SQL Question

From: <operationsengineer1(at)yahoo(dot)com>
To: "pgsql-novice(at)postgresql(dot)org" <pgsql-novice(at)postgresql(dot)org>
Subject: SQL Question
Date: 2006-05-24 22:06:53
Message-ID: 20060524220653.76449.qmail@web33307.mail.mud.yahoo.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-novice

i'm setting up a system that will record passes or
fails for a given inspection.

i have created a t_inspect_result table with three
columns (to keep it simple) - inspect_id (fk to
t_inspect.inspect_id), inspect_pass (boolean),
inspect_date.

to keep it simple, let's say i have qa (quality
assurance) and ft (functional test) inspections.

i need to generate a report that will let me know if a
given serial number has any inspections that have not
yet had a pass (still a failed unit for that test node
- qa or ft).

i must add that a unit can go through qa twice... for
example

qa test 1
fail
fail
pass

ft 1
pass

... more work done ...

qa test 2
pass

given this structure, and hopefully i've communicated
it well enough, i'm concerned that it might be
difficult to weed out qa 1 fail followed by qa 2 pass.

that is...

qa test 1
fail

... the bizrre happens ...

qa test 2
pass

how can i go about finding the qa test 1 failure?

if i did a inspect_date descending query with a limit
of 1, i'd get a pass and miss the failure.

i think it has to do with "group," but i'm not mind
gripping the concept.

__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

Responses

Browse pgsql-novice by date

  From Date Subject
Next Message Bruno Wolff III 2006-05-25 03:30:40 Re: SQL Question
Previous Message Brad Nicholson 2006-05-24 21:10:45 Re: PostgreSQL Update + (PGISDBRU) PGSQL is Slow Debunked