I'm storing facts about an entity (e.g., "company") in a normalized
form with various repeating groups, link tables, etc. My workflow
requires that after (or as part of) collecting these facts, I be able
to "sign off" as having verified all of the facts that pertain to a
given company. I understand this as meaning I need to sign off on each
row that was used in answering the query "select * from company left
join ..."

An inverted way of thinking about the problem is the notion of getting a
source document (say, a "company registration form") and parsing and
storing it in a normalized format. How do you "tag" the origin of each
and every row in every table that resulted from that source document?
It is possible to do so by associating an extra column with each
inserted or modified value (yuck).

It seems to me that the elegant way to do this would be to get the
entire graph of dependencies for not only all tables that reference the
entity, but only those rows within those tables that refer specifically
to my entity.

The query engine should have a pretty good idea about which tables and
which rows would actually be used in forming the responses to a given
query. How can I get this information? Or am I off the deep end (-ency
graph) with this one?



Randall Lucas Tercent, Inc. DF93EAD1

Search Discussions

Discussion Posts

Follow ups

Related Discussions

Discussion Navigation
viewthread | post
posts ‹ prev | 1 of 6 | next ›
Discussion Overview
grouppgsql-general @
postedAug 30, '06 at 11:23p
activeSep 5, '06 at 6:02p

2 users in discussion

Merlin Moncure: 3 posts Randall Lucas: 3 posts



site design / logo © 2022 Grokbase