Re: beginning hackers (was: indexes spanning multiple - Mailing list pgsql-hackers

From Rod Taylor
Subject Re: beginning hackers (was: indexes spanning multiple
Date
Msg-id 1124750932.89252.23.camel@home
Whole thread Raw
In response to beginning hackers (was: indexes spanning multiple tables)  (Rosser Schwarz <rosser.schwarz@gmail.com>)
Responses Re: beginning hackers
Re: beginning hackers (was: indexes spanning multiple
List pgsql-hackers
On Mon, 2005-08-22 at 15:24 -0500, Rosser Schwarz wrote:
> while you weren't looking, Tom Lane wrote:
> 
> [indexes spanning multiple tables]
> 
> > Wouldn't recommend it as a project for a beginning backend hacker;
> > the locking considerations alone are a bit daunting.
> 
> That being the case, is there a list anywhere of open/wish list/TODO
> items that are suitable for beginning pg hackers?  I've been over the
> TODO list and found myself fairly daunted by what I see, but would
> still like to take a stab at contributing.

Utility commands (CREATE, ALTER, DROP) seem to be the easiest to deal
with since they are pretty much self contained:
     * Allow TRUNCATE ... CASCADE/RESTRICT     * Add a separate TRUNCATE permission     * Add COMMENT ON for all
clusterglobal objects (roles, databases       and tablespaces)     * Allow ALTER TABLE ... ALTER CONSTRAINT ... RENAME
  * Have ALTER TABLE RENAME rename SERIAL sequence names
 

Another source of items on the TODO list is the Unsupported Features
portion of the SQL Conformance documentation:

http://www.postgresql.org/docs/8.0/interactive/unsupported-features-sql-standard.html

Identity and Generator support or possibly simple Assertions shouldn't
have too significant of a learning curve to implement. The amount of
work may be large but you don't need to dig into the difficult to do
right bits of code like the optimizer.



pgsql-hackers by date:

Previous
From: Greg Stark
Date:
Subject: Re: enable_constraint_exclusion GUC name
Next
From: "Jim C. Nasby"
Date:
Subject: Re: distributed performance testing