[Biojava-dev] biojava v2 alpha1 release
Michael Heuer
heuermh at acm.org
Wed May 5 16:39:48 EDT 2004
Got another one.
bash-2.03$ svn diff
Index: core/src/org/bjv2/util/WeakValueHashMap.java
===================================================================
--- core/src/org/bjv2/util/WeakValueHashMap.java (revision 26)
+++ core/src/org/bjv2/util/WeakValueHashMap.java (working copy)
@@ -53,16 +53,16 @@
extends AbstractMap<K,V>
{
private final Map<K, Reference<V>> keyToRefMap;
- private final ReferenceQueue queue;
+ private final ReferenceQueue<V> queue;
private final Set iteratorRefs;
- private final ReferenceQueue iteratorRefQueue;
+ private final ReferenceQueue<V> iteratorRefQueue;
public WeakValueHashMap()
{
keyToRefMap = new HashMap<K, Reference<V>>();
- queue = new ReferenceQueue();
+ queue = new ReferenceQueue<V>();
iteratorRefs = new HashSet();
- iteratorRefQueue = new ReferenceQueue();
+ iteratorRefQueue = new ReferenceQueue<V>();
}
private void diddleReferenceQueue()
On Wed, 5 May 2004, Michael Heuer wrote:
>
> fixed.
>
> bash-2.03$ svn diff
> Index: bytecode/src/org/bjv2/util/bytecode/IntrospectedCodeClass.java
> ===================================================================
> --- bytecode/src/org/bjv2/util/bytecode/IntrospectedCodeClass.java
> (revision 26)
> +++ bytecode/src/org/bjv2/util/bytecode/IntrospectedCodeClass.java
> (working copy)
> @@ -168,10 +168,10 @@
> private void initMethods()
> {
> if (_methods == null) {
> - Map meths = new HashMap();
> + Map meths = new HashMap<Object,CodeMethod>();
> popMeths(this.clazz, meths);
> popIMeths(this.clazz, meths);
> - _methods = new HashSet<CodeMethod>(meths.values());
> + _methods = new HashSet<CodeMethod>((Collection<CodeMethod>)
> meths.values());
> _methsByNameSig = new HashMap<Object, CodeMethod>();
> _methsByName = new HashMap<String, Set<CodeMethod>>();
> for(CodeMethod m: _methods) {
>
>
> On Wed, 5 May 2004, Michael Heuer wrote:
>
> > Hello Matthew,
> >
> > Neither the trunk nor the medusa branch is compiling for me right now,
> > with ant 1.6.1 and jdk 1.5.0-beta on solaris.
> >
> > [javac]
> > /home/heuer/working/bjv2/branches/medusa/bytecode/src/org/bjv2/util/bytecode/IntrospectedCodeClass.java:174:
> > cannot find symbol
> > [javac] symbol : constructor HashSet(java.util.Collection)
> > [javac] location: class
> > java.util.HashSet<org.bjv2.util.bytecode.CodeMethod>
> > [javac] _methods = new HashSet<CodeMethod>(meths.values());
> > [javac] ^
> >
> > A few questions -- what parts of the code are auto-generated, if
> > any? Do you have an OWL model for the API as we were discussing before?
> > I noticed some spelling errors (or at least uk/us english differences) in
> > some class and method names, would you like me to fix those? What are the
> > .iml files for?
> >
> > It's a lot to take in all at once, but I see some cool things right off
> > the bat -- the SPI design looks good, and the @NotNull and @Immutable
> > and so on annotations look quite useful.
> >
> > Unfortunately the SPI architecture/build process doesn't look well suited
> > to maven at first glance, I may have to grind on that for a while longer.
> >
> > michael
> >
> >
> > On Wed, 5 May 2004, Matthew Pocock wrote:
> >
> > > Hi,
> > >
> > > I have just committed the alpha1 release of biojava 2 (code-named
> > > medusa) to svn. To compile & run BJV2, you will need a 1.5 java sdk, SVN
> > > (1.0 or newer) and a reasonably up-to-date ANT. I'm releasing this
> > > revision under lGPL. I am happy to move to another open-source license
> > > if people feel it would be helpfull. GPL and OSL appear to be no good -
> > > to viral. BSD and friends are no good - too easy to split/pilfer. I'm
> > > reading through licences on http://www.opensource.org/licenses/ in the
> > > vain hope of finding something suitable.
> > >
> > > All bugs/questions either direct to me or to biojava-dev.
> > >
> > > Matthew
> > >
> > > ----
> > >
> > > Getting bjv2:
> > >
> > > the complete thing
> > > http://www.derkholm.net/svn/repos/bjv2
> > >
> > > development version
> > > http://www.derkholm.net/svn/repos/bjv2/trunk
> > >
> > > alpha1 release
> > > http://www.derkholm.net/svn/repos/bjv2/branches/medusa
> > >
> > > ----
> > >
> > > Things I want in the license:
> > >
> > > 1) "we" own the code - a 3rd party can't take it, change the formatting
> > > and stop us using it
> > >
> > > 2) "you" can incorporate the library into your app. this requries you to:
> > > acknowledge that you use us & what license you use us under
> > > provide access to the source - presumably as a link to our repository
> > >
> > > 3) if "you" have modified "our" source and then distribute the
> > > modification you must:
> > > notify the user that you've done this
> > > make the source of the modifications available under this license
> > > allow the modifications to be rolled back into the main copy of the
> > > code (we may chose not to though)
> > >
> > > 4) "you" can use example code without tainting your apps with our license
> > >
> > > 5) none of this lGPL linking rubbish - what's that all about anyway?
> > >
> > > 6) "we" are not liable for anything - we don't claim the code is fit for
> > > any purpose and if it craps out, you pay the consequences
> > >
> > > _______________________________________________
> > > biojava-dev mailing list
> > > biojava-dev at biojava.org
> > > http://biojava.org/mailman/listinfo/biojava-dev
> > >
> >
> > _______________________________________________
> > biojava-dev mailing list
> > biojava-dev at biojava.org
> > http://biojava.org/mailman/listinfo/biojava-dev
> >
>
>
More information about the biojava-dev
mailing list