Re: Versioning policy PgJDBC - discussion

From: Dave Cramer <pg(at)fastcrypt(dot)com>
To: Gavin Flower <GavinFlower(at)archidevsys(dot)co(dot)nz>
Cc: Vladimir Sitnikov <sitnikov(dot)vladimir(at)gmail(dot)com>, Jorge Solórzano <jorsol(at)gmail(dot)com>, List <pgsql-jdbc(at)postgresql(dot)org>
Subject: Re: Versioning policy PgJDBC - discussion
Date: 2016-11-27 11:31:25
Message-ID: CADK3HH+Dr8hkX_BzJtVbb9BmO7RJ_FH=5HvFGYGP=HFb0-Xn9g@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-jdbc

This is from one of the packagers:

I'd be more concerned with dependencies in the Java/Maven world. If
anyone was a dependency in there like jdbc-postgresql >= 9, you'll
kill them. (This is not a problem in the Debian version number
namespace because 1:4 > 9, but unless you'd be using 1:4.2.0 in the
Maven world as well...)

And the other

I'm in favor of that. Even I, as a packager, almost fail all the times when
I
see "9.4" there.

I think the maven issue is real, which may put Jorge's idea in the lead?

Dave Cramer

davec(at)postgresintl(dot)com
www.postgresintl.com

On 25 November 2016 at 14:02, Gavin Flower <GavinFlower(at)archidevsys(dot)co(dot)nz>
wrote:

> On 26/11/16 01:08, Vladimir Sitnikov wrote:
>
>>
>> >We've changed the numbering scheme once already
>>
>> AFAIK, the change from 9.4-1210 to 9.4.1211 was made to follow common
>> convention where version number is separated with dots.
>>
>> I would agree that it is still common for end-users to confuse 9.4 part
>> with PostgreSQL version.
>>
>
> My instinctive reaction IS to think that the 9.4 refers to the pg version,
> though i know it is not for several years already!!!
>
> I suggest that credit should be given to Douglas Adams who wrote THGTTG,
> for enlightening us as to the significance of '42' as the answer to 'the
> Ultimate Question of Life, the Universe and Everything'! See:
> http://hitchhikers.wikia.com/wiki/42
>
> I once had to construct some indexes to Index Sequential files on an ICL
> mainframe, shortly after the BBC had aired the THGTTG, and at least 3 of
> the index keys turned out to be 42 bytes long - suspicious omens???
>
>
>
>
>> So moving to pgjdbc 42.0.0 would probably make sense.
>>
>> Just in case: for current pgjdbc 9.4.1212, "9.4" mean nothing. "1212"
>> is just a sequence number.
>> So 42.0.0 would not harm much.
>>
>> However, it would enable us to use 42.0.1 vs 42.1.0 for "bugfix" vs "new
>> features" releases.
>> Current pgjdbc versioning scheme does not leave much room for pgjdbc
>> 9.5.0 or alike.
>>
> +1
>
>
>> Vladimir
>>
>> пт, 25 нояб. 2016 г. в 14:52, Dave Cramer <pg(at)fastcrypt(dot)com <mailto:
>> pg(at)fastcrypt(dot)com>>:
>>
> [...]
>
>
> Cheers,
> Gavin
>
>

In response to

Responses

Browse pgsql-jdbc by date

  From Date Subject
Next Message Vladimir Sitnikov 2016-11-27 13:40:10 Re: Versioning policy PgJDBC - discussion
Previous Message Gavin Flower 2016-11-25 19:02:40 Re: Versioning policy PgJDBC - discussion