>>I thought we were just talking about some definitions.
I was expanding on your "#1" question, which was directly talking about "shared"
headers rather than just cleaning out HAVE_INT64_TIMESTAMP. I had the same
experience but also ran into the need for "shared" library code; which BTW ecpg
could benefit from as well.
>> performance penalty inside the backend
The idea requires reorganizing, not reimplementing. There shouldn't be a change
in performance in either direction.
--
Andrew Chernow
eSilo, LLC
every bit counts
http://www.esilo.com/