in tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/ooxml/OOXMLExtractorFactory.java [92:272]
public static void parse(InputStream stream, ContentHandler baseHandler, Metadata metadata,
ParseContext context) throws IOException, SAXException, TikaException {
Locale locale = context.get(Locale.class, LocaleUtil.getUserLocale());
//if there's a problem opening the zip file;
//create a tmp file, and copy what you can read of it.
File tmpRepairedCopy = null;
OPCPackage pkg = null;
//if the pkg is in the opencontainer of a TikaInputStream, it will get closed.
//However, if a regular inputstream has been sent in, we need to revert the pkg.
boolean mustRevertPackage = false;
try {
OOXMLExtractor extractor = null;
// Locate or Open the OPCPackage for the file
TikaInputStream tis = TikaInputStream.cast(stream);
if (tis != null && tis.getOpenContainer() instanceof OPCPackageWrapper) {
pkg = ((OPCPackageWrapper) tis.getOpenContainer()).getOPCPackage();
} else if (tis != null && tis.hasFile()) {
try {
pkg = OPCPackage.open(tis.getFile().getPath(), PackageAccess.READ);
} catch (InvalidOperationException e) {
tmpRepairedCopy = Files.createTempFile("tika-ooxml-repair-", "").toFile();
ZipSalvager.salvageCopy(tis.getFile(), tmpRepairedCopy);
pkg = OPCPackage.open(tmpRepairedCopy, PackageAccess.READ);
}
tis.setOpenContainer(new OPCPackageWrapper(pkg));
} else {
//OPCPackage slurps rris into memory so we can close rris
//without apparent problems
mustRevertPackage = true;
try (RereadableInputStream rereadableInputStream = new RereadableInputStream(stream,
MAX_BUFFER_LENGTH, false)) {
try {
pkg = OPCPackage.open(CloseShieldInputStream.wrap(rereadableInputStream));
} catch (UnsupportedZipFeatureException e) {
if (e.getFeature() !=
UnsupportedZipFeatureException.Feature.DATA_DESCRIPTOR) {
throw e;
}
rereadableInputStream.rewind();
tmpRepairedCopy = Files.createTempFile("tika-ooxml-repair-", "").toFile();
ZipSalvager.salvageCopy(rereadableInputStream, tmpRepairedCopy, false);
//if there isn't enough left to be opened as a package
//throw an exception -- we may want to fall back to streaming
//parsing
pkg = OPCPackage.open(tmpRepairedCopy, PackageAccess.READ);
} catch (IOException e) {
if (e instanceof EOFException) {
//keep going
} else if (e instanceof IOException && e.getMessage() != null &&
e.getMessage().contains("Truncated")) {
//keep going
} else {
throw e;
}
rereadableInputStream.rewind();
tmpRepairedCopy = Files.createTempFile("tika-ooxml-repair-", "").toFile();
ZipSalvager.salvageCopy(rereadableInputStream, tmpRepairedCopy, false);
//if there isn't enough left to be opened as a package
//throw an exception -- we may want to fall back to streaming
//parsing
pkg = OPCPackage.open(tmpRepairedCopy, PackageAccess.READ);
}
}
}
if (pkg != null) {
PackageRelationshipCollection prc =
pkg.getRelationshipsByType(OOXMLParser.SIGNATURE_RELATIONSHIP);
if (prc != null && prc.size() > 0) {
metadata.set(TikaCoreProperties.HAS_SIGNATURE, "true");
}
}
MediaType type = null;
String mediaTypeString = metadata.get(Metadata.CONTENT_TYPE);
if (mediaTypeString != null) {
type = MediaType.parse(mediaTypeString);
}
if (type != null && OOXMLParser.UNSUPPORTED_OOXML_TYPES.contains(type)) {
// Not a supported type, delegate to Empty Parser
EmptyParser.INSTANCE.parse(stream, baseHandler, metadata, context);
return;
}
if (type == null || !OOXMLParser.SUPPORTED_TYPES.contains(type)) {
// Get the type, and ensure it's one we handle
type = OPCPackageDetector.detectOfficeOpenXML(pkg);
}
if (type == null || OOXMLParser.UNSUPPORTED_OOXML_TYPES.contains(type)) {
// Not a supported type, delegate to Empty Parser
EmptyParser.INSTANCE.parse(stream, baseHandler, metadata, context);
return;
}
metadata.set(Metadata.CONTENT_TYPE, type.toString());
// Have the appropriate OOXML text extractor picked
POIXMLTextExtractor poiExtractor = null;
// This has already been set by OOXMLParser's call to configure()
// We can rely on this being non-null.
OfficeParserConfig config = context.get(OfficeParserConfig.class);
if (config.isUseSAXDocxExtractor()) {
poiExtractor = trySXWPF(pkg);
}
if (poiExtractor == null) {
poiExtractor = tryXSLF(pkg, config.isUseSAXPptxExtractor());
}
if (type.equals(OOXMLParser.XPS)) {
poiExtractor = new XPSTextExtractor(pkg);
}
if (poiExtractor == null) {
poiExtractor = EXTRACTOR_FACTORY.create(pkg);
}
POIXMLDocument document = poiExtractor.getDocument();
if (poiExtractor instanceof XSSFBEventBasedExcelExtractor) {
extractor = new XSSFBExcelExtractorDecorator(context, poiExtractor, locale);
} else if (poiExtractor instanceof XSSFEventBasedExcelExtractor) {
extractor = new XSSFExcelExtractorDecorator(context, poiExtractor, locale);
} else if (poiExtractor instanceof XWPFEventBasedWordExtractor) {
extractor = new SXWPFWordExtractorDecorator(metadata, context,
(XWPFEventBasedWordExtractor) poiExtractor);
metadata.add(TikaCoreProperties.TIKA_PARSED_BY,
XWPFEventBasedWordExtractor.class.getCanonicalName());
} else if (poiExtractor instanceof XSLFEventBasedPowerPointExtractor) {
extractor = new SXSLFPowerPointExtractorDecorator(metadata, context,
(XSLFEventBasedPowerPointExtractor) poiExtractor);
metadata.add(TikaCoreProperties.TIKA_PARSED_BY,
XSLFEventBasedPowerPointExtractor.class.getCanonicalName());
} else if (poiExtractor instanceof XPSTextExtractor) {
extractor = new XPSExtractorDecorator(context, poiExtractor);
} else if (document == null) {
throw new TikaException(
"Expecting UserModel based POI OOXML extractor with a document, but none" +
" found. " +
"The extractor returned was a " + poiExtractor);
} else if (document instanceof XMLSlideShow) {
extractor = new XSLFPowerPointExtractorDecorator(metadata, context,
(org.apache.poi.xslf.extractor.XSLFExtractor) poiExtractor);
} else if (document instanceof XWPFDocument) {
extractor = new XWPFWordExtractorDecorator(metadata, context,
(XWPFWordExtractor) poiExtractor);
} else {
extractor = new POIXMLTextExtractorDecorator(context, poiExtractor);
}
// Get the bulk of the metadata first, so that it's accessible during
// parsing if desired by the client (see TIKA-1109)
extractor.getMetadataExtractor().extract(metadata);
// Extract the text, along with any in-document metadata
extractor.getXHTML(baseHandler, metadata, context);
} catch (IllegalArgumentException e) {
if (e.getMessage() != null &&
e.getMessage().startsWith("No supported documents found")) {
throw new TikaException("TIKA-418: RuntimeException while getting content" +
" for thmx and xps file types", e);
} else {
throw new TikaException("Error creating OOXML extractor", e);
}
} catch (OpenXML4JException | XmlException e) {
throw new TikaException("Error creating OOXML extractor", e);
} catch (RuntimeSAXException e) {
throw (SAXException) e.getCause();
} finally {
if (pkg != null && mustRevertPackage) {
pkg.revert();
}
if (tmpRepairedCopy != null) {
boolean deleted = tmpRepairedCopy.delete();
if (!deleted) {
LOG.warn("failed to delete tmp (repair) file: " +
tmpRepairedCopy.getAbsolutePath());
}
}
}
}