A database schema conversion tool

dbmv.py helps you to move one database schema to another database if you have a 'live' connection available to both source and target databases. ODBC is used to both manage metadata about the schemas and to transfer data (if required), so Python-accessible ODBC drivers are required on the execution machine.

Source schemas can be from:

    ORACLE Mysql
    MICROSOFT SqlServer
    ORACLE
    POSTGRES
    SAP Sybase IQ (V15+)
    TERADATA

Schema description can be easily extended to any database in the destination list. You need to add missing description for "tbDefinition", etc, ... in file "dbmx.xml".

SQL output, field names, and SELECT descriptions must respect the pattern used for other existing databases.

Target schema can be:

    ACTIAN VectorH, Vector
    ACTIAN Ingres
    EMC Greenplum
    IBM Netezza (UDB, not sure about iseries and Zos)
    IBM Netezza (4.xx ?)
    ORACLE Mysql
    MICROSOFT SqlServer
    ORACLE
    POSTGRES
    PROGRESS Database
    SAP Sybase IQ (V15+)
    SAP Sybase ASA (Anywhere)
    SAP Sybase ASE
    SAP Hana
    SAP MaxDB
    TERADATA

Without parameters dbmv.py diplays a simple help about parameters

$ ./dbmv.py

Usage : ['src=', 'dest=', 'loadata', 'cretab', 'creall', 'loaddl', 'parfile=', 'fdelim=', 'unload', 'translation=', 'quote=', 'cmdsep=']

• src : Source database (--src=URL)

• dest : Destination database (--src=URL)

  URL : db_driver[-odbc]://db_host[:port]/dbname?db_login&login_password

  db_driver list:

    ACTIAN VectorH, Vector -> vectorwise
    ACTIAN Ingres          -> ingres
    EMC Greenplum          -> greenplum
    IBM Netezza            -> netezza
    ORACLE Mysql           -> mysql
    MICROSOFT SqlServer    -> mssql
    ORACLE                 -> oracle
    POSTGRES               -> postgres
    PROGRESS Database      -> progress
    SAP Sybase IQ (V15+)   -> iq
    SAP Sybase ASA         -> asa
    SAP Sybase ASE         -> ase
    SAP Hana               -> hana
    SAP MaxDB              -> maxdb
    TERADATA               -> teradata
  

  About ODBC : Linux database drivers path and configuration must be added in file

        "../etc/driverTools.xml" (see in-file examples)
  

  On Window systems, ODBC drivers and Datasource must exist. No additional configuration is required.

• cretab : Switch to Create tables only

• creall : Switch to all objects TABLES, INDEXES, PRIMARY KEYS, REFERENTIAL CONSTRAINTS

• loaddl : Load ddl into destination database. If not specified, a file containing all objects is created in the current directory. In this latest case, you can reload.

• parfile: All parameters can be stored is a parfile. Parfiles examples can be find in the "wrk" directory.

• translation : Schema-name translation pairs for tables, constraints, indexes

  Example: --translation=scname:public,vw_user,postgres,vw_user; iscname:public,vw_user,postgres,vw_user; rscname:public,vw_user,postgres,vw_user

  • scname : Tables owned by "public" are moved to schema "vw_user", and from "postgres" to "vw_user"

  • iscname: Indexes owner by "public" are moved to schema "vw_user" and from "postgres" to "vw_user"

  • rscname: Constraints owner by "public" are moved to schema "vw_user" and from "postgres" to "vw_user"

• quote : Quote DDL (Ex: --quote='"' => CREATE TABLE "my_table")

• cmdsep: Command separator (default=';') used in DDL file (Ex: --cmdsep='\g' or --cmdsep='go')

• unload/loadata : Very basic unload/load facilities for data. Absolutly unefficient for massive data but can be useful for very small tables (< 1000 records)

• fdelim : Field delimiter used for data unloads.

Supported Python drivers are:

pyodbc # Module for ODBC

or:

   ingresdbi                      # Module for Ingres transactional database
   cx_Oracle                      # Module for Oracle
   Sybase                         # Module for Sybase ASE
   pymssql                        # Module for MsSql
   MySQLdb                        # Module for Mysql
   psycopg2, psycopg2.extensions  # Postgres module
   import DB2                     # Module for DB2