Examples with MultipartEntityBuilder org.apache.http.entity.mime.MultipartEntityBuilder used on opensource projects

Example 21 with MultipartEntityBuilder

use of org.apache.http.entity.mime.MultipartEntityBuilder in project stanbol by apache.

the class MultipartRequestTest method testUploadMultipleContents.

/**
     * This uploads the HTML as well as the plain text version of an content.
     * This allows it CMS to parse already available alternate content versions
     * in a single request. Stanbol can than still use the original content
     * (e.g. to extract metadata) but other engines that require the alternate
     * version (e.g. plain text version) of an document will directly use the
     * parsed version .<p>
     * This UnitTest ensures this by adding a "secret" extension the to plain
     * text version and than checks if the two entities mentioned in that
     * part are included in the extracted entities.
     * @throws IOException
     */
@Test
public void testUploadMultipleContents() throws IOException {
    //It is a secret, that Berlin is the capital of Germany
    String extraTextConent = TEXT_CONTENT + "\nIt is a secret, that the city of Berlin is the capital of Germany since 1990.";
    //The multipartBuilder used to construct the contentItem for the contentItem
    MultipartEntityBuilder ciBuilder = MultipartEntityBuilder.create();
    String boundary = "contentItem-47jjksnbue73fnis";
    ciBuilder.setBoundary(boundary);
    //use a small extension to deal with multipart/alternate
    Map<String, ContentBody> alternates = new LinkedHashMap<String, ContentBody>();
    alternates.put("http://www.example.com/test.html", new StringBody(HTML_CONTENT, ContentType.TEXT_HTML.withCharset(UTF8)));
    alternates.put("http://www.example.com/test.txt", new StringBody(extraTextConent, ContentType.TEXT_PLAIN.withCharset(UTF8)));
    ciBuilder.addPart("content", new MultipartContentBody(alternates, "contentParts", ContentType.create("multipart/alternate")));
    String receivedContent = executor.execute(builder.buildPostRequest(getEndpoint()).withHeader("Accept", "text/rdf+nt").withEntity(ciBuilder.build())).assertStatus(200).assertContentRegexp(//and the expected enhancements in the metadata
    "http://purl.org/dc/terms/creator.*LanguageDetectionEnhancementEngine", "http://purl.org/dc/terms/language.*en", "http://fise.iks-project.eu/ontology/entity-label.*Paris", "http://purl.org/dc/terms/creator.*org.apache.stanbol.enhancer.engines.opennlp.*NamedEntityExtractionEnhancementEngine", "http://fise.iks-project.eu/ontology/entity-label.*Bob Marley", //check also for expeted entities extracted from the secret Text part!
    "http://fise.iks-project.eu/ontology/entity-label.*Berlin", "http://fise.iks-project.eu/ontology/entity-label.*Germany").getContent();
    log.debug("Content:\n{}\n", receivedContent);
}

Example 22 with MultipartEntityBuilder

use of org.apache.http.entity.mime.MultipartEntityBuilder in project stanbol by apache.

the class MultipartRequestTest method testUploadWithMetadata.

/**
     * Stanbol also supports to upload pre-existing metadata with the content.
     * This UnitTest uses an example that parsed TextAnnotations for free text
     * tags provided by users that are than linked to Entities in DBPedia
     * @throws IOException
     */
@Test
public void testUploadWithMetadata() throws IOException {
    //create the metadata
    RDFTerm user = new PlainLiteralImpl("Rupert Westenthaler");
    final IRI contentItemId = new IRI("http://www.example.com/test.html");
    Graph metadata = new SimpleGraph();
    addTagAsTextAnnotation(metadata, contentItemId, "Germany", DBPEDIA_PLACE, user);
    addTagAsTextAnnotation(metadata, contentItemId, "Europe", DBPEDIA_PLACE, user);
    addTagAsTextAnnotation(metadata, contentItemId, "NATO", DBPEDIA_ORGANISATION, user);
    addTagAsTextAnnotation(metadata, contentItemId, "Silvio Berlusconi", DBPEDIA_PERSON, user);
    String rdfContentType = SupportedFormat.RDF_XML;
    ByteArrayOutputStream out = new ByteArrayOutputStream();
    serializer.serialize(out, metadata, rdfContentType);
    String rdfContent = new String(out.toByteArray(), UTF8);
    MultipartEntityBuilder ciBuilder = MultipartEntityBuilder.create();
    //add the metadata
    /*
         * NOTE: We need here to override the getFilename, because this MUST
         *       BE the URI of the ContentItem. This is important, because the
         *       Metadata do contain triples about that ContentItem and therefore
         *       it MUST BE assured that the URI of the ContentItem created by
         *       the Stanbol Enhancer is the same of as the URI used in the
         *       Metadata!
         */
    ciBuilder.addPart("metadata", new StringBody(rdfContent, ContentType.create(rdfContentType).withCharset(UTF8)) {

