[EISA] const correctness in EISA

[James Hill]

Hey Will, 

> -----Original Message-----
> From: William R. Otte [mailto:wotte at dre.vanderbilt.edu] 
> Sent: Wednesday, December 03, 2008 3:11 PM
> To: James Hill
> Cc: eisa at list.isis.vanderbilt.edu
> Subject: Re: [EISA] const correctness in EISA
> 
> Hi James -
> 
> Thanks!  I will take care of that.
> 
> Which examples/tests should I run to test functionality?

Please use the application under:

  $EISA_ROOT/tests/sprint

Instructions on running the system are at the following location:

  https://www.dre.vanderbilt.edu/eisa/index.php/Deployment

BTW, please make sure you have build the $EISA_ROOT/examples.mwc
workspace before building the $EISA_ROOT/tests/sprint.mwc workspace.

Thanks,

James

> 
> thanks,
> /-Will
> 
> On Dec 3, 2008, at 3:35 PM, James Hill wrote:
> 
> > Hey Will,
> >
> >> -----Original Message-----
> >> From: eisa-bounces at list.isis.vanderbilt.edu
> >> [mailto:eisa-bounces at list.isis.vanderbilt.edu] On Behalf 
> Of William 
> >> R. Otte
> >> Sent: Wednesday, December 03, 2008 2:15 PM
> >> To: eisa at list.isis.vanderbilt.edu
> >> Subject: [EISA] const correctness in EISA
> >>
> >> Hi folks -
> >>
> >> I am looking through the EISA code, and I had a question....
> >> there are several instances similar to the following where 
> the intent 
> >> of the const is essentially meaningless:
> >>
> >> EISA_EINode_i * const einode ();
> >>
> >> which means the method returns a const pointer to a 
> mutable object, 
> >> which in this case the const pointer in question is a 
> temporary and 
> >> ignored by some compilers.
> >>
> >> Is the intent to make the return value pointer to a 
> constant object, 
> >> ie:
> >> const EISA_EINode_i * einode ();
> >>
> >> or the method:
> >> EISA_EINode_i * einode () const;
> >>
> >> or is it exactly as intended?
> >
> > My understanding of that was incorrect then. We need to remove the 
> > 'const' modifier b/c you don't want it to go at the end of 
> the method.
> >
> > Thanks,
> >
> > James
> >
> >>
> >> thanks,
> >> /-Will
> >>
> 
>