use of org.apache.tika.sax.LinkContentHandler in project tika by apache.
the class RollbackSoftware method rollback.
public void rollback(File deployArea) throws IOException, SAXException, TikaException {
LinkContentHandler handler = new LinkContentHandler();
Metadata met = new Metadata();
DeploymentAreaParser parser = new DeploymentAreaParser();
parser.parse(IOUtils.toInputStream(deployArea.getAbsolutePath(), UTF_8), handler, met);
List<Link> links = handler.getLinks();
if (links.size() < 2)
throw new IOException("Must have installed at least 2 versions!");
Collections.sort(links, new Comparator<Link>() {
public int compare(Link o1, Link o2) {
return o1.getText().compareTo(o2.getText());
}
});
this.updateVersion(links.get(links.size() - 2).getText());
}
use of org.apache.tika.sax.LinkContentHandler in project tika by apache.
the class HtmlParserTest method testCustomHtmlSchema.
// TIKA-1193
@Test
public void testCustomHtmlSchema() throws Exception {
// Default schema does not allow tables inside anchors
String test = "<html><body><a><table><tr><td>text</tr></tr></table></a></body></html>";
Metadata metadata = new Metadata();
LinkContentHandler linkContentHandler = new LinkContentHandler();
new HtmlParser().parse(new ByteArrayInputStream(test.getBytes(ISO_8859_1)), linkContentHandler, metadata, new ParseContext());
// Expect no anchor text
assertEquals("", linkContentHandler.getLinks().get(0).getText());
// We'll change the schema to allow tables inside anchors!
Schema schema = new HTMLSchema();
schema.elementType("a", HTMLSchema.M_ANY, 65535, 0);
ParseContext parseContext = new ParseContext();
parseContext.set(Schema.class, schema);
linkContentHandler = new LinkContentHandler();
new HtmlParser().parse(new ByteArrayInputStream(test.getBytes(ISO_8859_1)), linkContentHandler, metadata, parseContext);
// Expect anchor text
assertEquals("\ttext\n\n", linkContentHandler.getLinks().get(0).getText());
}
use of org.apache.tika.sax.LinkContentHandler in project tika by apache.
the class TIAParsingExample method testTeeContentHandler.
public static void testTeeContentHandler(String filename) throws Exception {
InputStream stream = new ByteArrayInputStream(new byte[0]);
Metadata metadata = new Metadata();
ParseContext context = new ParseContext();
Parser parser = new AutoDetectParser();
LinkContentHandler linkCollector = new LinkContentHandler();
try (OutputStream output = new FileOutputStream(new File(filename))) {
ContentHandler handler = new TeeContentHandler(new BodyContentHandler(output), linkCollector);
parser.parse(stream, handler, metadata, context);
}
}
use of org.apache.tika.sax.LinkContentHandler in project nutch by apache.
the class TikaParser method getParse.
@SuppressWarnings("deprecation")
public ParseResult getParse(Content content) {
String mimeType = content.getContentType();
boolean useBoilerpipe = getConf().get("tika.extractor", "none").equals("boilerpipe");
String boilerpipeExtractorName = getConf().get("tika.extractor.boilerpipe.algorithm", "ArticleExtractor");
URL base;
try {
base = new URL(content.getBaseUrl());
} catch (MalformedURLException e) {
return new ParseStatus(e).getEmptyParseResult(content.getUrl(), getConf());
}
// get the right parser using the mime type as a clue
Parser parser = tikaConfig.getParser(MediaType.parse(mimeType));
byte[] raw = content.getContent();
if (parser == null) {
String message = "Can't retrieve Tika parser for mime-type " + mimeType;
LOG.error(message);
return new ParseStatus(ParseStatus.FAILED, message).getEmptyParseResult(content.getUrl(), getConf());
}
LOG.debug("Using Tika parser " + parser.getClass().getName() + " for mime-type " + mimeType);
Metadata tikamd = new Metadata();
HTMLDocumentImpl doc = new HTMLDocumentImpl();
doc.setErrorChecking(false);
DocumentFragment root = doc.createDocumentFragment();
ContentHandler domHandler;
// Check whether to use Tika's BoilerplateContentHandler
if (useBoilerpipe) {
BoilerpipeContentHandler bpHandler = new BoilerpipeContentHandler((ContentHandler) new DOMBuilder(doc, root), BoilerpipeExtractorRepository.getExtractor(boilerpipeExtractorName));
bpHandler.setIncludeMarkup(true);
domHandler = (ContentHandler) bpHandler;
} else {
DOMBuilder domBuilder = new DOMBuilder(doc, root);
domBuilder.setUpperCaseElementNames(upperCaseElementNames);
domBuilder.setDefaultNamespaceURI(XHTMLContentHandler.XHTML);
domHandler = (ContentHandler) domBuilder;
}
LinkContentHandler linkContentHandler = new LinkContentHandler();
ParseContext context = new ParseContext();
TeeContentHandler teeContentHandler = new TeeContentHandler(domHandler, linkContentHandler);
if (HTMLMapper != null)
context.set(HtmlMapper.class, HTMLMapper);
tikamd.set(Metadata.CONTENT_TYPE, mimeType);
try {
