Examples with ChmExtractor org.apache.tika.parser.chm.core.ChmExtractor used on opensource projects

Example 1 with ChmExtractor

use of org.apache.tika.parser.chm.core.ChmExtractor in project tika by apache.

the class TestChmExtraction method testExtractChmEntry.

protected void testExtractChmEntry(InputStream stream) throws TikaException, IOException {
    ChmExtractor chmExtractor = new ChmExtractor(stream);
    ChmDirectoryListingSet entries = chmExtractor.getChmDirList();
    final Pattern htmlPairP = Pattern.compile("\\Q<html\\E.+\\Q</html>\\E", Pattern.CASE_INSENSITIVE | Pattern.MULTILINE | Pattern.DOTALL);
    Set<String> names = new HashSet<String>();
    for (DirectoryListingEntry directoryListingEntry : entries.getDirectoryListingEntryList()) {
        byte[] data = chmExtractor.extractChmEntry(directoryListingEntry);
        //Entry names should be nice. Disable this if the test chm do have bad looking but valid entry names.
        if (!niceAscFileName(directoryListingEntry.getName())) {
            throw new TikaException("Warning: File name contains a non ascii char : " + directoryListingEntry.getName());
        }
        final String lowName = directoryListingEntry.getName().toLowerCase(Locale.ROOT);
        //check duplicate entry name which is seen before.
        if (names.contains(lowName)) {
            throw new TikaException("Duplicate File name detected : " + directoryListingEntry.getName());
        }
        names.add(lowName);
        if (lowName.endsWith(".html") || lowName.endsWith(".htm") || lowName.endsWith(".hhk") || lowName.endsWith(".hhc")) //|| name.endsWith(".bmp")
        {
            if (findZero(data)) {
                throw new TikaException("Xhtml/text file contains '\\0' : " + directoryListingEntry.getName());
            }
            //validate html
            String html = new String(data, ISO_8859_1);
            if (!htmlPairP.matcher(html).find()) {
                System.err.println(lowName + " is invalid.");
                System.err.println(html);
                throw new TikaException("Invalid xhtml file : " + directoryListingEntry.getName());
            }
        //                else {
        //                    System.err.println(directoryListingEntry.getName() + " is valid.");
        //                }
        }
    }
}

Example 2 with ChmExtractor

use of org.apache.tika.parser.chm.core.ChmExtractor in project tika by apache.

the class ChmParser method parse.

@Override
public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
    ChmExtractor chmExtractor = new ChmExtractor(stream);
    // metadata
    metadata.set(Metadata.CONTENT_TYPE, "application/vnd.ms-htmlhelp");
    // content
    XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
    xhtml.startDocument();
    Parser htmlParser = EmbeddedDocumentUtil.tryToFindExistingLeafParser(HtmlParser.class, context);
    if (htmlParser == null) {
        htmlParser = new HtmlParser();
    }
    for (DirectoryListingEntry entry : chmExtractor.getChmDirList().getDirectoryListingEntryList()) {
        final String entryName = entry.getName();
        if (entryName.endsWith(".html") || entryName.endsWith(".htm")) {
            //                AttributesImpl attrs = new AttributesImpl();
            //                attrs.addAttribute("", "name", "name", "String", entryName);
            //                xhtml.startElement("", "document", "document", attrs);
            byte[] data = chmExtractor.extractChmEntry(entry);
            parsePage(data, htmlParser, xhtml, context);
        //                xhtml.endElement("", "", "document");
        }
    }
    xhtml.endDocument();
}