Examples with ExtractorException org.semanticdesktop.aperture.extractor.ExtractorException used on opensource projects

Example 6 with ExtractorException

use of org.semanticdesktop.aperture.extractor.ExtractorException in project stanbol by apache.

the class MetaxaEngine method computeEnhancements.

public void computeEnhancements(ContentItem ci) throws EngineException {
    // get model from the extraction
    URIImpl docId;
    Model m = null;
    ci.getLock().readLock().lock();
    try {
        docId = new URIImpl(ci.getUri().getUnicodeString());
        m = this.extractor.extract(ci.getStream(), docId, ci.getMimeType());
    } catch (ExtractorException e) {
        throw new EngineException("Error while processing ContentItem " + ci.getUri() + " with Metaxa", e);
    } catch (IOException e) {
        throw new EngineException("Error while processing ContentItem " + ci.getUri() + " with Metaxa", e);
    } finally {
        ci.getLock().readLock().unlock();
    }
    // the extracted plain text from the model
    if (null == m) {
        log.debug("Unable to preocess ContentItem {} (mime type {}) with Metaxa", ci.getUri(), ci.getMimeType());
        return;
    }
    ContentSink plainTextSink;
    try {
        plainTextSink = ciFactory.createContentSink("text/plain");
    } catch (IOException e) {
        m.close();
        throw new EngineException("Unable to initialise Blob for storing" + "the plain text content", e);
    }
    HashMap<BlankNode, BlankNode> blankNodeMap = new HashMap<BlankNode, BlankNode>();
    RDF2GoUtils.urifyBlankNodes(m);
    ClosableIterator<Statement> it = m.iterator();
    BufferedWriter out = new BufferedWriter(new OutputStreamWriter(plainTextSink.getOutputStream(), UTF8));
    // used to detect if some text was extracted
    boolean textExtracted = false;
    try {
        // first add to a temporary graph
        Graph g = new SimpleGraph();
        while (it.hasNext()) {
            Statement oneStmt = it.next();
            // the plain text Blob!
            if (oneStmt.getSubject().equals(docId) && oneStmt.getPredicate().equals(NIE_PLAINTEXT_PROPERTY)) {
                String text = oneStmt.getObject().toString();
                if (text != null && !text.isEmpty()) {
                    try {
                        out.write(oneStmt.getObject().toString());
                    } catch (IOException e) {
                        throw new EngineException("Unable to write extracted" + "plain text to Blob (blob impl: " + plainTextSink.getBlob().getClass() + ")", e);
                    }
                    textExtracted = true;
                    if (includeText) {
                        BlankNodeOrIRI subject = (BlankNodeOrIRI) asClerezzaResource(oneStmt.getSubject(), blankNodeMap);
                        IRI predicate = (IRI) asClerezzaResource(oneStmt.getPredicate(), blankNodeMap);
                        RDFTerm object = asClerezzaResource(oneStmt.getObject(), blankNodeMap);
                        g.add(new TripleImpl(subject, predicate, object));
                    }
                }
            } else {
                // add metadata to the metadata of the contentItem
                BlankNodeOrIRI subject = (BlankNodeOrIRI) asClerezzaResource(oneStmt.getSubject(), blankNodeMap);
                IRI predicate = (IRI) asClerezzaResource(oneStmt.getPredicate(), blankNodeMap);
                RDFTerm object = asClerezzaResource(oneStmt.getObject(), blankNodeMap);
                if (null != subject && null != predicate && null != object) {
                    Triple t = new TripleImpl(subject, predicate, object);
                    g.add(t);
                    log.debug("added " + t.toString());
                }
            }
        }
        // add the extracted triples to the metadata of the ContentItem
        ci.getLock().writeLock().lock();
        try {
            ci.getMetadata().addAll(g);
            g = null;
        } finally {
            ci.getLock().writeLock().unlock();
        }
    } finally {
        it.close();
        m.close();
        IOUtils.closeQuietly(out);
    }
    if (textExtracted) {
        // add plain text to the content item
        IRI blobUri = new IRI("urn:metaxa:plain-text:" + randomUUID());
        ci.addPart(blobUri, plainTextSink.getBlob());
    }
}

Example 7 with ExtractorException

use of org.semanticdesktop.aperture.extractor.ExtractorException in project stanbol by apache.

the class XsltExtractor method extract.

public synchronized void extract(String id, Document doc, Map<String, Object> params, RDFContainer result) throws ExtractorException {
    if (params == null) {
        params = new HashMap<String, Object>();
    }
    params.put(this.uriParameter, id);
    initTransformerParameters(params);
    Source source = new DOMSource(doc);
    StringWriter writer = new StringWriter();
    StreamResult output = new StreamResult(writer);
    try {
        this.transformer.transform(source, output);
        String rdf = writer.toString();
        LOG.debug(rdf);
        StringReader reader = new StringReader(rdf);
        result.getModel().readFrom(reader, this.syntax);
        reader.close();
    } catch (TransformerException e) {
        throw new ExtractorException(e.getMessage(), e);
    } catch (ModelRuntimeException e) {
        throw new ExtractorException(e.getMessage(), e);
    } catch (IOException e) {
        throw new ExtractorException(e.getMessage(), e);
    }
}