Re: Is this a bug?

[menakkis(at)my-dejanews.com]

steinhh@ulrik.uio.no (Stein Vidar Hagfors Haugan) wrote:
<...>
>  1. Brackets around anything *except* a simple variable name gives
>     an rvalue (expression).
>  2. Dereferenced pointers -- "*p" behave as a simple variable name,
>     -> see exception in first point
>  3. Rvalues should *not* be allowed on the left hand side of
>     assignments (fix it, RSI).

It looks like you have it.   I'd go so far as to add that:  brakets around a
simple variable name are effectively ignored / discarded.   e.g., A=INDGEN(4)
&(((A)))[2]=20 works.


> Finally, I guess that Peter Mason changing his mind in the middle
> of a paragraph caused him to say things about the
>     (<expression>)[index]
> construct whilst appearing to talk about the (*p) construct..?

I guess I should have thrown in a <NL> there for your paragraph parser :-)


Peter Mason

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own