Reviewing patch "URI connection string support for libpq"

From: Harold Giménez <harold(dot)gimenez(at)gmail(dot)com>
To: pgsql-hackers(at)postgresql(dot)org
Subject: Reviewing patch "URI connection string support for libpq"
Date: 2012-02-24 00:38:13
Message-ID: CABQCq-QXag7K3GpJNron2Uy+ZK5LJzEjVXNC3qosretewiTKAw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hello hackers,

I've been a reader of this list for some time, but have never posted.

I have interest in the URI connection string support patch[1], so I'm in
the process of reviewing it. I have a couple of comments and questions:

1. I see no tests in the patch. I'd like to start getting together a set of
tests, likely based on the connection string permutations found on Greg
Smith's response[2]. However I don't find an obvious place to put them.
They could possibly live in the test/examples directory. Another thought is
to use dblink in a test, although it may be problematic to depend on a
contrib package for a test, to say the least. Any thoughts on how to test
this are most welcome.
2. The documentation/manual was not updated as part of this patch, so this
is pending.
3. I for one do prefer the `postgres` prefix, as opposed to `postgresql`
for the reasons stated on an earlier thread [3]. In my opinion the best way
to move forward is to support them both.

The good news is the patch still applies fine on the 9.2 HEAD, and seems to
work well.

Regards,

-Harold

[1] https://commitfest.postgresql.org/action/patch_view?id=720
[2]
http://archives.postgresql.org/message-id/4F45253C.4030409@2ndQuadrant.com
[3] http://archives.postgresql.org/pgsql-hackers/2011-12/msg00499.php

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message David E. Wheeler 2012-02-24 01:35:09 row_to_json() Bug
Previous Message Tom Lane 2012-02-24 00:15:18 Re: psql \i tab completion initialization problem on HEAD