use of org.apache.stanbol.enhancer.servicesapi.EngineException in project stanbol by apache.
the class KeywordLinkingEngine method computeEnhancements.
@Override
public void computeEnhancements(ContentItem ci) throws EngineException {
if (isOfflineMode() && !entitySearcher.supportsOfflineMode()) {
throw new EngineException("Offline mode is not supported by the Component used to lookup Entities");
}
Entry<IRI, Blob> contentPart = ContentItemHelper.getBlob(ci, SUPPORTED_MIMETYPES);
if (contentPart == null) {
throw new IllegalStateException("No ContentPart with a supported Mime Type" + "found for ContentItem " + ci.getUri() + "(supported: '" + SUPPORTED_MIMETYPES + "') -> this indicates that canEnhance was" + "NOT called and indicates a bug in the used EnhancementJobManager!");
}
String text;
try {
text = ContentItemHelper.getText(contentPart.getValue());
} catch (IOException e) {
throw new InvalidContentException(String.format("Unable to extract " + " text from ContentPart %s of ContentItem %s!", contentPart.getKey(), ci.getUri()), e);
}
if (text.trim().length() == 0) {
// TODO: make the length of the data a field of the ContentItem
// interface to be able to filter out empty items in the canEnhance
// method
log.warn("ContentPart {} of ContentItem does not contain any Text to extract knowledge from", contentPart.getKey(), ci);
return;
}
// Determine the language
String language;
ci.getLock().readLock().lock();
try {
language = extractLanguage(ci);
} finally {
ci.getLock().readLock().unlock();
}
if (isProcessableLanguages(language)) {
log.debug("computeEnhancements for ContentItem {} language {} text={}", new Object[] { ci.getUri().getUnicodeString(), language, StringUtils.abbreviate(text, 100) });
EntityLinker entityLinker = new EntityLinker(analysedContentFactory.create(text, language), entitySearcher, linkerConfig);
// process
entityLinker.process();
// write results (requires a write lock)
ci.getLock().writeLock().lock();
try {
writeEnhancements(ci, entityLinker.getLinkedEntities().values(), language);
} finally {
ci.getLock().writeLock().unlock();
}
} else {
log.debug("ignore ContentItem {} because language '{}' is not configured to" + "be processed by this engine.", ci.getUri().getUnicodeString(), language);
}
}
use of org.apache.stanbol.enhancer.servicesapi.EngineException in project stanbol by apache.
the class LanguageDetectionEnhancementEngine method computeEnhancements.
public void computeEnhancements(ContentItem ci) throws EngineException {
Entry<IRI, Blob> contentPart = ContentItemHelper.getBlob(ci, SUPPORTED_MIMTYPES);
if (contentPart == null) {
throw new IllegalStateException("No ContentPart with Mimetype '" + TEXT_PLAIN_MIMETYPE + "' found for ContentItem " + ci.getUri() + ": This is also checked in the canEnhance method! -> This " + "indicated an Bug in the implementation of the " + "EnhancementJobManager!");
}
String text = "";
try {
text = ContentItemHelper.getText(contentPart.getValue());
} catch (IOException e) {
throw new InvalidContentException(this, ci, e);
}
// do not call trim() on long texts to check if the text is empty
if (text.length() < 50 && text.trim().length() == 0) {
log.info("No text contained in ContentPart {} of ContentItem {}", contentPart.getKey(), ci.getUri());
return;
}
// truncate text to some piece from the middle if probeLength > 0
int checkLength = probeLength;
if (checkLength > 0 && text.length() > checkLength) {
text = text.substring(text.length() / 2 - checkLength / 2, text.length() / 2 + checkLength / 2);
}
List<Language> languages = null;
try {
languages = languageIdentifier.getLanguages(text);
log.debug("language identified: {}", languages);
} catch (LangDetectException e) {
Enum<?> errorCode = e.getCode();
// ignore " 0 - NoTextError" and "5 - CantDetectError"
if (errorCode.ordinal() != 0 && errorCode.ordinal() != 5) {
StringBuilder msg = new StringBuilder("Could not identify language of text: ");
if (text.length() < 200) {
msg.append(text);
} else {
msg.append(text.subSequence(0, 199)).append("...");
}
msg.append(" (Error Code: ").append(errorCode.ordinal()).append(" - ").append(errorCode.name()).append(")");
throw new EngineException(this, ci, msg.toString(), e);
} else {
log.debug("No text to detect the language from present in ContentItem ", ci);
}
}
// add language to metadata
if (languages != null) {
Graph g = ci.getMetadata();
ci.getLock().writeLock().lock();
try {
for (int i = 0; i < maxSuggestedLanguages && i < languages.size(); i++) {
// add a hypothesis
Language hypothesis = languages.get(i);
IRI textEnhancement = EnhancementEngineHelper.createTextEnhancement(ci, this);
g.add(new TripleImpl(textEnhancement, DC_LANGUAGE, new PlainLiteralImpl(hypothesis.lang)));
g.add(new TripleImpl(textEnhancement, ENHANCER_CONFIDENCE, literalFactory.createTypedLiteral(hypothesis.prob)));
g.add(new TripleImpl(textEnhancement, DC_TYPE, DCTERMS_LINGUISTIC_SYSTEM));
g.add(new TripleImpl(textEnhancement, ENHANCER_CONFIDENCE, literalFactory.createTypedLiteral(hypothesis.prob)));
}
} finally {
ci.getLock().writeLock().unlock();
}
}
}
use of org.apache.stanbol.enhancer.servicesapi.EngineException in project stanbol by apache.
