[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Naming conventions?
From:       Richard Smith <kde () metafoo ! co ! uk>
Date:       2004-10-13 11:53:07
Message-ID: 200410131253.08169.kde () metafoo ! co ! uk
[Download RAW message or body]

On Wednesday 13 October 2004 03:10, Aaron Seigo wrote:
> On October 12, 2004 9:17, Richard Smith wrote:
> > <tab>someFunctionCall( arg, arg, arg, arg, arg, arg, ...
> > <tab><spaces>            arg, arg, arg, arg, arg, arg, ...
> >
> > and press return, it fills up to the appropriate place with tabs! Given
> > that the whole point of using tabs for indentation is to allow people to
> > set their own indent width without breaking the code's appearance, this
> > is clearly wrong.
>
> in the example you give above where you have tabs and then a bunch spaces
> to fudge the indent to line up properly, as soon as someone changes the
> tabstop width your space-fudges will no longer be right and the code will
> missalign.

I'm sorry? The whole point of using spaces there (which is not a fudge, it's 
simply the only right way of doing it) is exactly so, if someone changes the 
tab width, things still line up. The philosophy is simple: use tabs to indent 
only (ie, one tab in for every nested scope) and use spaces for everything 
else.

> i've yet to see very many coders stick to tabs-only;

That's because tabs-only is broken. As should be obvious from the example 
above - if you replaced the spaces with tabs, and changed your tab width, the 
code would be misaligned.

You've clearly misunderstood the purpose of the spaces above. They work fine 
if your tab size is set to 8, 2 or pi. Please read my message again. If 
you're still mystified, Charles Samuels (I assume) has written an excellent 
summary of the tabs issue here: http://www.derkarl.org/why_to_tabs.html

Thanks,
Richard
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]