Re: has my bug report & fix been submitted (in NSProcessInfo.m)?

[Richard Frith-Macdonald]

On 2005-07-19 15:18:19 +0100 Andrew Ruder <aeruder@ksu.edu> wrote:

> Lloyd Dupont wrote:
> > here was the fix:
> > NSProcessInfo.m: 669
> > + (void) initialize
> > {
> >    if (self == [NSProcessInfo class] && !_gnu_environment)
> >      _gnu_process_args(__argc, __argv, _environ);
> > }
> > I add the "&& !_gnu_environment" test
>
> I don't think this could possibly be the correct fix.  This is just 
> duct-taping a fix over something else which is obviously broken... Aren't + 
> initialize methods only supposed to be called once?

The 'self == [NSProcessInfo class]' clause is a standard/common thing ... to 
protect against the case where a subclass calls +initialize (yes, it can get 
called more than once ... it's just that the runtime guaranteese to call it 
once before calling other methods of the class).

The !_gnu_environment is the important bit ... to prevent 
_gnu_process_args(__argc, __argv, _environ) being called by +initialize if 
it has already been called.  On windows, I think it is normally called in 
the fake main process, before +initialize is called.

I believe Lloyd needed to do this because he had initialised NSProcessInfo 
values with a custom version of the environment, and didn't want  a later 
call to +initialize to overwrite stuff with environment information from the 
_environ global variable.

Basically, _gnu_process_args() can be called before +initialize, so it's 
good for +initialise to not call it again  under those circumstances.