>> An example from my code, which returns a database connection:
>>
>> import pyodbc
>>
>> def connect_to_db2():
>> """
>> quick and dirty connect to db2
>> """
>> s='DRIVER={iSeries access ODBC Driver};SYSTEM=10.3.36.150;UID=%s;PWD=%s'
>> return
>> connect_to_db(connect_str=s%('myuser','mypassword'),autocommit=True)
>>
>> db=connect_to_db2()
>> c=db.cursor()
>> #
>> # insert working code here
>> #
>> c.close()
>> db.close()
>>
>
> You will also want to define a system data source on the machine
> your running the above code on, instead of passing the driver
> details to the Windows ODBC Manager directly. This makes your
> code both more portable and creates fewer issues with Windows
> permission management.
I disagree with the "DSN is best" generalization. They both have their places,
and it is up to the developer to chose which is best for a given situation. In
the case of sample code being posted to a mail list, I think DSNs just confuse
things, as where a DSNless connection string makes it clear exactly what
options
are being set.
I will agree that for production code, DSNs have something to offer, but even
then there are cases where they aren't needed.
Carl K
_______________________________________________
DB-SIG maillist - DB-SIG@python.org
http://mail.python.org/mailman/listinfo/db-sig
