mixed languages build / constructor lookup and boolean.class

[rvjansen]

Hi Mike and others,

I've got two problems, one of which is the secundary one. My main
problem is co-dependencies in Java and NetRexx. Co-dependencies are
resolved if we compile the modules together; I'd like to be able to
send Java code into the NetRexx compiler.

While trying to solve some of these, I resort to
mClass = Class.forname("ScopeDialog")
con = mClass.getConstructor(Class[] [JFrame.class, boolean.class,
InScopeTableModel.class])
inscopeDialog =  Maintaining con.newInstance(java.lang.Object[]
[inScopeFrame, boolean 0, ist])

This works fine when I have one argument, in our case mostly a JFrame.
The compiler complains in this case on the boolean.class, with the
message : Error: Partial term is primitive type 'boolean', which
cannot have methods or properties

The java samples use int.class anyway, which also does not work in
NetRexx.

What do I do now?

best regards,

Rene Jansen.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

[rvjansen]

>
> Co-dependencies don't really have a solution.
>
> int and boolean (etc.) are Java reserved words.   They cannot be used for
> identifiers.
>
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> Mike Cowlishaw, IBM Fellow
Mike,

isn't it possible to send .nrx and .java into the compiler, and have the
generated java and the .java files go unmodified into the java compiler so
it can resolve the signatures?

On the second point, it *is* valid use in java to use int.class and
boolean.class as members of an array of Class[] . It is NetRexx that is
flagging it down, saying that you cannot have methods or properties on
primitive types. I think that just might be wrong in this context.

best regards,

Rene.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

[rvjansen]

>
> The Java language spec. (Gosling & co., 1996) says these cannot be used as
> identifiers (section 3.9), and so NetRexx enforces that rule to save
> people
> from generating Java source files that are illegal.   Possibly there's
> some
> later change, but that would be an incompatible change, so I hope not :-(.
> It's not something I would change at this point.
>
I typed in the following sample from The Java Developers Almanac 1.4 ,
Chan et al., page 111, Getting a Constructor of a Class Object:

class test {

    public static void main(String[] args){
        try{
            java.lang.reflect.Constructor con =
java.awt.Point.class.getConstructor(new Class[]
{int.class,int.class});
        } catch (Exception e) {};
    }
}

This compiles without any pain.

I spent some time browsing "The Java Language Specification", and would
like to quote paragraph 15.8.2: A *class literal* is an expression
consisting of the name of a class, interface, array or primitive type
followed by a '.' and the token *class*.

This is from the 2000 edition. The concept of an int.class is first
mentioned in the 1.1 update to the 1996 edition.

(http://java.sun.com/docs/books/jls/first_edition/html/1.1Update.html ,
D.7.3)

I added a constructor that makes the dialog always modal (so I would not
need the boolean.class right now), but I really like to be able to use
this.

best regards,

Rene-Vincent Jansen.




~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

[Patric Bechtel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tue, 18 Feb 2003 17:07:01 +0000, Mike Cowlishaw wrote:

Hello,

shouldn't this be Integer.TYPE ? I used it several times, and it works...

Patric


>Ah, sorry .. yes, it's coming back to me.    Yes I think int.class is
>allowed -- see 'Special names' in the NetRexx supplement document.

>NetRexx does anything that 1.1 can do, I believe.

>- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
>Mike Cowlishaw, IBM Fellow
>IBM UK (MP5), PO Box 31, Birmingham Road, Warwick, CV34 5JL
>mailto:[hidden email]  --  http://www2.hursley.ibm.com/decimal

>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
>[hidden email]
>with the following message in the body of the note
>unsubscribe ibm-netrexx <e-mail address>




PGP Public Key Fingerprint: 2636 F26E F523 7D62  4377 D206 7C68 06BB

-----BEGIN PGP SIGNATURE-----
Version: PGPsdk version 1.7.1 (C) 1997-1999 Network Associates, Inc. and its affiliated companies.

iQA/AwUBPlJ1vHxoBrvMu8qQEQLgrwCfS4JNiBM6okmMFMh6M2Xunmg47WYAoKjw
2TnBnE2QTX6HUVUqvm91mEQ4
=bmmf
-----END PGP SIGNATURE-----


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>