Examples with InputStreamEntity org.apache.http.entity.InputStreamEntity used on opensource projects

Example 16 with InputStreamEntity

use of org.apache.http.entity.InputStreamEntity in project stanbol by apache.

the class RestfulLangidentEngine method computeEnhancements.

/**
     * Compute enhancements for supplied ContentItem. The results of the process
     * are expected to be stored in the metadata of the content item.
     * <p/>
     * The client (usually an {@link org.apache.stanbol.enhancer.servicesapi.EnhancementJobManager}) should take care of
     * persistent storage of the enhanced {@link org.apache.stanbol.enhancer.servicesapi.ContentItem}.
     * <p/>
     * This method creates a new POSContentPart using {@link org.apache.stanbol.enhancer.engines.pos.api.POSTaggerHelper#createContentPart} from a text/plain part and
     * stores it as a new part in the content item. The metadata is not changed.
     *
     * @throws org.apache.stanbol.enhancer.servicesapi.EngineException
     *          if the underlying process failed to work as
     *          expected
     */
@Override
public void computeEnhancements(final ContentItem ci) throws EngineException {
    //get the plain text Blob
    Map.Entry<IRI, Blob> textBlob = getPlainText(this, ci, false);
    Blob blob = textBlob.getValue();
    //send the text to the server
    final HttpPost request = new HttpPost(serviceUrl);
    request.setEntity(new InputStreamEntity(blob.getStream(), blob.getContentLength(), ContentType.create(blob.getMimeType(), blob.getParameter().get("charset"))));
    //execute the request
    List<LangSuggestion> detected;
    try {
        detected = AccessController.doPrivileged(new PrivilegedExceptionAction<List<LangSuggestion>>() {

            public List<LangSuggestion> run() throws ClientProtocolException, IOException {
                return httpClient.execute(request, new LangIdentResponseHandler(ci, objectMapper));
            }
        });
    } catch (PrivilegedActionException pae) {
        Exception e = pae.getException();
        if (e instanceof ClientProtocolException) {
            throw new EngineException(this, ci, "Exception while executing Request " + "on RESTful Language Identification Service at " + serviceUrl, e);
        } else if (e instanceof IOException) {
            throw new EngineException(this, ci, "Exception while executing Request " + "on RESTful Language Identification Service at " + serviceUrl, e);
        } else {
            throw RuntimeException.class.cast(e);
        }
    }
    Graph metadata = ci.getMetadata();
    log.debug("Detected Languages for ContentItem {} and Blob {}");
    ci.getLock().writeLock().lock();
    try {
        //write TextAnnotations for the detected languages
        for (LangSuggestion suggestion : detected) {
            // add a hypothesis
            log.debug(" > {}@{}", suggestion.getLanguage(), suggestion.hasProbability() ? suggestion.getProbability() : "-,--");
            IRI textEnhancement = EnhancementEngineHelper.createTextEnhancement(ci, this);
            metadata.add(new TripleImpl(textEnhancement, DC_LANGUAGE, new PlainLiteralImpl(suggestion.getLanguage())));
            metadata.add(new TripleImpl(textEnhancement, DC_TYPE, DCTERMS_LINGUISTIC_SYSTEM));
            if (suggestion.hasProbability()) {
                metadata.add(new TripleImpl(textEnhancement, ENHANCER_CONFIDENCE, literalFactory.createTypedLiteral(suggestion.getProbability())));
            }
        }
    } finally {
        ci.getLock().writeLock().unlock();
    }
}

Example 17 with InputStreamEntity

use of org.apache.http.entity.InputStreamEntity in project stanbol by apache.

the class TikaXmlDocumentOsgiTest method testDocx2.

/**
     * Tests docx format
     * engine 
     * @throws Exception
     */
@Test
public void testDocx2() throws Exception {
    InputStream in = EngineEnhancementRequestTest.class.getClassLoader().getResourceAsStream("Vorlage_Protokoll.docx");
    Assert.assertNotNull("Unable to find test resource 'Vorlage_Protokoll.docx'", in);
    executor.execute(builder.buildPostRequest(getEndpoint() + "/engine/tika?omitMetadata=true").withHeader("Accept", "text/plain").withHeader("Content-Type", "application/vnd.openxmlformats-officedocument.wordprocessingml.document").withEntity(new InputStreamEntity(in, -1))).assertStatus(//not interested in the results, just that it worked
    200);
}

Example 18 with InputStreamEntity

use of org.apache.http.entity.InputStreamEntity in project stanbol by apache.

the class EntityhubTest method buildImportRdfData.

/**
     * Imports/updates RDF data of the file to the entityhub with the possibility
     * to restrict imports/updates to the parsed uri
     * @param file the file with the RDF data (needs to be in the classpath)
     * @param create if <code>true</code> the data are created (POST) otherwise
     * updated (PUT). 
     * @param uri if not <code>null</code> only data of this URI are imported by
     * specifying the id parameter
     */
protected Request buildImportRdfData(InputStream in, String contentType, boolean create, String uri) {
    Assert.assertNotNull(in);
    Assert.assertNotNull(contentType);
    Request request;
    String path;
    if (uri != null) {
        path = builder.buildUrl("/entityhub/entity", "id", uri);
    } else {
        path = builder.buildUrl("/entityhub/entity");
    }
    if (create) {
        request = builder.buildOtherRequest(new HttpPost(path));
    } else {
        request = builder.buildOtherRequest(new HttpPut(path));
    }
    //set the HttpEntity (both PUT and POST are HttpEntityEnclosingRequests)
    ((HttpEntityEnclosingRequest) request.getRequest()).setEntity(new InputStreamEntity(in, -1));
    //finally set the correct content-type of the provided data
    //currently fixed to "application/rdf+xml"
    request.getRequest().setHeader("Content-Type", contentType);
    return request;
}

Example 19 with InputStreamEntity

use of org.apache.http.entity.InputStreamEntity in project stanbol by apache.

the class EngineEnhancementRequestTest method testPlainTextExtraction.

/**
     * Tests plain text extraction for an request directly sent to the tika 
     * engine 
     * @throws Exception
     */
@Test
public void testPlainTextExtraction() throws Exception {
    InputStream in = EngineEnhancementRequestTest.class.getClassLoader().getResourceAsStream("test.pdf");
    Assert.assertNotNull("Unable to find test resource 'test.pdf'", in);
    executor.execute(builder.buildPostRequest(getEndpoint() + "/engine/tika?omitMetadata=true").withHeader("Accept", "text/plain").withEntity(new InputStreamEntity(in, -1))).assertStatus(200).assertContentRegexp(//only that the Enhancer REST API works also with engines!
    "The Apache Stanbol Enhancer", "The Stanbol enhancer can detect famous cities such as Paris");
}

Example 20 with InputStreamEntity

use of org.apache.http.entity.InputStreamEntity in project stanbol by apache.

the class TikaXmlDocumentOsgiTest method testDocx.

/**
     * Tests docx format
     * engine 
     * @throws Exception
     */
@Test
public void testDocx() throws Exception {
    InputStream in = EngineEnhancementRequestTest.class.getClassLoader().getResourceAsStream("testWORD.docx");
    Assert.assertNotNull("Unable to find test resource 'testWORD.docx'", in);
    executor.execute(builder.buildPostRequest(getEndpoint() + "/engine/tika?omitMetadata=true").withHeader("Accept", "text/plain").withHeader("Content-Type", "application/vnd.openxmlformats-officedocument.wordprocessingml.document").withEntity(new InputStreamEntity(in, -1))).assertStatus(//not interested in the results, just that it worked
    200);
}