the class TopicClassificationEngine method computeEnhancements.
@Override
public void computeEnhancements(ContentItem ci) throws EngineException {
Entry<IRI, Blob> contentPart = ContentItemHelper.getBlob(ci, SUPPORTED_MIMETYPES);
if (contentPart == null) {
throw new IllegalStateException("No ContentPart with a supported Mime Type" + "found for ContentItem " + ci.getUri() + "(supported: '" + SUPPORTED_MIMETYPES + "') -> this indicates that canEnhance was" + "NOT called and indicates a bug in the used EnhancementJobManager!");
}
String language = EnhancementEngineHelper.getLanguage(ci);
if (!(acceptedLanguageSet.isEmpty() || acceptedLanguageSet.contains(language) || acceptedLanguageSet.contains(""))) {
throw new IllegalStateException("The language '" + language + "' of the ContentItem is not configured as " + " active for this Engine (active: " + acceptedLanguageSet + ").");
}
String text;
try {
text = ContentItemHelper.getText(contentPart.getValue());
} catch (IOException e) {
throw new InvalidContentException(String.format("Unable to extract " + " textual content from ContentPart %s of ContentItem %s!", contentPart.getKey(), ci.getUri()), e);
}
if (text.trim().isEmpty()) {
log.warn("ContentPart {} of ContentItem {} does not contain any " + "text to extract topics from", contentPart.getKey(), ci.getUri());
return;
}
Graph metadata = ci.getMetadata();
List<TopicSuggestion> topics;
try {
topics = suggestTopics(text);
if (topics.isEmpty()) {
return;
}
} catch (ClassifierException e) {
throw new EngineException(e);
}
IRI precision = new IRI(NamespaceEnum.fise + "classifier/precision");
IRI recall = new IRI(NamespaceEnum.fise + "classifier/recall");
IRI f1 = new IRI(NamespaceEnum.fise + "classifier/f1");
LiteralFactory lf = LiteralFactory.getInstance();
ci.getLock().writeLock().lock();
try {
// Global text annotation to attach all the topic annotation to it.
IRI textAnnotation = EnhancementEngineHelper.createTextEnhancement(ci, this);
metadata.add(new TripleImpl(textAnnotation, org.apache.stanbol.enhancer.servicesapi.rdf.Properties.DC_TYPE, OntologicalClasses.SKOS_CONCEPT));
for (TopicSuggestion topic : topics) {
IRI enhancement = EnhancementEngineHelper.createEntityEnhancement(ci, this);
metadata.add(new TripleImpl(enhancement, org.apache.stanbol.enhancer.servicesapi.rdf.Properties.RDF_TYPE, TechnicalClasses.ENHANCER_TOPICANNOTATION));
metadata.add(new TripleImpl(enhancement, org.apache.stanbol.enhancer.servicesapi.rdf.Properties.DC_RELATION, textAnnotation));
// add link to entity
metadata.add(new TripleImpl(enhancement, org.apache.stanbol.enhancer.servicesapi.rdf.Properties.ENHANCER_ENTITY_REFERENCE, new IRI(topic.conceptUri)));
metadata.add(new TripleImpl(enhancement, org.apache.stanbol.enhancer.servicesapi.rdf.Properties.ENHANCER_ENTITY_TYPE, OntologicalClasses.SKOS_CONCEPT));
// add confidence information
metadata.add(new TripleImpl(enhancement, org.apache.stanbol.enhancer.servicesapi.rdf.Properties.ENHANCER_CONFIDENCE, lf.createTypedLiteral(Double.valueOf(topic.score))));
// add performance estimates of the classifier if available
ClassificationReport perf = getPerformanceEstimates(topic.conceptUri);
if (perf.uptodate) {
metadata.add(new TripleImpl(enhancement, precision, lf.createTypedLiteral(Double.valueOf(perf.precision))));
metadata.add(new TripleImpl(enhancement, recall, lf.createTypedLiteral(Double.valueOf(perf.recall))));
metadata.add(new TripleImpl(enhancement, f1, lf.createTypedLiteral(Double.valueOf(perf.f1))));
}
// fetch concept label from the entityhub or a referenced site if available
Entity entity = entityhub.getEntity(topic.conceptUri);
if (entity == null) {
entity = referencedSiteManager.getEntity(topic.conceptUri);
}
if (entity != null) {
Representation representation = entity.getRepresentation();
// TODO: extract all languages based on some configuration instead of hardcoding English
Text label = representation.getFirst(NamespaceEnum.skos + "prefLabel", "en", "en-US", "en-GB");
if (label == null) {
label = representation.getFirst(NamespaceEnum.rdfs + "label", "en", "en-US", "en-GB");
}
if (label != null) {
metadata.add(new TripleImpl(enhancement, org.apache.stanbol.enhancer.servicesapi.rdf.Properties.ENHANCER_ENTITY_LABEL, new PlainLiteralImpl(label.getText())));
}
}
}
} catch (ClassifierException e) {
throw new EngineException(e);
} catch (IllegalArgumentException e) {
throw new EngineException(e);
} catch (EntityhubException e) {
throw new EngineException(e);
} finally {
ci.getLock().writeLock().unlock();
}
}
use of org.apache.stanbol.enhancer.servicesapi.EngineException in project stanbol by apache.
