Re: REVIEW Range Types - Mailing list pgsql-hackers

From Jeff Davis
Subject Re: REVIEW Range Types
Date
Msg-id 1297230733.27157.401.camel@jdavis
Whole thread Raw
In response to Re: REVIEW Range Types  ("Erik Rijkers" <er@xs4all.nl>)
List pgsql-hackers
On Tue, 2011-02-08 at 20:43 +0100, Erik Rijkers wrote:
> --------------
> Documentation:
> --------------
> 
> Section 9.18:
>   table 9-42. range functions:
>   The following functions are missing (I encountered them in the regression tests):
>     contained_by()
>     range_eq()

I opted not to document functions that mostly exist to implement
operators. Some people might prefer the names, but then they don't
benefit from GiST index searches. So, I left this as-is for now.

> section 'Constructing Ranges' (8.16.6):
>   In the code example, remove the following line:
>     "-- the int4range result will appear in the canonical format"
>   it doesn't make sense there.  At this place "canonical format" has not been discussed;
> maybe it is not even discussed anywhere.

Thank you. I have added a forward reference.

> also (same place):
>        'where "_" is used to mean "exclusive" and "" is used to mean "inclusive".'
> should be:
>        'where "_" is used to mean "exclusive" and "i" is used to mean "inclusive".'

Thank you, fixed.

> And btw: it should mention here that the range*inf* functions,
> an underscore to separate 'range' from the rest of the function name, e.g.:
>    range_linfi_()  =>  infinite lower bound,  inclusive upper bound

I tried to clarify that section.

Thank you for the review! Updated patch forthcoming.

Regards,Jeff Davis



pgsql-hackers by date:

Previous
From: Jeff Davis
Date:
Subject: Range Type constructors
Next
From: Fujii Masao
Date:
Subject: Re: Named restore points