I learned SQL before the SQL-92 standard was in place. The database
engines I used (SQL Server and Oracle) have their own proprietary
methods for indicating forced inner outer joins. I haven't looked at
this code in over 10 years, so I'm a little unclear about exactly what
the result set returned was, and I don't have a database in place to
play with these as yet.
The SELECT I have is:
SELECT T1.COLUMN,
T2.COLUMN
FROM TABLE1 T1,
TABLE2 T2
WHERE T1.COLUMN *= T2.COLUMN
In SQL Server, the *= indicates a forced inner join which would cause
a record to be generated regardless if it existed in the T1 table or
not. The result set generated (if I remember correctly) would set the
T1.COLUMN to null if it did not appear in the T1 table and both
columns would be populated if the record existed in both tables.
Changing the *= to =* caused the reverse of this.
So my questions are how would this be expressed in SQL-92 (or more to
the point SQLite) syntax, and am I mistating the result set that would
be returned by this (there was another post in the mailing list on how
to determine whether a record existed in a primary table or not, and I
didn't want to spread bad advice).
Thanks in advance.
-----------------------------------------------------------------------------
To unsubscribe, send email to [EMAIL PROTECTED]
-----------------------------------------------------------------------------
