> -----Original Message-----
> From: Stefan Sperling [mailto:stsp_at_elego.de]
> Sent: maandag 14 maart 2011 17:55
> To: Daniel Shahaf
> Cc: Kamesh Jayachandran; dev_at_subversion.apache.org
> Subject: Re: [PATCH] Add --dry-run flag to "svn update" client command
>
> On Mon, Mar 14, 2011 at 06:34:15PM +0200, Daniel Shahaf wrote:
> > Kamesh Jayachandran wrote on Fri, Mar 11, 2011 at 18:12:21 +0530:
> > > Unless there are no objection I will commit this patch post I
> > > receive a r1075802 regression fix(and testcase) from either Daniel
> > > Becroft or Arwin.
> >
> > Hold on, are you suggesting to commit the two thousand, six hundred and
> > sixty-three line diff that started this thread?
>
> So what if it's long? I'd be much more worried if this diff was shorter.
> It needs to make changes in lots of places.
It adds an insane amount of conditionals. How are you going to test this and
maintain it?
The complexity of sourcecode increases by every conditional.
I don't see the added benefit of having this feature compared to the
complexity this implementation of this feature adds. There must be simpler
ways to add this, without slowing down the work that still has to be done on
the update editor before releasing 1.7.
This implementation makes me starting to think about when we have to start
on WC-NG-NG to remove the complexity added here.
Bert
Received on 2011-03-14 18:22:44 CET