Re: [HACKERS] Proposed change to the JDBC driver

From: Peter Mount <peter(at)retep(dot)org(dot)uk>
To: "PostgreSQL Interfaces (E-mail)" <pgsql-hackers(at)postgresql(dot)org>
Cc: "PostgreSQL Developers List (E-mail)" <hackers(at)postgresql(dot)org>
Subject: Re: [HACKERS] Proposed change to the JDBC driver
Date: 2000-01-24 17:33:58
Message-ID: Pine.LNX.4.10.10001241731400.4761-100000@maidast.retep.org.uk
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


As I haven't seen any replies to this, can I assume no body objects to me
making this fairly major change to the driver?

I need to know sometime in the next few hours, as I'm about to fit the
pieces together.

Peter

On Wed, 19 Jan 2000, Peter Mount wrote:
> I'm sending this to both the hackers and interfaces lists as this
> affects the 7.0 release and an interface.
>
> Ok, up until now the driver has operated under a base package of
> postgresql. This has worked fine so far but technically breaks Sun's
> rules on package naming. The rule is that any organisations package
> names begins with their domain name. This prevents two different package
> names from clashing.
>
> Ie: My own classes always begin with uk.org.retep as my own domain is
> retep.org.uk. The classes I write here begin with uk.gov.maidstone.
>
> Now, what I'm thinking is that as the 7.0 driver isn't going to be
> compatible with earlier backends (mainly due to the core changes like
> date/time handling, but there are others), I'm proposing to change our
> base package name from postgresql to org.postgresql so that we comply
> with this rule (which has been around since Java first came out).
>
> All this involves in the source is to create an empty directory called
> org, and move the original postgresql directory into it. Then each .java
> file will need org. prefixed to the package name.
>
> The down side, is that any existing source that uses the driver will
> need amending so that either the Class.forName() line reads:
>
> Class.forName("org.postgresql");
>
> or if it's supplied as a parameter (which is my prefered way) the org.
> added.
>
> Now because of this downside, I want to see what everyone thinks about
> making this change before I do it, as I have a lot of things to do to
> the source to implement it, but it would be better to do it now,
> especially as it's the first new major release since JDBC was included.
>
> Peter
>
> --
> Peter Mount
> Enterprise Support
> Maidstone Borough Council
> Any views stated are my own, and not those of Maidstone Borough Council.
>
>
>
> ************
>

--
Peter T Mount peter(at)retep(dot)org(dot)uk
Main Homepage: http://www.retep.org.uk
PostgreSQL JDBC Faq: http://www.retep.org.uk/postgres
Java PDF Generator: http://www.retep.org.uk/pdf

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2000-01-24 17:34:19 Re: [HACKERS] fatal copy in/out error (6.5.3)
Previous Message Bruce Momjian 2000-01-24 17:33:34 Re: [HACKERS] Well, then you keep your darn columns