1 files changed, 26 insertions, 30 deletions

diff --git a/pdf-as-lib/src/main/java/at/knowcenter/wag/egov/egiz/pdf/TextualSignature.java b/pdf-as-lib/src/main/java/at/knowcenter/wag/egov/egiz/pdf/TextualSignature.java
index 35a0768..3ce690b 100644
--- a/pdf-as-lib/src/main/java/at/knowcenter/wag/egov/egiz/pdf/TextualSignature.java
+++ b/pdf-as-lib/src/main/java/at/knowcenter/wag/egov/egiz/pdf/TextualSignature.java
@@ -53,7 +53,7 @@ import com.lowagie.text.pdf.PdfWriter;
 

 /**

  * Contains helper function for textual signatures.

- * 

+ *

  * @author wprinz

  */

 public class TextualSignature

@@ -62,21 +62,22 @@ public class TextualSignature
   * The logger definition.

   */

   private static final Logger logger_ = ConfigLogger.getLogger(TextualSignature.class);

-  

+

   /**

    * Extracts the document text from a given pdf.

-   * 

+   *

    * @param pdf_stream

    *          The pdf_input stream.

    * @return Returns the extracted document text.

- * @throws PDFDocumentException 

+ * @throws PDFDocumentException

    * @throws TextExtractionException

    *           Forwarded exception.

    */

-  public static String extractTextTextual(PdfDataSource pdfDataSource, String encoding) throws PDFDocumentException 

+  public static String extractTextTextual(PdfDataSource pdfDataSource, String encoding) throws PDFDocumentException

   {

     PerformanceCounters.textExtractions.increment();

-    

+    PDDocument doc  = null;

+    Document document = null;

     try

     {

       int first_page_rotation = 0;

@@ -90,12 +91,8 @@ public class TextualSignature
 

 

      // byte[] bytes = normalizePDF(pdf_stream);

-    	//iText  

-    	

-        byte [] pdf_data = pdfDataSource.getAsByteArray();

-        PdfReader reader = new PdfReader(pdf_data);

-        PDFASUtils.checkReaderPermissions(reader);

-        //pdf_stream.close();

+

+        PdfReader reader = PDFASUtils.createPdfReaderCheckingPermissions(pdfDataSource);

 

         // PERF: PDF normalization needs byte array - this is costy

         ByteArrayOutputStream baos = new ByteArrayOutputStream(4096);

@@ -108,7 +105,7 @@ public class TextualSignature
         // this method (although it works when a Table is appended)... very

         // fragile.

 

-        Document document = new Document();

+        document = new Document();

 

         PdfWriter writer = PdfWriter.getInstance(document, baos);

         document.open();

@@ -126,7 +123,7 @@ public class TextualSignature
         	first_page_rotation = new_size_withrot.getRotation();

         	//logger_.info("iText first_page_rotation="+new_size_withrot.getRotation());

           }

-          //logger_.info("iText set PageSize of page:"+page_num+" to: "+new_size_withrot); 

+          //logger_.info("iText set PageSize of page:"+page_num+" to: "+new_size_withrot);

           //document.setPageSize(new_size);

           document.setPageSize(new_size_withrot);

           document.newPage();

@@ -163,15 +160,15 @@ public class TextualSignature
       //logger_.info("temporary_dir="+temporary_dir.getAbsolutePath());

       parser.setTempDirectory(temporary_dir);

       parser.parse();

-      

-      PDDocument doc = parser.getPDDocument();

+

+      doc = parser.getPDDocument();

       //System.out.println("pdfBox.getNumberOfPages()"+doc.getNumberOfPages());

-      

+

       PDFTextStripper stripper = new PDFTextStripper();

       stripper.setSortByPosition(false);

       stripper.setGetFirstPageRotationFromThis(true);

       stripper.setFirstPageRotation(first_page_rotation);

-      

+

       // stripper.setStartPage(4);

       // stripper.setEndPage(4);

       logger_.debug("TextualSignator extractTextTextual: Begin stripping text");

@@ -182,8 +179,7 @@ public class TextualSignature
     	  throw new PDFDocumentException(ErrorCode.TEXT_EXTRACTION_EXCEPTION, "Unable to extract textual content.", e);

       }

       logger_.debug("TextualSignator extractTextTextual: Stripping text ended");

-      

-      doc.close();

+

       //logger_.debug("TextualSignator extractTextTextual="+text);

       return text;

 

@@ -199,39 +195,39 @@ public class TextualSignature
     catch (DocumentException e)

     {

        throw new PDFDocumentException(ErrorCode.DOCUMENT_CANNOT_BE_READ, e);

+    } finally {

+    	PDFASUtils.closeQuietly(doc);

+    	PDFASUtils.closeQuietly(document);

     }

   }

-  

+

   /**

    * Normalizes a given binary PDF to a version PDFbox can handle correctly.

-   * 

+   *

    * <p>

    * PDFbox has serious problems with documents that use incremental updates or

    * XObject forms. Therefor use this to remove incremental updates and create a

    * streamlined document.

    * </p>

-   * 

+   *

    * <p>

    * Note that this has nothing to do with text normalization. It just unifies

    * the PDF documents that are fed into PDFbox for text extraction and page

    * length determination.

    * </p>

-   * 

+   *

    * @param input_pdf

    *          The input pdf to be normalized.

    * @return Returns the normalized pdf.

    * @throws IOException

    * @throws DocumentException

- * @throws PDFDocumentException 

+ * @throws PDFDocumentException

    */

   public static byte[] normalizePDF(PdfDataSource pdfDataSource) throws IOException, DocumentException, PDFDocumentException

   {

 	  //iText

-    byte [] pdf_data = pdfDataSource.getAsByteArray();

-    PdfReader reader = new PdfReader(pdf_data);

-    PDFASUtils.checkReaderPermissions(reader);

-    //input_pdf.close();

-    

+    PdfReader reader = PDFASUtils.createPdfReaderCheckingPermissions(pdfDataSource);

+

     // PERF: PDF Normalization needs byte array

     ByteArrayOutputStream baos = new ByteArrayOutputStream();

     // For some reason the Reader -> ImportPage -> Writer mechanism produces