User: Password:
Subscribe / Log in / New account

PostgreSQL and the SQL standards process

PostgreSQL and the SQL standards process

Posted Sep 22, 2011 4:00 UTC (Thu) by BenHutchings (subscriber, #37955)
Parent article: PostgreSQL and the SQL standards process

The irony of Microsoft and Oracle using SQL standardisation as a weapon is that neither of them are seem to be in much of a hurry to follow the standards themselves. When I last had the misfortune to work on a data access layer (about 2005), SQL Server was still using SQL-89 semantics in some cases where SQL-92 differed, and Oracle still considered that '' IS NULL. PostgreSQL seemed to be the most standard-conformant of the lot. Has this changed?

(Log in to post comments)

PostgreSQL and the SQL standards process

Posted Sep 24, 2011 20:38 UTC (Sat) by butlerm (guest, #13312) [Link]

>Oracle still considered that '' IS NULL

Oracle still does, and it probably won't ever change because it would break far too much software if they did. In Oracle's defense, most languages have extremely poor or non-existent support for nullable character strings, and so this decision tends to make working with Oracle easier in some respects.

In particular, you really don't want to index the "no relationship" values of an optional foreign key. Like an empty email address, for example. To accomplish this in most databases, you have to convert email addresses that are empty strings to nulls when inserting data into the database, and then convert nulls back to empty strings when you are done.

Personally, I wouldn't even like to use a database that distinguished between nulls and empty strings unless there was a convenient facility to do just that.

Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds