Re: bit strings - anyone working on them?
От | Philip Warner |
---|---|
Тема | Re: bit strings - anyone working on them? |
Дата | |
Msg-id | 5.1.0.14.0.20030423111934.0580d038@mail.rhyme.com.au обсуждение исходный текст |
Ответ на | Re: bit strings - anyone working on them? ("Christopher Kings-Lynne" <chriskl@familyhealth.com.au>) |
Ответы |
Re: bit strings - anyone working on them?
|
Список | pgsql-hackers |
At 09:15 AM 23/04/2003 +0800, Christopher Kings-Lynne wrote: >Do you right pad that 101? No way!!! You left pad it! See my earlier post in this thread questioning which was the low order bit. The current implementation is schizophrenic about this: it seems to treat the left-hand bits as low order sometimes, and the right-hand bits as low order when converting integer. It can't make up it's mind if it is a string or a number. If we treat left as low order, then: 1 | 10 => 10 Cast(8 as varbit(10)) => 0001000000 Cast(8 as varbit(32)) => 00010000000000000000000000000000 etc Which, on the whole, makes more sense. It's just hard to get around the notion of '10' being decimal 1, not 2. Also, I think substring is the only way to extract bits, and I would expect: Substring(Cast(8 as bit(X)), 4,1) should produce 1, which it does not. But if we adopt the left-bits-are-low philosophy, it works. Don't get me wrong, I don't mind if we adopt left-bits-are-high, then we just need to change the way substring & padding works. No idea which is better. ---------------------------------------------------------------- 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 |/
В списке pgsql-hackers по дате отправления: