Examples with BoilerpipeContentHandler org.apache.tika.sax.boilerpipe.BoilerpipeContentHandler used on opensource projects

Example 1 with BoilerpipeContentHandler

use of org.apache.tika.sax.boilerpipe.BoilerpipeContentHandler in project nutch by apache.

the class TikaParser method getParse.

ParseResult getParse(Content content, HTMLDocumentImpl doc, DocumentFragment root) {
    String mimeType = content.getContentType();
    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
    CompositeParser compositeParser = (CompositeParser) tikaConfig.getParser();
    Parser parser = compositeParser.getParsers().get(MediaType.parse(mimeType));
    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 {} for mime-type {}.", parser.getClass().getName(), mimeType);
    byte[] raw = content.getContent();
    Metadata tikamd = new Metadata();
    ContentHandler domHandler;
    // Check whether to use Tika's BoilerplateContentHandler
    if (useBoilerpipe && boilerpipeMimeTypes.contains(mimeType)) {
        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();
    if (parseEmbedded) {
        context.set(Parser.class, new AutoDetectParser(tikaConfig));
    }
    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(TikaCoreProperties.TITLE.toString()))
            continue;
        String[] values = tikamd.getValues(tikaMDName);
        for (String v : values) {
            nutchMetadata.add(tikaMDName, v);
            if (tikaMDName.equalsIgnoreCase(Nutch.ROBOTS_METATAG) && nutchMetadata.get(Nutch.ROBOTS_METATAG) == null) {
                // NUTCH-2720 force lowercase robots directive
                nutchMetadata.add(Nutch.ROBOTS_METATAG, 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;
}