Re: Views, views, views! (long) - Mailing list pgsql-hackers

From Josh Berkus
Subject Re: Views, views, views! (long)
Date
Msg-id 200505050911.14080.josh@agliodbs.com
Whole thread Raw
In response to Re: Views, views, views! (long)  (Peter Eisentraut <peter_e@gmx.net>)
Responses Re: Views, views, views! (long)  (Andreas Pflug <pgadmin@pse-consulting.de>)
Re: Views, views, views! (long)  (Peter Eisentraut <peter_e@gmx.net>)
Re: Views, views, views! (long)  ("Greg Sabino Mullane" <greg@turnstep.com>)
List pgsql-hackers
Tom, Peter,

> That said, I don't particularly care for this proposal.  If you want a
> human-readable version of the system catalogs, I suggest you work on
> extensions of the information schema, not a completely new interface.

So, both of your would prefer that we break the SQL spec with the information 
schema?  In order to cover all PG objects?  Because that's what your 
proposing.  Either the information schema adheres to the spec, or it only 
covers 25% of PostgreSQL objects.   There isn't a 3rd alternative.  I'm fine 
with merging this with the information_schema (some of these views are 
derived from the same code) but it's either/or.

> So?  If you want reality, look at the catalogs.

The system catalogs are NOT user-friendly, nor are they meant to be.  The 
purpose of the new system views is to answer questions like, "what objects 
does user "gregory" have permissions on?"  and "do any of my fuctions use 
custom type 'joebert' before I change it?"  and "I need a way to query all of 
my functions in a loop so that I can change their permissions."  You can get 
that info from the system catalogs, but only if you're a SQL wizard and know 
them very well.

These are all things that users (NOT pg hackers) have to do for applications 
daily, and that we currently don't provide any easy, comprehensible way to 
access.  It's certainly easy for pg hackers to say, "oh, use the system 
tables" but those tables are baffling and awkward for the many thousands of 
users who are not PG hackers.   Elein's series on General Bits covering 
queries which were the inspiration for many of the views was immensely 
popular.

Frankly, this is sounding a lot like "Who needs OpenOffice.org?  Use vi!"

Further, Jim and I went over the purpose of the views on this list two months 
ago and even Tom seemed positive to the idea.  What the hell changed?   No 
wonder the discussion of the decision process dragged on so.   If we 
introduce a proposal, do an implementation, and come back with it 2 months 
later, I don't expect a bunch of core members to act like they've never heard 
of it before.  

If there are things (like not dropping columns) that you want to change about 
the spec, fine.   But if you think that nobody needs these views, it's 
because you haven't had much contact with end users lately.

-- 
Josh Berkus
Aglio Database Solutions
San Francisco


pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: [pgsql-advocacy] Increased company involvement
Next
From: "Marc G. Fournier"
Date:
Subject: Re: [pgsql-advocacy] Increased company involvement