this will cause SQL*Plus to display a comment
               about each statement before it executes, which can be
               helpful in a multi-statement file.  "1" means include
               the prompt; "0" or "" means to suppress the prompt.

new

The new method is the object constructor. The two mandatory object definitions are supplied with this method, to wit:

      type    The type of object (e.g., TABLE, INDEX, SYNONYM, family,
              etc.).

              For 'table family', supply the name(s) of tables -- the
              DDL will include the table and its:
                  Comments (Table and Column)
                  Indexes
                  Constraints
                  Triggers

      list    An arrayref to an array of arrayrefs (as in the DBI's 
             "fetchall_arrayref" method) containing pairs of owner and
              name.

Object methods

create

The create method generates the DDL to create the list of Oracle objects.

drop

The drop method generates the DDL to drop the list of Oracle objects.

resize

The resize method generates the DDL to resize the list of Oracle objects. The 'type' defined in the 'new' method is limited to 'index' and 'table'. For tables, this generates an ALTER TABLE MOVE statement; for indexes, it generates an ALTER INDEX REBUILD statement. If the table or index is partitioned, then a statement for each partition is generated.

To generate DDL for a single partition of an index or table, define the 'name' as a colon delimited field (e.g., 'name:partition').

compile

The compile method generates the DDL to compile the list of Oracle objects. The 'type' defined in the 'new' method is limited to 'function', 'package', 'procedure', 'trigger' and 'view'.

show_space

The show_space method produces a report showing used/unused bytes and blocks above/below the high water mark in a segment. It includes the free blocks below the high water mark. For partitioned objects, it shows the information for each partition, with grand totals for the table/index. The object does NOT need to be analyzed for this report to be accurate -- it uses package sys.DBMS_SPACE to collect the data.

BUGS ^

FILES ^

 copy_user.pl
 copy_user.sh
 ddl.pl
 defrag.pl
 query.pl

AUTHOR ^

 Richard V. Sutherland
 rvsutherland@yahoo.com

COPYRIGHT ^

Copyright (c) 2000, 2001 Richard V. Sutherland. All rights reserved. This module is free software. It may be used, redistributed, and/or modified under the same terms as Perl itself. See:

    http://www.perl.com/perl/misc/Artistic.html
syntax highlighting: