Skip site navigation (1) Skip section navigation (2)

Re: SSI patch version 14

From: Markus Wanner <markus(at)bluegap(dot)ch>
To: Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
Cc: Simon Riggs <simon(at)2ndQuadrant(dot)com>, Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: SSI patch version 14
Date: 2011-02-02 12:48:54
Message-ID: (view raw, whole thread or download thread mbox)
Lists: pgsql-hackers

On 02/02/2011 11:20 AM, Heikki Linnakangas wrote:
> I couldn't get it to run on my basic Debian system, clearly I'm
> doing something wrong, but it will be even harder for people on more
> exotic platforms.

On Debian you only need 'python-twisted' and the dtester sources [1],
IIRC.  What issue did you run into?

> The tests don't seem very complicated, and they don't seem to depend
> much on the dtester features. How hard would it be to rewrite the test
> engine in C or perl?

I've taken a quick look at a perl framework similar to twisted (POE).
Doesn't seem too complicated, but I don't see the benefit of requiring
one over the other.  Another async, event-based framework that I've been
playing with is asio (C++).

I don't think it's feasible to write anything similar without basing on
such a framework.  (And twisted seems more mature and widespread (in
terms of supported platforms) than POE, but that's probably just my
subjective impression).

However, the question here probably is whether or not you need all of
the bells and whistles of dtester (as if there were that many) [2].

> I'm thinking of defining each test in a simple text file, and write a
> small test engine to parse that.

AFAIUI there are lots of possible permutations, so you don't want to
have to edit files for each manually (several hundreds, IIRC).  So using
a scripting language to generate the permutations makes sense, IMO.

Pretty much everything else that Kevin currently uses dtester for (i.e.
initdb, running the postmaster, connecting with psql) is covered by the
existing regression testing infrastructure already, I think.  So it
might be a matter of integrating the permutation generation and test
running code into the existing infrastructure.  Kevin?


Markus Wanner

[1]: dtester project site:

[2]: side note:
Dtester mainly serves me to test Postgres-R, where the current
regression testing infrastructure simply doesn't suffice.  With dtester
I tried to better structure the processes and services, their
dependencies and the test environment.

In response to


pgsql-hackers by date

Next:From: Heikki LinnakangasDate: 2011-02-02 12:56:55
Subject: Re: SSI patch version 14
Previous:From: Dimitri FontaineDate: 2011-02-02 11:29:10

Privacy Policy | About PostgreSQL
Copyright © 1996-2018 The PostgreSQL Global Development Group