use of org.apache.poi.hemf.extractor.HemfExtractor in project tika by apache.
the class EMFParser method parse.
@Override
public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
EmbeddedDocumentExtractor embeddedDocumentExtractor = null;
XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
xhtml.startDocument();
try {
HemfExtractor ex = new HemfExtractor(stream);
long lastY = -1;
long lastX = -1;
//derive this from the font or frame/bounds information
long fudgeFactorX = 1000;
StringBuilder buffer = new StringBuilder();
for (HemfRecord record : ex) {
if (record.getRecordType() == HemfRecordType.comment) {
AbstractHemfComment comment = ((HemfCommentRecord) record).getComment();
if (comment instanceof HemfCommentPublic.MultiFormats) {
if (embeddedDocumentExtractor == null) {
embeddedDocumentExtractor = EmbeddedDocumentUtil.getEmbeddedDocumentExtractor(context);
}
handleMultiFormats((HemfCommentPublic.MultiFormats) comment, xhtml, embeddedDocumentExtractor);
} else if (comment instanceof HemfCommentPublic.WindowsMetafile) {
if (embeddedDocumentExtractor == null) {
embeddedDocumentExtractor = EmbeddedDocumentUtil.getEmbeddedDocumentExtractor(context);
}
handleWMF((HemfCommentPublic.WindowsMetafile) comment, xhtml, embeddedDocumentExtractor);
}
} else if (record.getRecordType().equals(HemfRecordType.exttextoutw)) {
HemfText.ExtTextOutW extTextOutW = (HemfText.ExtTextOutW) record;
if (lastY > -1 && lastY != extTextOutW.getY()) {
xhtml.startElement("p");
xhtml.characters(buffer.toString());
xhtml.endElement("p");
buffer.setLength(0);
lastX = -1;
}
if (lastX > -1 && extTextOutW.getX() - lastX > fudgeFactorX) {
buffer.append(" ");
}
String txt = extTextOutW.getText();
buffer.append(txt);
lastY = extTextOutW.getY();
lastX = extTextOutW.getX();
}
}
if (buffer.length() > 0) {
xhtml.startElement("p");
xhtml.characters(buffer.toString());
xhtml.endElement("p");
}
} catch (RecordFormatException e) {
//POI's hemfparser can throw these for "parse exceptions"
throw new TikaException(e.getMessage(), e);
} catch (RuntimeException e) {
//convert Runtime to RecordFormatExceptions
throw new TikaException(e.getMessage(), e);
}
xhtml.endDocument();
}
use of org.apache.poi.hemf.extractor.HemfExtractor in project poi by apache.
the class HemfPlusExtractorTest method getCommentRecord.
private HemfCommentEMFPlus getCommentRecord(String testFileName, int recordIndex) throws Exception {
InputStream is = null;
HemfCommentEMFPlus returnRecord = null;
try {
is = POIDataSamples.getSpreadSheetInstance().openResourceAsStream(testFileName);
HemfExtractor ex = new HemfExtractor(is);
int i = 0;
for (HemfRecord record : ex) {
if (i == recordIndex) {
HemfCommentRecord commentRecord = ((HemfCommentRecord) record);
returnRecord = (HemfCommentEMFPlus) commentRecord.getComment();
break;
}
i++;
}
} finally {
is.close();
}
return returnRecord;
}
Aggregations