svn commit: r806888 - /lucene/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java

[jukka]

Author: jukka
Date: Sat Aug 22 20:57:07 2009
New Revision: 806888

URL: http://svn.apache.org/viewvc?rev=806888&view=rev
Log:
TIKA-267: encrypted pdf files aren't handled properly

Only try to decrypt the document if the encryption prevents text extraction. 
Fixed as suggested by Sascha Szott.

Also added an input metadata key for entering a document-specific decryption 
password.

Modified:
    
lucene/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java

Modified: 
lucene/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java
URL: 
http://svn.apache.org/viewvc/lucene/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java?rev=806888&r1=806887&r2=806888&view=diff
==============================================================================
--- 
lucene/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java
 (original)
+++ 
lucene/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java
 Sat Aug 22 20:57:07 2009
@@ -30,18 +30,35 @@
 import org.xml.sax.SAXException;
 
 /**
- * PDF parser
+ * PDF parser.
+ * <p>
+ * This parser can process also encrypted PDF documents if the required
+ * password is given as a part of the input metadata associated with a
+ * document. If no password is given, then this parser will try decrypting
+ * the document using the empty password that's often used with PDFs.
  */
 public class PDFParser implements Parser {
 
+    /**
+     * Metadata key for giving the document password to the parser.
+     *
+     * @since Apache Tika 0.5
+     */
+    public static final String PASSWORD = 
"org.apache.tika.parser.pdf.password";
+
     public void parse(
             InputStream stream, ContentHandler handler, Metadata metadata)
             throws IOException, SAXException, TikaException {
         PDDocument pdfDocument = PDDocument.load(stream);
         try {
-            if (pdfDocument.isEncrypted()) {
+            if (pdfDocument.isEncrypted()
+                    && 
!pdfDocument.getCurrentAccessPermission().canExtractContent()) {
                 try {
-                    pdfDocument.decrypt("");
+                    String password = metadata.get(PASSWORD);
+                    if (password == null) {
+                        password = "";
+                    }
+                    pdfDocument.decrypt(password);
                 } catch (Exception e) {
                     // Ignore
                 }