The IBM_DB_SA adapter provides the Python/SQLAlchemy interface to IBM DB2.

The IBM_DB_SA adapter is packaged as a Python Egg component and is dependent on:

• SQLAlchemy 0.7.3 and above

We are assuming that you have DB2 or Informix 11.10 and SQLAlchemy already installed.

If you need to connect from SQLAlchemy to a DB2 server on the same local machine, just proceed to the Installation section below. If you need to connect from SQLAlchemy to either a DB2 server on different machine or to Informix (local or remote) then you will need to at minimum install the IBM Data Server Driver Package (DS Driver) on the machine where you installed SQLAlchemy.

To install the IBM_DB_SA Python egg component (.egg), use the standard setuptools provided by the Python Easy Install, which is available through the Python Enterprise Application Kit community portal.

After installing Easy Install, perform these additional steps to install the IBM_DB_SA egg component:

• To install the IBM_DB_SA egg component available in the remote repository:

Windows:

 > easy_install ibm_db_sa

Linux or UNIX:

 $ sudo easy_install ibm_db_sa

• To install the IBM_DB_SA egg component from the downloaded .egg file:

Windows:

 > easy_install ibm_db_sa-x.x.x-py2.x.egg

Linux or UNIX:

 $ sudo easy_install ibm_db_sa-x.x.x-py2.x.egg

• To uninstall the IBM_DB_SA egg component:

Windows:

 > easy_install -m ibm_db_sa==x.x.x
 > rmdir Python25\Lib\site-packages\ibm_db_sa-x.x.x-py2.x.egg

Linux or UNIX:

 $ sudo easy_install -m ibm_db_sa==x.x.x   
 $ sudo rm -rf /usr/lib/python2.5/site-packages/ibm_db_sa-x.x.x-py2.x.egg

Before you open a Python prompt, you need to ensure that IBM CLI (which the Python driver uses to connect to DB2/Informix) is accessible to SQLAlchemy. On Linux, set the LD_LIBRARY_PATH variable (for the user executing Python) to include the folder where the IBM CLI shared library (libdb2.so) resides

For local DB2 access: export LD_LIBRARY_PATH=<DB2_HOME>/sqllib/lib:$LD_LIBRARY_PATH .For remote DB2 or any Informix access with DS Driver: export LD_LIBRARY_PATH=<DS_DRIVER_FOLDER>/odbc_cli_driver/linux/clidriver/lib:$LD_LIBRARY_PATH

$ python
Python 2.5.1 (r251:54863, Oct  5 2007, 13:36:32)
[GCC 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import sqlalchemy
>>> from sqlalchemy import *
>>> db2 = sqlalchemy.create_engine('ibm_db_sa://db2inst1:secret@host.name.com:50000/pydev')
>>> metadata = MetaData()
>>> users = Table('users', metadata, 
    Column('user_id', Integer, primary_key = True),
    Column('user_name', String(16), nullable = False),
    Column('email_address', String(60), key='email'),
    Column('password', String(20), nullable = False)
)
>>> metadata.bind = db2
>>> metadata.create_all()
>>> users_table = Table('users', metadata, autoload=True, autoload_with=db2)
>>> users_table

$ python
Python 2.5.1 (r251:54863, Oct  5 2007, 13:36:32)
[GCC 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import ibm_db
>>> ibm_db_conn = ibm_db.connect('pydev', 'db2inst1', 'secret')
>>> import ibm_db_dbi
>>> conn = ibm_db_dbi.Connection(ibm_db_conn)
>>> conn.tables('SYSCAT', '%')

$ cat ./clidriver/db2cli.ini
...
[pydb]
Database=pydev
Protocol=tcpip
Hostname=host.name.com
Servicename=50000
uid=db2inst1
pwd=secret

$ python
Python 2.5.1 (r251:54863, Oct  5 2007, 13:36:32)
[GCC 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import sqlalchemy
>>> from sqlalchemy import *
>>> db2 = sqlalchemy.create_engine('ibm_db_sa://db2inst1:secret@/pydb')
>>> ...

• SuSE Linux Server 9 32-bit
• Ubuntu Linux 7.04 32-bit
• MS Windows 32-bit

• Minimum Supported version of IBM DB2 is V9fp2 for Linux, UNIX, and Windows
• IBM DB2 Version 9.5 for Linux, UNIX, and Windows
• IBM DB2 Version 9.1 for Linux, UNIX, and Windows

Your feedback is very much appreciated and expected through project ibm-db:

• ibm-db project: http://code.google.com/p/ibm-db/
• ibm-db wiki: http://code.google.com/p/ibm-db/w/list
• ibm-db issues reports: http://code.google.com/p/ibm-db/issues/list
• ibm-db developers: opendev@us.ibm.com