parser.parse(new ByteArrayInputStream(raw), (ContentHandler) teeContentHandler, tikamd, context);
} catch (Exception e) {
LOG.error("Error parsing " + content.getUrl(), e);
return new ParseStatus(ParseStatus.FAILED, e.getMessage()).getEmptyParseResult(content.getUrl(), getConf());
}
HTMLMetaTags metaTags = new HTMLMetaTags();
String text = "";
String title = "";
Outlink[] outlinks = new Outlink[0];
org.apache.nutch.metadata.Metadata nutchMetadata = new org.apache.nutch.metadata.Metadata();
// we have converted the sax events generated by Tika into a DOM object
// so we can now use the usual HTML resources from Nutch
// get meta directives
HTMLMetaProcessor.getMetaTags(metaTags, root, base);
if (LOG.isTraceEnabled()) {
LOG.trace("Meta tags for " + base + ": " + metaTags.toString());
}
// check meta directives
if (!metaTags.getNoIndex()) {
// okay to index
StringBuffer sb = new StringBuffer();
if (LOG.isTraceEnabled()) {
LOG.trace("Getting text...");
}
// extract text
utils.getText(sb, root);
text = sb.toString();
sb.setLength(0);
if (LOG.isTraceEnabled()) {
LOG.trace("Getting title...");
}
// extract title
utils.getTitle(sb, root);
title = sb.toString().trim();
}
if (!metaTags.getNoFollow()) {
// okay to follow links
// extract outlinks
ArrayList<Outlink> l = new ArrayList<Outlink>();
URL baseTag = base;
String baseTagHref = tikamd.get("Content-Location");
if (baseTagHref != null) {
try {
baseTag = new URL(base, baseTagHref);
} catch (MalformedURLException e) {
LOG.trace("Invalid <base href=\"{}\">", baseTagHref);
}
}
if (LOG.isTraceEnabled()) {
LOG.trace("Getting links (base URL = {}) ...", baseTag);
}
// pre-1233 outlink extraction
// utils.getOutlinks(baseTag != null ? baseTag : base, l, root);
// Get outlinks from Tika
List<Link> tikaExtractedOutlinks = linkContentHandler.getLinks();
utils.getOutlinks(baseTag, l, tikaExtractedOutlinks);
outlinks = l.toArray(new Outlink[l.size()]);
if (LOG.isTraceEnabled()) {
LOG.trace("found " + outlinks.length + " outlinks in " + content.getUrl());
}
}
// populate Nutch metadata with Tika metadata
String[] TikaMDNames = tikamd.names();
for (String tikaMDName : TikaMDNames) {
if (tikaMDName.equalsIgnoreCase(Metadata.TITLE))
continue;
String[] values = tikamd.getValues(tikaMDName);
for (String v : values) nutchMetadata.add(tikaMDName, v);
}
if (outlinks.length == 0) {
outlinks = OutlinkExtractor.getOutlinks(text, getConf());
}
ParseStatus status = new ParseStatus(ParseStatus.SUCCESS);
if (metaTags.getRefresh()) {
status.setMinorCode(ParseStatus.SUCCESS_REDIRECT);
status.setArgs(new String[] { metaTags.getRefreshHref().toString(), Integer.toString(metaTags.getRefreshTime()) });
}
ParseData parseData = new ParseData(status, title, outlinks, content.getMetadata(), nutchMetadata);
ParseResult parseResult = ParseResult.createParseResult(content.getUrl(), new ParseImpl(text, parseData));
// run filters on parse
ParseResult filteredParse = this.htmlParseFilters.filter(content, parseResult, metaTags, root);
if (metaTags.getNoCache()) {
// not okay to cache
for (Map.Entry<org.apache.hadoop.io.Text, Parse> entry : filteredParse) entry.getValue().getData().getParseMeta().set(Nutch.CACHING_FORBIDDEN_KEY, cachingPolicy);
}
return filteredParse;
}
use of org.apache.tika.sax.LinkContentHandler in project acs-aem-commons by Adobe-Consulting-Services.
the class BrokenLinksReport method collectPaths.
/**
* Collect references from a JCR property.
* A property can be one of:
* <ol>
* <li>A string containing a reference, e.g, fileReference=/content/dam/image.png. </li>
* <li>An array of strings, e.g, fileReference=[/content/dam/image1.png, /content/dam/image2.png]</li>
* <li>An html fragment containing links , e.g,
* <pre>
* <p>
* <a href="/content/site/page.html">hello</a>
* <img src="/content/dam/image1.png">hello</a>
* </p>
* </pre>
* </li>
* </ol>
*
* @param property an entry from a ValueMap
* @param htmlFields lst of properties containing html
* @return stream containing extracted references
*/
static Stream<String> collectPaths(Map.Entry<String, Object> property, Set<String> htmlFields) {
Object p = property.getValue();
Stream<String> stream;
if (p.getClass() == String[].class) {
stream = Arrays.stream((String[]) p);
} else if (p.getClass() == String.class) {
stream = Stream.of((String) p);
} else {
stream = Stream.empty();
}
if (htmlFields.contains(property.getKey())) {
stream = stream.flatMap(val -> {
try {
// parse html and extract links via underlying tagsoup library
LinkContentHandler linkHandler = new LinkContentHandler();
HtmlParser parser = new HtmlParser();
parser.parse(new ByteArrayInputStream(val.getBytes("utf-8")), linkHandler, new Metadata(), new ParseContext());
return linkHandler.getLinks().stream().map(Link::getUri);
} catch (Exception e) {
return Stream.empty();
}
});
}
return stream;
}
Aggregations