Re: Range Types: empty ranges

From: "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
To: "Jan Wieck" <JanWieck(at)Yahoo(dot)com>
Cc: "Josh Berkus" <josh(at)agliodbs(dot)com>, "Jeff Davis" <pgsql(at)j-davis(dot)com>, <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Range Types: empty ranges
Date: 2011-02-14 16:16:26
Message-ID: 4D5900FA020000250003A97B@gw.wicourts.gov
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Jan Wieck <JanWieck(at)Yahoo(dot)com> wrote:

> Does ['15:15:00','15:15:00') make any more sense? Doesn't this
> essentially mean
>
> >= '15:15:00' && < '15:15:00'
>
> which again doesn't include a single point on the time line?

It defines a position in time with zero duration.

Some of the graphics programming I've done in the past was based on
a system where, at the pixel level, the coordinates referred to the
boundaries *between* the pixels. If you were to have a number of
horizontal bars, for example, the size of which you would adjust to
represent fluctuating data, you might have an x coordinate of 20 for
the left edge, and [20,30) would paint 10 pixels. I guess you
*could* destroy and recreate the object when the number dropped to
zero and came off it again; but the concept of [20,20) to draw zero
pixels but maintain the positional anchor can be convenient. I see
parallel concepts for some data domains in a database.

-Kevin

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Dimitri Fontaine 2011-02-14 16:17:48 Re: [HACKERS] "Extension" versus "module"
Previous Message Tom Lane 2011-02-14 16:08:54 Re: Extensions vs PGXS' MODULE_PATHNAME handling