RFC: Make javax.jnlp package visible to javac
martinrb at google.com
Tue Jun 16 09:47:45 PDT 2009
The Java world has always made a bigger distinction
between interface and implementation than other projects,
especially since (most of) the interface is (at least in principle)
controlled by the JCP (this includes javax.* but not com.sun.*).
Moving an API to a more public location is as big a deal
as creating that new API in the first place.
Currently the process for making small (i.e. non-JSR)
changes to the JDK is in flux. In the past (and still at present)
one needs a Sun-internal sponsor to push such a change.
In future probably there will be a more public process that
anyone can participate in.
In the particular case of base64, this is a long-standing RFE.
An inhibitor may be the fact
that base64 is a byte-to-byte coder, while the JDK is used to
handling byte-to-char coders, which makes the API design
On Tue, Jun 16, 2009 at 02:05, Florian Weimer <fweimer at bfk.de> wrote:
> * Joe Darcy:
> > The ct.sym file is a mechanism used be javac for a variety of reasons,
> > including to serve as a stepping stone toward modules. Peter Ahe wrote
> > about it his blog entry:
> > http://blogs.sun.com/ahe/entry/b86_compiler_fixes
> Will you accept patches which move particularly useful functionality
> (the base64 decoder comes to my mind, or the X500Name class and other
> ASN.1-related helpers) to com.sun.* or javax.*?
> Florian Weimer <fweimer at bfk.de>
> BFK edv-consulting GmbH http://www.bfk.de/
> Kriegsstraße 100 tel: +49-721-96201-1
> D-76133 Karlsruhe fax: +49-721-96201-99
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the distro-pkg-dev