svn commit: r442588 - in head/www: nginx nginx-full

[Bartek Rutkowski]

> On 6 Jun 2017, at 10:39, Alexey Dokuchaev <danfe at FreeBSD.org> wrote:
> 
> On Mon, Jun 05, 2017 at 05:50:06PM -0600, Adam Weinberger wrote:
>>> On 4 Jun, 2017, at 18:18, Sergey A. Osokin <osa at FreeBSD.org> wrote:
>>> 
>>> Hi Bartek and Adam,
>>> 
>>> I don't think I can get this, so two questions for you guys:
>>> o) what was the reason to bump PORTREVISION in www/nginx?
>>> o) wouldn't it btter to just bump PORTREVISION in www/nginx-full?
>> 
>> Hi Sergey,
> 
> [ Wrapping very long lines ]
> 
>> I'll give Bartek a chance to explain in more detail, but I supported an
>> nginx bump because it was less complex for the future.
>> 
>> If nginx-full got a bump, then it would need to be bumped every time
>> nginx got bumped, or nginx would have to be bumped by two and nginx-full's
>> PORTREVISION line gets removed, and then the line has to be removed at the
>> next nginx update or reset. At the end of the day, bumping nginx was more
>> straightforward. It triggers an update for everyone else, but becomes less
>> invasive over the long haul.
> 
> It seems that everyone bumps port revisions whenever they please these days;
> wondering about it just a waste of time.  Just an exampler: r442562, where
> it was bumped for pkg-descr change (sic!) in a port that takes considerable
> time to build. :-(
> 
> ./danfe

This wasn't the case here, so I'd take your comment as a general one ;) I can't speak for reasons behind other bumps, so I won't, but I am personally aware of the docs on when the bump should happen and I try to adhere to these. When in doubt, I seek inspiration in portmgr members and their insight was promptly provided every time, including this one.

Kind regards,
Bartek Rutkowski