ooc-dev team mailing list archive

[Oddity007 <oddity007@xxxxxxxxx>]

A question I'm not sure has been asked is whether it would be easy to  
distiquish whether a member is that of the class, of it's instance, a  
local variable, or a global variable.

Imagine the following that is several functions down the file within a  
class declaration.

...
foo: func->Int{
  ....  couple lines in
  return a+b+c+d
}

If I ever saw this line, I would remove all hair upon my noggin  
barehanded!  "a" could be "This a", "this a", the global var "a", or a  
local, same applies to the other variables, with which the poor reader  
would have to manually trackdown which is declared where.

I'm not really one to like forcing of a coding style (much of the  
reason why I disagree with enforcing lowercase function names), but  
this is something to think of.

And while no sane coder would allow such vile code to be written, it  
is common knowledge that not everyone is sane.  But why am I  
explaining the reasoning of PERL?

On Apr 9, 2010, at 6:05 PM, Scott Olson <scott@xxxxxxxxxxxxxxx> wrote:

> In j/ooc you can't access static members without prepending "This ".
> I don't think this should be enforced, but rather simply mentioned  
> in the
> style guide ORC (which still needs to be written, btw). It  
> especially doesn't
> make sense with constants, eg. in oos, functions inside Display  
> class have to
> use "This VIDEO_MEMORY" instead of just "VIDEO_MEMORY". According to  
> the erorr
> message given when you don't use "This", this was done for 'clean  
> code', but
> in my opinion it is sometimes cleaner without "This" and the option  
> should be
> left to the user.
>
> See http://github.com/tsion/oos/blob/covers-and-
> statics/Src/Kernel/Hal/Display.ooc
>
> I don't believe it would make it any cleaner to pepper that code  
> with "This".
> In fact quite the opposite.
>
> In j/ooc, all you have to do is remove 3 lines of code that call that
> exception and it all works fine. I don't know what must be done in  
> rock, but it
> should also be able to do unqualified member access.
>
> if anyone else has any thoughts or arguments, I'd like to hear them.
>
> - Scott Olson
>
> _______________________________________________
> Mailing list: https://launchpad.net/~ooc-dev
> Post to     : ooc-dev@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~ooc-dev
> More help   : https://help.launchpad.net/ListHelp