[PATCH 2 of 2] patchbomb: add --no-intro option

Matt Mackall mpm at selenic.com
Sun Mar 4 17:34:40 CST 2012


On Thu, 2012-03-01 at 20:31 +0100, Yann E. MORIN wrote:
> Greg, All,
> 
> On Thursday 01 March 2012 04:17:54 Greg Ward wrote:
> > On 29 February 2012, Yann E. MORIN said:
> > > # HG changeset patch
> > > # User "Yann E. MORIN" <yann.morin.1998 at free.fr>
> > > # Date 1330467959 -3600
> > > # Node ID 18cdc18bd6cc4bbd9cb9a0512faa6b1151c000e7
> > > # Parent  e6fb31285eabdb9f3092860791be954ea826d82e
> > > patchbomb: add --no-intro option
> > 
> > I sent a similar patch back in Sept 2011, and Matt did not like the
> > extra complication added. See
> >   http://thread.gmane.org/gmane.comp.version-control.mercurial.devel/44293
> 
> Ah, OK. I missed that.
> 
> (Note: in the following, Mercurial = the project, while hg = the tool)
> 
> The problem is that the intro/no-intro policy centers (IMHO) a lot around
> Mercurial's use-case, while a lot of other projects may have a different
> policy of requiring introduction message (at least the project I maintain
> does highly favor introduction messages, and accepts single-patch with
> introduction, because it makes sense in some cases).

I'm confused. Greg says "here's why mpm rejected this last time", you
say "ah, ok" and then.. talk about something completely different? My
earlier mail is primarily about the accrual of poorly thought-out,
conflicting options and NOT project policy. Hence all the detailed truth
tables. Please read it again, in particular this line:

        Horrors: we've now taken something that should have been
        completely
        trivial and made it possible for users to get errors.
        
http://thread.gmane.org/gmane.comp.version-control.mercurial.devel/44293/focus=44309

Note that you've really got an uphill battle to convince me that having
--foo and --no-foo options is ever a good design, regardless of what
"foo" is. Also note that your patch fails to complain if conflicting
options are provided, which is just a bug report waiting to happen.

My inclination right now is to simply add "(optional)" to the intro
prompt and move on.

As for your first patch, I might be inclined to accept something that
simply disables intro messages, but I think that adding
nearly-but-not-quite-Boolean options is a similar class of design
mistake.

-- 
Mathematics is the supreme nostalgia of our time.




More information about the Mercurial-devel mailing list