[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Virtual Packages proposal



[You (Hamish Moffatt)]
>On Thu, Feb 19, 1998 at 01:03:38PM +0100, Javier Fernandez-Sanguino Pen~a wrot
>> jvm1.0		Java virtual machines (I agree that programs for jvm1.0
>> jvm1.1		run with jvm1.1, but not viceversa)
>
>I thought we already had java-virtual-machine. guavac suggests
>java-virtual-machine, for example. I think I found some package which
>provided it when I put in the suggests.

!!
You are correct.   Thanks for pointing that out.

>> jdk1.0		Java compiler
>> jdk1.1
>
>ANy need to depend: on jdk1.0 or jdk1.1 yet though?

Sure.   I've been consider packaging jCVS, which would require 
java-virtual-machine and the classes from the 1.1 JDK.

However, I agree with Anand WRT naming.  I think we need the following.  
Proposed new virtual package names have + in the left-most column:

+  java-compiler-1.0
      conforms to java language spec 1.0
 + java-compiler-1.1
      conforms to java language spec 1.1 (yes, there is a different spec)

+  java-virtual-machine-1.0
     conforms to JVM specification 1.0.  This is generally couple with a 
JDK release but there's no reason why it has to be.  From the JVM 
spec:
| The Java Virtual Machine knows nothing of the Java programming language,
| only of a particular file format, the class file format. A class file
| contains Java Virtual Machine instructions (or bytecodes) and a symbol
| table, as well as other ancillary information. 
     (is there a JVM spec 1.1?  I have never seen one.)

+  java-classes-1.0
+  java-classes-1.1 
These are also called by JavaSoft "Java core classes", sometimes "Java
Platform Core API".  It conforms, i.e., in the latter case, to the "Java
Platform 1.1.5 Core API Specification" <URL:http:// java.sun.com:80/
products/jdk/1.1/docs/api/packages.html> This virtual package would be
provided by both the JDK and the JRE distributions.  This is to be
distinguished from the add-on packages, like the JFC, JDBC, etc.

The point is to have specifications-driven virtual package names.  I 
think this is the right abstraction to embrace.

Note that I may have gotten some details wrong here.  

..A. P. Harris...apharris@onShore.com...<URL:http://www.onShore.com/>



--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-devel-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .


Reply to: