[Mesa-dev] [PATCH mesa] Introduce .editorconfig
Rob Clark
robdclark at gmail.com
Tue Aug 30 16:32:19 UTC 2016
On Tue, Aug 30, 2016 at 12:16 PM, Emil Velikov <emil.l.velikov at gmail.com> wrote:
> On 29 August 2016 at 22:30, Eric Engestrom <eric at engestrom.ch> wrote:
>
>> A couple of people have also suggested I add `tab_width`, but IMO this concept
>> should die: it comes from the confusion between indentation (a logic concept
>> meant to convey code structure) and alignment (an aesthetic concept).
>> If you need a specific size for your tabs, you're not indenting, you're
>> aligning, which makes no sense to do with a character that will have a varying
>> size in the first place.
>> There is a (pointless IMO, who cares?) war between tab and space indentation
>> with various arguments on each side, but I have yet to find anyone with a single
>> argument in favour of using varying-width characters to align code... yet I keep
>> seeing people doing it.
>> (OK, that's not true: I do have a preference for space indentation, because it
>> makes it harder for people to do the wrong thing when they get confused)
>> </rant> (sorry)
>
> Hehe [1] comes to mind when people start talking on the topic. Then
> again some prefer to separate the two - use tabs for indent and space
> for alignment, but that's likely a minority so I won't bother.
>
> But seriously: tab_width serves more than just alignment purposes -
> people have different setups/visual comprehension thus wide tabs can
> make things easier to digest or (on the other end of the spectrum) too
> annoying/distracting ;-)
I end up using tabs for indent, spaces for alignment (/me ducks)
but anyway, I'd say to just go ahead and push it, and I'll fixup the
freedreno bits after the fact ;-)
BR,
-R
> That said, we can tweak things as/if needed.
>
> Thanks,
> Emil
>
> [1] https://camo.githubusercontent.com/96d51190854058fe8f6953dd4f1a50047ef47bd9/687474703a2f2f6576616465666c6f772e636f6d2f77702d636f6e74656e742f75706c6f6164732f323031312f30332f54616273537061636573426f74682e706e67
More information about the mesa-dev
mailing list