Q: Features of 6.5 - Mailing list pgsql-hackers

From Hannu Krosing
Subject Q: Features of 6.5
Date
Msg-id 3741DD48.A3446B99@trust.ee
Whole thread Raw
List pgsql-hackers
As I have lately been swamped with work, I must confess note having 
read the mailing-list very constantly so some things may have slipped.

I have some questions about new/missing features in 6.5 that
I did not find a clear answer to in the changes list:

the list claims:

New SELECT FOR UPDATE 

does this just lock the table, or can I use it within a cursor to do 
UPDATE ... WHERE CURRENT also ?

Other questions 

1. What is the state of OUTER JOINS ?

2 Which constraints and constraint ops are (not) supported and how ?

2.1. Are FOREIGN KEYs supported
2.1.a - in the parser
2.1.b - in system tables
2.1.c - actually effective

2.2. Are constraints still 'write once - drop and recreate table if must
change', or is some of the ALTER TABLE .. ADD/REMOVE/DISABLE CONSRTRAINT ...
syntax  supported

3. Is there a way to get the source code for views,functions,rules, etc. Oracle does this by simply keeping the
originalcode (this enables one  to later recompile them as well if underlying tables change)
 

4. Are views still dumped as table+view ? I can read it but various reverse-engineering tools would like to see the
actualview  definition (could probably be solved by storing the code as well)
 

BTW, the code storing described in 3,4 could be done quite elegantly and
(unlike Oracle, which has different ways for different object types) 
consistently, by using the following table, 

CREATE TABLE pg_source_code (   objoid   oid,   rownr    int,   rowtext  text,   constraint primary key(objoid,rownr)
);



---------------
Hannu


pgsql-hackers by date:

Previous
From: Peter T Mount
Date:
Subject: I thought this was picked up ages ago?
Next
From: Tom Lane
Date:
Subject: Re: [HACKERS] I've got it, now should I commit it?