From: | Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> |
---|---|
To: | Robert Haas <robertmhaas(at)gmail(dot)com> |
Cc: | Nick Rudnick <joerg(dot)rudnick(at)t-online(dot)de>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: [pgsql-general 2011-1-21:] Are there any projects interested in object functionality? (+ rule bases) |
Date: | 2011-02-01 04:41:25 |
Message-ID: | AANLkTimAgSNx3wtkvgeB9=drXYJOtJHRpGyjuCB50PrV@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
2011/2/1 Robert Haas <robertmhaas(at)gmail(dot)com>:
> On Mon, Jan 31, 2011 at 5:09 PM, Nick Rudnick <joerg(dot)rudnick(at)t-online(dot)de> wrote:
>> Interesting... I remember that some years ago, I fiddled around with
>> functions, operators etc. to allow a method like syntax -- but I ever was
>> worried this approach would have serious weaknesses -- are there any
>> principal hindrances to having methods, if no, can this be implemented in a
>> straightforward way?
>
> It would help if you were a bit more specific. Do you mean you want
> to write something like foo.bar(baz) and have that mean call the bar
> method of foo and pass it baz as an argument?
>
> If so, that'd certainly be possible to implement for purposes of a
> college course, if you're so inclined - after all it's free software -
> but we'd probably not make such a change to core PG, because right now
> that would mean call the function bar in schema baz and pass it foo as
> an argument. We try not to break people's code to when adding
> nonstandard features.
>
I has not a standard, so I am not sure what is in standard and what
not. It was a popular theme about year 2000 and OOP was planed to
SQL3. You can find a some presentation from this time. Oracle
implemented these features.
J. Melton: SQL:1999: Understanding Object-Relational and
Other Advanced Features, Morgan Kaufmann, 2003.
CREATE METHOD next_color (n INT)
RETURNS INT
FOR colored_part_t
RETURN SELF.color_id + n
SELECT partno, color_id, DEREF(oid).next_color(1) AS next
FROM colored_parts
some other databases implemented a dereferenced data (it's not only
Oracle's subject)
http://www.java2s.com/Code/Oracle/Object-Oriented-Database/DEREFDereferencetheRowAddresses.htm
Probably DB2 implements this functionality too. See doc for CREATE
TYPE statement, REF USING, NOT FINAL, method specification
CREATE TYPE type-name
...
METHOD attribute-name()
RETURNS attribute-type
these features are very nice - but is not well documented and probably not used.
Pavel
> --
> Robert Haas
> EnterpriseDB: http://www.enterprisedb.com
> The Enterprise PostgreSQL Company
>
From | Date | Subject | |
---|---|---|---|
Next Message | Pavel Stehule | 2011-02-01 04:43:59 | Re: [pgsql-general 2011-1-21:] Are there any projects interested in object functionality? (+ rule bases) |
Previous Message | Hitoshi Harada | 2011-02-01 04:08:16 | Re: Add ENCODING option to COPY |