Re: [classlib] generics puzzler

[Stepan Mishura <st...@gmail.com>]

On 5/11/06, Tim Ellison wrote:

> But the question remains in my mind whether there is any generic type
> definition you could write that would allow you to cast the  entrySet()
> to a "Set<Map.Entry<capture-of ? extends K, capture-of ? extends V>>"
> equivalent?
>
> To put it another way, I would expect that
>        for (Map.Entry<? extends K, ? extends V> entry : map.entrySet())
>
> should complain that the declaration of 'entry' is not compatible with
> the 'capture-of ? extends K' etc.


Basing on intuition (but it put me in a spot quite often when I work with
generics :-)) I'd say that the declaration of 'entry' in for-loop is
equivalent the the following:
Map.Entry<? extends K, ? extends V> entry = map.entrySet
().iterator().next();

And there are no compiler complains like "cannot convert from ... capture-of
? extends K ... ". So I assume that there is something 'hidden logic' in new
for-loop syntax that lets to iterate on collections in this way. And it
would be good to figure out what exactly is going on.




> (It's still a useful cheat though)


Yep!

Thanks,
Stepan.





> Regards,
> Tim
>
>
> Neil Bartlett wrote:
> > Nathan,
> >
> > It's a tricky one to be sure.
> >
> > The problem is that the entrySet() method is returning a
> > "Set<Map.Entry<capture-of ? extends K, capture-of ? extends V>>",
> > which is incompatible with the type "Set<Map.Entry<? extends K, ?
> > extends V>>". It's easier to describe why if I drop the "extends K"
> > and "extends V" part. So we have "Set<Map.Entry<?, ?>" and
> > "Set<Map.Entry<capture-of ?, capture-of ?>>".
> >
> > The first one, "Set<Map.Entry<?, ?>>" is a set of Map.Entries of
> > different types - ie it is a heterogeneous collection. It could
> > contain a Map.Entry<Long, Date> and a Map.Entry<String, ResultSet>>
> > and any other pair of types, all in the same set.
> >
> > On the other hand, "Set<Map.Entry<capture-of ?, capture-of ?>>" is a
> > homogenous collection of the same (albeit unknown) pair of types. Eg
> > it might be a Set<Map.Entry<Long, Date>>, so all of the entries in the
> > set MUST be Map.Entry<Long, Date>.
> >
> > In general, even if you could assign this to a local variable without
> > unchecked type safety warnings, you wouldn't be able to call any
> > methods on it because the compiler cannot check whether the access is
> > type-safe. For example you would not be able to call add() because the
> > Set must contain a specific type, but the compiler doesn't know which
> > specific type that is!
> >
> > Regards
> > Neil
> >
> > On 5/11/06, Nathan Beyer <nb...@kc.rr.com> wrote:
> >> Does someone understand why this works this way? This seems so odd. I
> >> know
> >> there are quirks to the generics syntax, but this in an edge I haven't
> >> run
> >> into yet. I haven't been able to make this one click in my head yet.
> >>
> >> This compiles fine:
> >>         public synchronized void putAll(Map<? extends K,? extends V>
> >> map) {
> >>         for (Map.Entry<? extends K,? extends V> entry : map.entrySet())
> {
> >>               ...
> >>         }
> >>         }
> >>
> >> This won't compile at all:
> >>         public synchronized void putAll(Map<? extends K,? extends V>
> >> map) {
> >>         Set<Map.Entry<? extends K, ? extends V>> entries =
> >> map.entrySet();
> >>         ...
> >>       }
> >> The error is: Type mismatch: cannot convert from
> >> Set<Map.Entry<capture-of ?
> >> extends K,capture-of ? extends V>> to Set<Map.Entry<? extends K,?
> extends
> >> V>>
> >> The suggested quick fix in Eclipse is "Set<?> entries = ...".
> >>
> >> This reports a warning:
> >>         public synchronized void putAll(Map<? extends K,? extends V>
> >> map) {
> >>         Map<K,V> map2 = (Map<K,V>)map;
> >>       }
> >> The warning is: Type safety: The cast from Map<capture-of ? extends
> >> K,capture-of ? extends V> to Map<K,V> is actually checking against the
> >> erased type Map
> >> The suggested quick fix in Eclipse is to the annotation:
> >> @SuppressWarnings("unchecked").
> >>
> >> -Nathan
> >>
> >>
> >> > -----Original Message-----
> >> > From: Tim Ellison [mailto:t.p.ellison@gmail.com]
> >> > Sent: Wednesday, May 10, 2006 6:59 AM
> >> > To: harmony-dev@incubator.apache.org
> >> > Subject: Thanks Stepan! (was: Re: [jira] Resolved: (HARMONY-454)
> >> > [classlib][luni] java.util.Set generics uplift and related changes)
> >> >
> >> > Stepan Mishura (JIRA) wrote:
> >> > <snip>
> >> > > 2) To avoid casting while-loop was replaced with for-loop. Could
> you
> >> > review the change?
> >> >
> >> > I was scratching my head about this cast, so I was very pleased to
> see
> >> > your elegant solution.
> >> >
> >> > I must admit that I don't really understand why the for-loop version
> is
> >> > inherently different (other than it 'hides' the casting) -- but I've
> >> > learned a new pattern there :-)
> >> >
> >> > Regards,
> >> > Tim
> >> >
> >> > --
> >> >
> >> > Tim Ellison (t.p.ellison@gmail.com)
> >> > IBM Java technology centre, UK.
> >> >
> >> >
> >> > ---------------------------------------------------------------------
> >> > Terms of use : http://incubator.apache.org/harmony/mailing.html
> >> > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> >> > For additional commands, e-mail:
> harmony-dev-help@incubator.apache.org
> >>
> >>
> >> ---------------------------------------------------------------------
> >> Terms of use : http://incubator.apache.org/harmony/mailing.html
> >> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> >> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> >>
> >>
> >
> >
>
> --
>
> Tim Ellison (t.p.ellison@gmail.com)
> IBM Java technology centre, UK.
>
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>


-- 
Thanks,
Stepan Mishura
Intel Middleware Products Division

------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org