Hi Claude,
Thanks for the advice - your Comparator-based solution is cleaner,
compact
and powerful. We understand utilizing the Comparator is always better
way
to deal with any sorting needs, as one can customize the detail
process.
BTW, the HashMap is for table-lookup to find a Resource with the
toString()
name, and to provide the set of Resource names to the TreeSet. The
hashing
mechanism doesn't have anything to do with the actual name sorting to
order
the Resources.
Best
Chan
Chan,
You are using a HashMap, this means that the objects will be retrieved
in
hashCode order. Not what you want.
Look at the java.util.Comparator interface. Implement one of those
that
extracts the name for comparison and returns the results in the order
you
want.
something like:
new Comparator<Resource>(){
public int compare(Resource o1, Resource o2) {
return -1 * o1.toString().compareTo( o2.toString() );
}
};
You then pass that implementation to the TreeSet constructor. This
will
yield a sorted map with the items in the order you want. Then just
call
the iterator.
Claude
On Fri, Sep 13, 2013 at 3:21 AM, <[email protected]> wrote:
Hi Claude,
Thanks for your guidance - it worked nicely.
I made a couple of changes to make it as I wanted:
1) Need to sort using the name of the subject resources,
not the resource itself. Have to build a map to do that
2) Need to take iterator in descending order after sort,
to have the named resources come up first
The changed & tested code:
------------------------------**------------------------------**-------
public void writeRDFStatements(Model model, PrintWriter writer) {
Map<String, Resource> subjMap = new HashMap<String, Resource>();
ResIterator subjIter = model.listSubjects();
while (subjIter.hasNext()) {
Resource subj = subjIter.next();
Resource typeDefn = subj.getPropertyResourceValue(**RDF.type);
subjMap.put(subj.toString(), subj);
}
TreeSet<String> subjs = new TreeSet<String>(subjMap.**keySet());
Iterator<String> rIter = subjs.descendingIterator();
while (rIter.hasNext())
writeRDFStatements(model, subjMap.get(rIter.next()), writer);
}
------------------------------**------------------------------**-------
Chan
something like
protected void writeRDFStatements( Model model, PrintWriter writer )
{
Set<Resource> subjects = new TreeSet<Resource>();
subjects.addAll( model.listSubjects().asSet() );
Iterator<Resource> rIter = subjects.iterator()
while (rIter.hasNext()) writeRDFStatements( model, rIter.nex(),
writer );
}
Might do the trick, assuming that you want the statements in
"natural"
order and that you don't have so many as to overflow memory.
If you want a different order you could implement a comparator and
pass it
to the TreeSet constructor.
Claude
On Thu, Sep 12, 2013 at 8:09 AM, <[email protected]> wrote:
Hi Claude,
It's close as this code at least group the triples
per subject basis and the "model.write(stream, lang)"
prints out in that manner. Still, there's no explicit
ordering in the subjects (other than the iterator's
own implicit one). Looks like I need to write my own.
Thanks
Chan
The code you are looking for is probably in
com.hp.hpl.jena.xmloutput.****impl.Basic.java
protected void writeRDFStatements( Model model, PrintWriter
writer )
{
ResIterator rIter = model.listSubjects();
while (rIter.hasNext()) writeRDFStatements( model,
rIter.nextResource(),
writer );
}
protected void writeRDFStatements
( Model model, Resource subject, PrintWriter writer )
{
StmtIterator sIter = model.listStatements( subject, null,
(RDFNode) null
);
writeDescriptionHeader( subject, writer );
while (sIter.hasNext()) writePredicate( sIter.nextStatement(),
writer );
writeDescriptionTrailer( subject, writer );
}
However, it just prints the statements out in the order the model
returns
them by listSubjects() and listStatements() for each subject
returned.
So
as Chris said the order is model implementation specific.
Claude
On Thu, Sep 12, 2013 at 7:01 AM, <[email protected]> wrote:
On 2013-09-11 23:15, Chris_Dollin wrote:
On Wednesday, September 11, 2013 07:28:33 PM
[email protected]:
Is there a Jena stmt/triple iterator feeding items in order ?
I looked around the Jena API list to find, but no luck.
However,
the 'Model.write(outputStream, lang)' method produces triple
listing in hierarchy with indentation - wonder how the method
produces such listing.
Is Model.listStatements() what you're looking for?
[Note: there is no "in order"; the statements come out in some
unspecified implementation-specific order.]
Chris
What I'm looking for is another iterator (or a Jena utility ?)
which can order the triples in hierarchy - not in random sequence
from 'listStatements()'. I just need to traverse a resource from
the top to the bottom, and it'd be nice to have an iterator to
feed
the triples that way. The model.write() produces triples in
sequence
and I just want to utilize that ordering mechanism.
Chan
