Re: bit strings - anyone working on them? - Mailing list pgsql-hackers

From Philip Warner
Subject Re: bit strings - anyone working on them?
Date
Msg-id 5.1.0.14.0.20030424022221.044b4e38@mail.rhyme.com.au
Whole thread Raw
In response to Re: bit strings - anyone working on them?  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: bit strings - anyone working on them?  (Hannu Krosing <hannu@tm.ee>)
List pgsql-hackers
At 12:02 PM 23/04/2003 -0400, Tom Lane wrote:
>and instead
>add a separate function that maps an integer to an LSB-first bitstring?

This would work, but it's not going to be very friendly without automatic 
cast support. Does the spec mention LSB and/or numeric conversions? Also, I 
just noticed that B'0' is not equal to B'00', which also seems unfortunate.

Perhaps we could also add an LSB-last function, and output some kind of 
warning when using numeric casts to the effect that the application should 
use the LSB-last function because the casts will start using LSB-first in a 
future version.

Alternatively, we could consider creating a new type that works as one 
would expect:
 - Automatic casting (sizes and numeric) - Automatic padding - numeric & text conversion - extract_bits using LSB as
position1.
 

and one of: - LSB on right - pad on left
or - LSB on left - pad on right

Another option: can we use some aspect of the type definition to specify 
the LSB position to store in attyypmod?



----------------------------------------------------------------
Philip Warner                    |     __---_____
Albatross Consulting Pty. Ltd.   |----/       -  \
(A.B.N. 75 008 659 498)          |          /(@)   ______---_
Tel: (+61) 0500 83 82 81         |                 _________  \
Fax: (+61) 03 5330 3172          |                 ___________ |
Http://www.rhyme.com.au          |                /           \|                                 |    --________--
PGP key available upon request,  |  /
and from pgp5.ai.mit.edu:11371   |/



pgsql-hackers by date:

Previous
From: "Hiroshi Saito"
Date:
Subject: Re: linking problem with gcc-mingw
Next
From: Peter Eisentraut
Date:
Subject: Re: bit strings - anyone working on them?