Re: Bug 1500 - Mailing list pgsql-hackers

From Steve Crawford
Subject Re: Bug 1500
Date
Msg-id 200503251253.54061.scrawford@pinpointresearch.com
Whole thread Raw
In response to Bug 1500  (Lyubomir Petrov <lpetrov@sysmaster.com>)
Responses Re: Bug 1500
Re: Bug 1500
Re: Bug 1500
List pgsql-hackers
> So this bug actually brings the issue of interval to_char()
> formatting. Opinions?

In digging around I discovered that it appears a decision was made to 
remove to_char(interval) at the 8.1 release but I've been unable to 
find the replacement for this functionality. This alarms me.

Given the messages I've seen regarding to_char(interval), it's clearly 
a function that is used. As an example, in our telephony systems 
there is a column for start_time and for end_time. Billing involves a 
sum(end_time-start_time) for the appropriate project/client/period. 
Naturally, that interval needs to be displayed appropriately.

The most common request I've seen (and it would be very helpful for me 
as well) is the ability to fill the largest displayed time increment 
with all remaining time in the interval.

In other words when the total increment is 7 days, 7 hours, 28 
minutes, 12 seconds the desired output would be 10528 minutes 12 
seconds. Think phone-billing, race times, mission clocks, etc.

So...

1) Is there really a plan to eliminate to_char(interval)?

2) If so, what is the replacement?

3) If there isn't a replacement and it's just scheduled for 
elimination, what harm was to_char(interval) causing to require its 
removal and what's the best way to lobby for its retention and 
improvement?

Cheers,
Steve



pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Bug 1500
Next
From: Palle Girgensohn
Date:
Subject: Re: Patch for collation using ICU