Re: Network Flow Schema + Bulk Import/Updates - Mailing list pgsql-general

From Michael L. Artz
Subject Re: Network Flow Schema + Bulk Import/Updates
Date
Msg-id 4331F4C6.7070609@october29.net
Whole thread Raw
In response to Re: Network Flow Schema + Bulk Import/Updates  (Tony Wasson <ajwasson@gmail.com>)
Responses Re: Network Flow Schema + Bulk Import/Updates
List pgsql-general
Tony Wasson wrote:

>You can use the merge trigger below to do this. You'll need to add
>some code to update the count. You may also benefit from using the new
>constraint exclusion (table partitioning) in PostgreSQL 8.1. I am not
>sure if CE works against the inet datatype -- if not, try converting
>the IP to an integer.
>
>

CE looked like it was just for parent/child relationships ... did I read
that right?  I'm not sure how it applies.  And the table partitioning
looks like its still on the todo list ... is that really the case?

And as for getting data into the DB ... from earlier posts it sounded
like standard practice was to bulk load the new data into a temporary
table and then do an INSERT ... SELECT to load the data into the new
table.  Is this still the case with the trigger, or can/should I just
COPY the data straight into the final database?  And I assume that I
should *not* delete my indexes while I'm loading the table, since the
queries in the trigger can take advantage of them ... right?

Also, as a slight aside, has anyone created a data type for single IPs
that is essentially an integer (i.e. 4 bytes) that works with the
standard functions for INET?

-Mike

pgsql-general by date:

Previous
From: vishal saberwal
Date:
Subject: Re: Problem with libpq3 & postgresql8
Next
From: Ismael Perdido
Date:
Subject: Re: JDBC error w/ WebObjects