the class ZemantaEnhancementEngine method computeEnhancements.
public void computeEnhancements(ContentItem ci) throws EngineException {
Entry<IRI, Blob> contentPart = ContentItemHelper.getBlob(ci, SUPPORTED_MIMETYPES);
if (contentPart == null) {
throw new IllegalStateException("No ContentPart with a supported Mime Type" + "found for ContentItem " + ci.getUri() + "(supported: '" + SUPPORTED_MIMETYPES + "') -> this indicates that canEnhance was" + "NOT called and indicates a bug in the used EnhancementJobManager!");
}
String text;
try {
text = ContentItemHelper.getText(contentPart.getValue());
} catch (IOException e) {
throw new InvalidContentException(this, ci, e);
}
if (text.trim().length() == 0) {
log.warn("ContentPart {} of ContentItem {} does not contain any text to enhance", contentPart.getKey(), ci.getUri());
return;
}
Graph graph = ci.getMetadata();
IRI ciId = ci.getUri();
// we need to store the results of Zemanta in an temp graph
Graph results = new SimpleGraph();
ZemantaAPIWrapper zemanta = new ZemantaAPIWrapper(key);
try {
results.addAll(zemanta.enhance(text));
} catch (IOException e) {
throw new EngineException("Unable to get Enhancement from remote Zemanta Service", e);
}
// now we need to process the results and convert them into the Enhancer
// annotation structure
ci.getLock().writeLock().lock();
try {
processRecognition(results, graph, text, ciId);
processCategories(results, graph, ciId);
} finally {
ci.getLock().writeLock().unlock();
}
}
use of org.apache.stanbol.enhancer.servicesapi.EngineException in project stanbol by apache.
the class SmartcnTokenizerEngine 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(ContentItem ci) throws EngineException {
final AnalysedText at = initAnalysedText(this, analysedTextFactory, ci);
String language = getLanguage(this, ci, false);
if (!("zh".equals(language) || (language != null && language.startsWith("zh-")))) {
throw new IllegalStateException("The detected language is NOT 'zh'! " + "As this is also checked within the #canEnhance(..) method this " + "indicates an Bug in the used EnhancementJobManager implementation. " + "Please report this on the dev@apache.stanbol.org or create an " + "JIRA issue about this.");
}
if (!at.getSentences().hasNext()) {
// no sentences ... use this engine to detect
// first the sentences
TokenStream sentences = new SentenceTokenizer(new CharSequenceReader(at.getText()));
try {
while (sentences.incrementToken()) {
OffsetAttribute offset = sentences.addAttribute(OffsetAttribute.class);
Sentence s = at.addSentence(offset.startOffset(), offset.endOffset());
if (log.isTraceEnabled()) {
log.trace("detected {}:{}", s, s.getSpan());
}
}
} catch (IOException e) {
String message = String.format("IOException while reading from " + "CharSequenceReader of AnalyzedText for ContentItem %s", ci.getUri());
log.error(message, e);
throw new EngineException(this, ci, message, e);
}
}
// now the tokens
TokenStream tokens = new WordTokenFilter(new AnalyzedTextSentenceTokenizer(at));
try {
tokens.reset();
while (tokens.incrementToken()) {
OffsetAttribute offset = tokens.addAttribute(OffsetAttribute.class);
Token t = at.addToken(offset.startOffset(), offset.endOffset());
log.trace("detected {}", t);
}
} catch (IOException e) {
String message = String.format("IOException while reading from " + "CharSequenceReader of AnalyzedText for ContentItem %s", ci.getUri());
log.error(message, e);
throw new EngineException(this, ci, message, e);
}
}
Aggregations