Re: Question about short circuit operators

[Robert T. Short]

Jordi Gutiérrez Hermoso wrote:
> On 24 February 2011 16:23, Robert T. Short
> <address@hidden>  wrote:
>    
> > Before I submit a bug report, maybe someone can tell me that I am
> > missing something.  Why would the following happen?
> >      
> [snip]
>    
> > warning: /home/rtshort/Mirrored/octave/Sources/00Bugs/somefunc.m:
> > possible Matlab-style short-circuit operator at line 3, column 20
> >      
> Just that, you're using operators that short circuit but only if you
> use them as a condition in an if or while.
>
> E.g.
>
>       a = b | c; ## No short circuiting in Matlab or Octave
>
>       if b | c  ## Short circuiting in Matlab, use --braind-dead
>                 ## options to enable it in Octave
>
> HTH,
> - Jordi G. H.
>
>
>    
Ahh.  I see.  However, I would never have got that message from the 
octave documentation.  I read the MATLAB documentation to understand 
about "short-circuiting in element-wise...".  I realize this was 
discussed earlier, but I didn't get involved because I didn't understand 
the implications based on the documentation.

It seems to me that the crux of the problem is that

x = logical_expression;
if(x)
  ...
end

and

if(logical_expression)
  ...
end

might have different results, yes?  That is certainly braindead behavior.


Now, here is my problem with this warning.  The example I submitted was 
clearly contrived simply to get to the root of the problem, but it is 
NOT "braindead".  It is a simple and straightforward statement.  Why 
produce a warning for something like this?  The current approach is 
simply too heavy handed.  Similarly the "braindead" option is far, far 
too draconian to be a solution.

IMHO.  Don't try to protect people from themselves.  Document the 
problem, explain what the problem is and why there is a problem and 
leave it alone.  Get rid of that warning.


Just my two cents.

Bob