[PATCH] Fix the build

[Rowland Penny]

On Fri, 2 Sep 2016 09:05:27 -0700
Jeremy Allison <jra at samba.org> wrote:

> On Fri, Sep 02, 2016 at 09:21:42AM +0100, Rowland Penny wrote:
> > On Fri, 2 Sep 2016 09:34:49 +0200
> > Volker Lendecke <vl at samba.org> wrote:
> > 
> > > Hi!
> > > 
> > > Review appreciated!
> > > 
> > > Looking at commit 1c636532874da from a few weeks ago I begin to
> > > question the value of our README.Coding file. I've asked a few
> > > times to fix patches to follow the 80-column rule, I even
> > > provided patches to assist.
> > > 
> > > There's a reason why we have this rule: It's not that we are all
> > > sitting at 3270 or vt100 terminals. We want to avoid arbitrarily
> > > deeply nested control structures. It might be more work, but
> > > well-named factored out subfunctions foster unterstanding of
> > > complex code. Looking at dsdb_garbage_collect_tombstones(), we
> > > have four (!!)  levels of nested for-loops. One line I've just
> > > come across almost touches twice the 80-columns with its length
> > > of 157 chars.
> > > 
> > > So, shall we drop the README.Coding section on 80 chars, as it is
> > > not generally seen as worthwhile following?
> > > 
> > > Volker
> > 
> > Hi Volker, No, but you could fix the link to PEP8, it now seems to
> > be at: https://www.python.org/dev/peps/pep-0008/
> > 
> > Until you raised this, I didn't know about README.Coding, nobody
> > told me, perhaps this is one reason why the 80 columns rule gets
> > broken.
> 
> Ah that's just the Samba way Rowland. You break the rules and
> then someone complains about it and then you find the rule existed.
> 
> Been working well for *years* :-).

Ah, you mean it works like local government, you only find out you have
broken a rule (that you didn't know existed) after you break the rule
;-)

Rowland