        @Override
        public String getFilename() {
            //uri of the ContentItem
            return contentItemId.getUnicodeString();
        }
    });
    //add the content
    ciBuilder.addTextBody("content", HTML_CONTENT, ContentType.TEXT_HTML.withCharset(UTF8));
    //send the request
    String receivedContent = executor.execute(builder.buildPostRequest(getEndpoint()).withHeader("Accept", "text/rdf+nt").withEntity(ciBuilder.build())).assertStatus(200).assertContentRegexp(//and the expected enhancements based on the parsed content
    "http://purl.org/dc/terms/creator.*LanguageDetectionEnhancementEngine", "http://purl.org/dc/terms/language.*en", "http://fise.iks-project.eu/ontology/entity-label.*Paris", "http://purl.org/dc/terms/creator.*org.apache.stanbol.enhancer.engines.opennlp.*NamedEntityExtractionEnhancementEngine", "http://fise.iks-project.eu/ontology/entity-label.*Bob Marley", //additional enhancements based on parsed metadata
    "http://fise.iks-project.eu/ontology/entity-reference.*http://dbpedia.org/resource/Germany.*", "http://fise.iks-project.eu/ontology/entity-reference.*http://dbpedia.org/resource/NATO.*", "http://fise.iks-project.eu/ontology/entity-reference.*http://dbpedia.org/resource/Silvio_Berlusconi.*", "http://fise.iks-project.eu/ontology/entity-reference.*http://dbpedia.org/resource/Europe.*").getContent();
    log.debug("Content:\n{}\n", receivedContent);
}

Example 23 with MultipartEntityBuilder

use of org.apache.http.entity.mime.MultipartEntityBuilder in project stanbol by apache.

the class MultipartRequestTest method testMissingContent.

@Test
public void testMissingContent() throws IOException {
    final IRI contentItemId = new IRI("http://www.example.com/test.html");
    String rdfContentType = SupportedFormat.RDF_XML;
    String rdfContent = getDummyRdfMetadata(contentItemId, rdfContentType);
    MultipartEntityBuilder ciBuilder = MultipartEntityBuilder.create();
    ciBuilder.addTextBody("metadata", rdfContent, ContentType.create(rdfContentType, UTF8));
    String receivedContent = executor.execute(builder.buildPostRequest(getEndpoint()).withHeader("Accept", "text/rdf+nt").withEntity(ciBuilder.build())).assertStatus(//BAD request
    400).getContent();
    //check also the error message
    Assert.assertTrue(receivedContent.contains("The parsed multipart content item does not contain any content."));
}

Example 24 with MultipartEntityBuilder

use of org.apache.http.entity.mime.MultipartEntityBuilder in project stanbol by apache.

the class MultipartRequestTest method testContentBeforeMetadata.

@Test
public void testContentBeforeMetadata() throws IOException {
    final IRI contentItemId = new IRI("http://www.example.com/test.html");
    String rdfContentType = SupportedFormat.RDF_XML;
    String rdfContent = getDummyRdfMetadata(contentItemId, rdfContentType);
    MultipartEntityBuilder ciBuilder = MultipartEntityBuilder.create();
    ciBuilder.addTextBody("content", HTML_CONTENT, ContentType.TEXT_HTML.withCharset(UTF8));
    ciBuilder.addTextBody("metadata", rdfContent, ContentType.create(rdfContentType, UTF8));
    String receivedContent = executor.execute(builder.buildPostRequest(getEndpoint()).withHeader("Accept", "text/rdf+nt").withEntity(ciBuilder.build())).assertStatus(//BAD request
    400).getContent();
    //check also the error message
    Assert.assertTrue(receivedContent.contains("The Multipart MIME part with the 'metadata' MUST BE before the " + "MIME part containing the 'content'"));
}