Path: csiph.com!eternal-september.org!feeder.eternal-september.org!reader02.eternal-september.org!.POSTED!not-for-mail
From: Eric Sosman <esosman@comcast-dot-net.invalid>
Newsgroups: comp.lang.java.programmer
Subject: Re: Casting without Generics
Date: Mon, 9 Jul 2018 17:59:41 -0400
Organization: A noiseless patient Spider
Lines: 44
Message-ID: <pi0lsf$lkk$1@dont-email.me>
References: <casting-20180709205142@ram.dialup.fu-berlin.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Mon, 9 Jul 2018 21:59:44 -0000 (UTC)
Injection-Info: reader02.eternal-september.org; posting-host="f4774f41d776d8a1f45f2a9326b95747"; logging-data="22164"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/qlYDRfh9fWbBZIq9zdGA7"
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.8.0
Cancel-Lock: sha1:w54hQgxDTfVFbIU9hZO9JoNdVaY=
In-Reply-To: <casting-20180709205142@ram.dialup.fu-berlin.de>
Content-Language: en-US
Xref: csiph.com comp.lang.java.programmer:38479

On 7/9/2018 3:54 PM, Stefan Ram wrote:
>    The following programm yields no compile-time error
>    using the JDK 10:
> 
>    Main.java
> 
> public final class Main
> { public static void main( final java.lang.String[] args )
>    { final java.lang.CharSequence cs = "abc";
>      java.lang.System.out.println( ( java.io.PrintStream )cs ); }}
> 
>    transcript
> 
> Exception in thread "main" java.lang.ClassCastException: java.base/java.lang.String cannot be cast to java.base/java.io.PrintStream
>      at Main.main(Main.java:4)
> 
>    Why?

     It surprised me, too -- at first.  But consider:

	class Perverse extends PrintStream implements CharSequence {...}

... is certainly not forbidden, and if you were to write such a thing
it would then be possible to have a CharSequence instance that would
survive the cast-check with no trouble at all.  The mere fact that some
reference is to a CharSequence does not imply that it cannot also be
to a PrintStream.

>    I used to believe that the compiler would complain that
>    »java.io.PrintStream« is neither a super- nor a subtype
>    of »java.lang.CharSequence«.
> 
>    Did I miss something?

     The rules are spelled out in JLS 5.1.6, with special reference
to 5.1.6.1-3.  Roughly speaking, the compiler need not object to a
conversion that is *possibly* valid if there exists a foolproof way
to check the validity at run-time.  (If there's no foolproof run-time
check the compiler must object unless the conversion is *assuredly*
valid.)

-- 
esosman@comcast-dot-net.invalid
Nine hundred twenty-six days to go.