Re: missing patch comments should not be classified as an error

[Alistair Crooks <agc%pkgsrc.org@localhost>]

On Sat, Sep 15, 2012 at 09:18:48PM +0200, John Marino wrote:
> On 9/15/2012 19:49, David Holland wrote:
> >On Sat, Sep 15, 2012 at 03:49:51PM +0200, John Marino wrote:
> >  >  I don't know when this changed, but I just installed the latest
> >  >  pkglint to clean up a makefile and it's reporting a missing comment
> >  >  on a patch as an error.
> >  >
> >  >  While I've softened my stance on patch comments in recent weeks, I
> >  >  can not agree that a patch lacking a comment constitutes an error.
> >  >  That's a bit too draconian.  Many of these patches are very old and
> >  >  many others frankly don't need a comment.
> >  >
> >  >  Can we get this put back to a warning?
> >
> >Given that the change was made intentionally in response to the last
> >round of flamage on this issue, I doubt you'll convince anyone.
> >
> >I think it's fine the way it is...
> 
> 
> Alright, then I'll just locally remove this error from pkglint and
> the "new pkglint" available warning and not update it anymore.
> Having it as a warning was obnoxious enough, but one could live with
> that.

I'm interested to know why you think that commenting why a change is
made is a bad thing.  Now, granted, not every patch has historically
had a comment, but that springs mostly from 15 years ago - I'd like to
think that over time, we gain as much information in the patch itself
about why it was needed. Information and its sharing is a good thing;
please continue to practise this.

On the "I don't consider your rules workable/viable/sane, so stuff
you" issue, I'd encourage you to reconsider.  I wouldn't want to be
thought of by others as being irresponsible in the way I approach
things in pkgsrc; I would hope that other pkgsrc developers would
share my attitude in this regard.

And finally, I'd also encourage you to treat your co-developers with
the respect that you would like to be shown yourself.

Thanks,
Alistair