Re: new procedural language - PL/R - Mailing list pgsql-hackers

From cbbrowne@cbbrowne.com
Subject Re: new procedural language - PL/R
Date
Msg-id 20030203205612.4EC3C51DC4@cbbrowne.com
Whole thread Raw
In response to Re: new procedural language - PL/R  (Joe Conway <mail@joeconway.com>)
Responses Re: new procedural language - PL/R  (Joe Conway <mail@joeconway.com>)
List pgsql-hackers
> >>Any strong feelings on whether this is necessary for a first release?
> > 
> > No.  I'm not sure you'd really need triggers written in R ever ;-)
> 
> Yeah, that's what I figured too.

Indeed.  R sounds like it might be an interesting platform from which to
do "data mining," and in that sort of context, you're almost exclusively
reading data, so the loss of triggers would be of no great importance.

What might be "nifty" would be to have some mappings that did Clever
Transformations of Queries Into Views, particularly if that allowed
harnessing the DBMS to do some of the statistical analysis behind your
back...
--
If this was helpful, <http://svcs.affero.net/rm.php?r=cbbrowne> rate me
http://www3.sympatico.ca/cbbrowne/rdbms.html
"After you've  heard two  eyewitness accounts of  an auto  accident it
makes you wonder about history." -- Bits and Pieces


pgsql-hackers by date:

Previous
From: "Dann Corbit"
Date:
Subject: Win32 and fsync()
Next
From: Jan Wieck
Date:
Subject: Re: [mail] Re: Windows Build System