From: | Chris Bitmead <chris(at)bitmead(dot)com> |
---|---|
To: | Oliver Elphick <olly(at)lfix(dot)co(dot)uk> |
Cc: | pgsql-hackers(at)postgreSQL(dot)org |
Subject: | Re: [HACKERS] Inheritance, referential integrity and other constraints |
Date: | 2000-01-28 00:10:39 |
Message-ID: | 3890DE7F.5409C776@bitmead.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Oliver Elphick wrote:
> No, the inheritance system doesn't allow them to be different types.
> You get an error if you try to create such a table:
Hmm. While it might allow it, I can't see the logic in it. Can't think
of any OO language that thinks this way. All other languages you get
two different variables either with :: scope resolution in C++ or
renaming in Eiffel.
> Because the column names are identical, they are overlaid and treated
> as the same column. This is so whether or not they ultimately derive
> from the same parent, so it isn't strictly a case of repeated inheritance
> as in Eiffel. (There, repeatedly inherited features of the same parent
> are silently combined, but identical names from unrelated classes are
> conflicts.)
Which seems like the right thing to me.
From | Date | Subject | |
---|---|---|---|
Next Message | The Hermit Hacker | 2000-01-28 00:51:09 | PostgreSQL 7.0beta Freeze Postponed ... (fwd) |
Previous Message | Chris Bitmead | 2000-01-28 00:08:00 | Re: OIDS (Re: [HACKERS] Well, then you keep your darn columns) |