Re: Another idea for dealing with cmin/cmax

Martijn van Oosterhout wrote:
> On Fri, Sep 29, 2006 at 10:59:13AM +0100, Heikki Linnakangas wrote:
>> Martijn van Oosterhout wrote:
>>> On Fri, Sep 29, 2006 at 09:35:31AM +0100, Heikki Linnakangas wrote:
>>>> We could get rid of t_hoff, because we should always be able to
>>>> calculate the header size. Then we're down to 18 bytes.
>>> Without t_hoff, how do you know the size of the null bitmap? You could
>>> probably do it only if you assume the null bitmap, if present, always
>>> covers all the columns...
>> I think we assume that already. heap_form_tuple reserves space for the
>> bitmap like this:
>>
>> if (hasnull)
>> len += BITMAPLEN(numberOfAttributes);
>
> Ok, now we do an ALTER TABLE blah ADD COLUMN ..., and we have to expand
> the bitmaps for the entire table?
No, you'd still have the the number of attributes (t_natts) in the header.

--
Heikki Linnakangas
EnterpriseDB http://www.enterprisedb.com