Re: Plan for compressed varlena headers

Gregory Stark wrote:
> 2) Replace VARATT* macros to store and retrieve the toast bits in a manner
> that will work for variable length headers. This either means storing the
> bits at the least-significant position or using network byte order.
>
> If we want to allow storing >1 headers unaligned which I think would be
> good then I still think we have to read them using bytewise lookups -- ie
> by casting to (char*). That means network byte order or using the low order
> bits is equally efficient.

I think the plan was to have the macro code conditional on big-little
endian. We can deal with doing >1 headers unaligned at some future
date if we feel we need it, and the macros will make it transparent.

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://www.enterprisedb.com

+ If your life is a hard drive, Christ can be your backup. +