/**
* Constructor sets the input and output file and convert the pdf, docx and doc files to text .
*
* @param infile,outfile
* @return
*/
public TextConvertor(String infile, String outfile) {
try {
File input = new File(infile); // The file from where you would like to extract
FileInputStream fis = new FileInputStream(input.getAbsolutePath());
int x = fis.read();
int y = fis.read();
fis = new FileInputStream(input.getAbsolutePath());
if (x == 37 && y == 80) {
filetype = "pdf";
pd = PDDocument.load(input);
PDF2Text(outfile);
} else if (x == 80 && y == 75) {
filetype = "docx";
dx = new XWPFDocument(fis);
DOCX2Text(outfile);
} else if (x == 208 && y == 207) {
filetype = "doc";
dc = new HWPFDocument(fis);
DOC2Text(outfile);
}
} catch (Exception e) {
e.printStackTrace();
}
}
/**
* This method scans the document for all WidgetAnnotation objects.
*
* <p>This is done because some writer do not create a correct list of all PDAcroFormField objects
* in the AcroForm. In the case that the list of children is empty, we go and search ourselves for
* candidates...
*
* @param doc The document to reconstruct.
*/
protected COSArray reconstruct(PDDocument doc) {
COSArray result = COSArray.create();
if (doc == null) {
return result;
}
PDPageTree pageTree = doc.getPageTree();
if (pageTree == null) {
return result;
}
boolean signatureExists = false;
for (PDPage page = pageTree.getFirstPage(); page != null; page = page.getNextPage()) {
List annotations = page.getAnnotations();
if (annotations == null) {
continue;
}
for (Iterator it = annotations.iterator(); it.hasNext(); ) {
PDAnnotation annot = (PDAnnotation) it.next();
if (annot.isWidgetAnnotation()) {
COSDictionary cosAnnot = annot.cosGetDict();
result.basicAddSilent(cosAnnot);
cosAnnot.basicRemoveSilent(PDAcroFormField.DK_Parent);
signatureExists |= cosAnnot.get(PDAcroFormField.DK_FT).equals(PDAcroFormField.CN_FT_Sig);
}
}
}
if (signatureExists) {
int flags = getFieldInt(PDAcroForm.DK_SigFlags, 0);
flags |= AcroFormSigFlags.Bit_AppendOnly | AcroFormSigFlags.Bit_SignatureExists;
cosGetDict().basicPutSilent(PDAcroForm.DK_SigFlags, COSInteger.create(flags));
}
return result;
}
/**
* The destination page. ATTENTION: it is common have dangling destinations to invalid (null)
* pages around!
*
* @return The destination page. Be sure to handle null return values.
*/
public PDPage getPage(PDDocument doc) {
COSArray definition = cosGetArray();
COSObject page = definition.get(0);
if (page.asNumber() != null) {
int pageIndex = page.asNumber().intValue();
return doc.getPageTree().getPageAt(pageIndex);
}
if (page.asDictionary() != null) {
return (PDPage) PDPageNode.META.createFromCos(page.asDictionary());
}
return null;
}
private static void extract(InputStream in) throws Exception {
PDDocument document = null;
try {
PDFParser parser = new PDFParser(in);
parser.parse();
document = parser.getPDDocument();
if (document.isEncrypted()) {
System.err.println("Document is Encrypted!");
}
PDDocumentCatalog cat = document.getDocumentCatalog();
PDMetadata metadata = cat.getMetadata();
if (metadata != null) {
// System.out.println(metadata.getStream().getStreamTokens());
// Levantamos la MetaData
DocumentBuilderFactory dbFactory = DocumentBuilderFactory.newInstance();
DocumentBuilder dBuilder = dbFactory.newDocumentBuilder();
Document doc = dBuilder.parse(metadata.createInputStream());
// Buscamos el tag de SEmployee y el Element -> CUIT
NodeList nList = doc.getElementsByTagName("foaf:SEmployee");
Element elem = (Element) nList.item(0);
String cuit = elem.getElementsByTagName("foaf:cuit").item(0).getTextContent();
System.out.println(cuit);
System.out.println("---");
System.out.println(metadata.getInputStreamAsString());
}
} catch (Exception err) {
throw err;
} finally {
if (document != null)
try {
document.close();
} catch (Throwable err2) {
}
}
}
/**
* save the converted text (without any processing) to the given file.
*
* @param filename
* @return
*/
public void PDF2Text(String filename) {
try {
File output =
new File(filename); // The text file where you are going to store the extracted data
stripper = new PDFTextStripper();
wr = new BufferedWriter(new OutputStreamWriter(new FileOutputStream(output)));
stripper.writeText(pd, wr);
if (pd != null) {
pd.close();
}
} catch (Exception e) {
e.printStackTrace();
}
}
public void PDF2TextPreProssesd(String filename) {
try {
stripper = new PDFTextStripper();
stripper.setParagraphStart("&*&");
stripper.setLineSeparator("#%#");
stripper.setPageSeparator("#%#");
String fulltxt = stripper.getText(pd);
String paras[] = fulltxt.split("&*&");
File file = new File(filename);
try {
BufferedWriter out = new BufferedWriter(new FileWriter(file));
int i = 0;
while (i < paras.length) {
if (paras[i].length() > 200) {
String para = paras[i].replace("#%#", " ");
out.write(para + "\r\n");
}
i++;
}
out.close();
} catch (IOException ex) {
ex.printStackTrace();
}
if (pd != null) {
pd.close();
}
} catch (Exception e) {
e.printStackTrace();
}
}
