Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #33387 > unrolled thread

Re: Today I invented ... the downcast!

Back to article view | Back to comp.lang.java.programmer

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: Today I invented ... the downcast! Eric Sosman <esosman@comcast-dot-net.invalid> - 2015-12-24 15:12 -0500
    Re: Today I invented ... the downcast! Lew <lewbloch@gmail.com> - 2015-12-24 13:11 -0800

#33387 — Re: Today I invented ... the downcast!

On 12/24/2015 2:39 PM, Stefan Ram wrote:
>    What do you all thing about it?

     Meh.

>    Example of the intended usage:
>
> final android.widget.LinearLayout.LayoutParams parameters
> = downcast( params );
>
>    Definition:
>
> static private final< T, O >T downcast( O x ){ return( T )x; }
>
>    Main Advantage (that's why I wrote it):
>
>        - the programmer conveys his intention that this
>          cast is intended to be a /down/cast

     Since an /up/cast requires no cast operator at all, the mere
presence of the cast is a pretty good indicator.  Besides,

	String s = "I feel sad and downcast. (:";
	Object o = downcast(s);  // "down?"

>    Advantages that are mere side-effects of the implementation:
>
>        - the programmer does not actually have to write
>          the target type, which is deduced
>
>    Disadvantage:
>
>        - It has not yet been extensively tested
>
>        - It does not check at compile time whether T
>          really is a /sub/type of O.

    ... thus converting a compile-time error into a run-time error,
a wrong in the step direction.  Hmm: I withdraw my "Meh" above, and
change it instead to "Ugh."

-- 
esosman@comcast-dot-net.invalid
"Don't be afraid of work. Make work afraid of you." -- TLM

[toc] | [next] | [standalone]

#33388

On Thursday, December 24, 2015 at 12:12:32 PM UTC-8, Eric Sosman wrote:
> On 12/24/2015 2:39 PM, Stefan Ram wrote:
> >    What do you all thing about it?
> 
>      Meh.
> 
> >    Example of the intended usage:
> >
> > final android.widget.LinearLayout.LayoutParams parameters
> > = downcast( params );
> >
> >    Definition:
> >
> > static private final< T, O >T downcast( O x ){ return( T )x; }
> >
> >    Main Advantage (that's why I wrote it):
> >
> >        - the programmer conveys his intention that this
> >          cast is intended to be a /down/cast
> 
>      Since an /up/cast requires no cast operator at all, the mere
> presence of the cast is a pretty good indicator.  Besides,
> 
> 	String s = "I feel sad and downcast. (:";
> 	Object o = downcast(s);  // "down?"
> 
> >    Advantages that are mere side-effects of the implementation:
> >
> >        - the programmer does not actually have to write
> >          the target type, which is deduced
> >
> >    Disadvantage:
> >
> >        - It has not yet been extensively tested
> >
> >        - It does not check at compile time whether T
> >          really is a /sub/type of O.
> 
>     ... thus converting a compile-time error into a run-time error,
> a wrong in the step direction.  Hmm: I withdraw my "Meh" above, and
> change it instead to "Ugh."

<O, T extends O>
But the unchecked cast 
(T) O
is problematic, and exactly the sort of thing generics are supposed to obviate.

Define the logic for which you think you need casting to use a <T extends U> in the first place.

-- 
Lew

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.java.programmer

csiph-web