Add support for protected ("encrypted") PDFs

* add BouncyCastle library * remove BouncyCastle installation warning string * add test file write-protected.pdf * better exception name
JabRef · Apr 12, 2016 · a39f11f · a39f11f
1 parent 39f7ed7
commit a39f11f
Show file tree

Hide file tree

Showing 14 changed files with 108 additions and 50 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -95,7 +95,7 @@ to [sourceforge feature requests](https://sourceforge.net/p/jabref/features/) by
 - Fixed [#1062](https://github.com/JabRef/jabref/issues/1062): Merge entry with DOI information now also applies changes to entry type
 - Fixed [#535](https://github.com/JabRef/jabref/issues/535): Add merge action to right click menu
 - Fixed [#1115](https://github.com/JabRef/jabref/issues/1115): Wrong warning message when importing duplicate entries
-
+- Fixed [#935](https://github.com/JabRef/jabref/issues/935): PDFs, which are readable, but carry a protection for editing, are treated by the XMP parser and the importer generating a BibTeX entry based on the content.
 
 ### Removed
 - Fixed [#627](https://github.com/JabRef/jabref/issues/627): The pdf field is removed from the export formats, use the file field

diff --git a/build.gradle b/build.gradle
@@ -71,6 +71,7 @@ dependencies {
     compile 'org.apache.pdfbox:pdfbox:1.8.11'
     compile 'org.apache.pdfbox:fontbox:1.8.11'
     compile 'org.apache.pdfbox:jempbox:1.8.11'
+    compile 'org.bouncycastle:bcprov-jdk15on:1.54'
 
     compile 'commons-cli:commons-cli:1.3.1'
 

diff --git a/external-libraries.txt b/external-libraries.txt
@@ -92,6 +92,11 @@ Project: Apache PDFBox
 URL:     http://pdfbox.apache.org
 License: Apache 2.0
 
+Id:      org.bouncycastle:bcprov-jdk15on
+Project: The Legion of the Bouncy Castle
+URL:     https://www.bouncycastle.org/
+License: MIT
+
 Id:      org.jsoup:jsoup
 Project: jsoup
 URL:     https://github.com/jhy/jsoup/

diff --git a/src/main/java/net/sf/jabref/importer/EntryFromPDFCreator.java b/src/main/java/net/sf/jabref/importer/EntryFromPDFCreator.java
@@ -7,17 +7,17 @@
 import java.util.List;
 import java.util.Optional;
 
-import net.sf.jabref.gui.IconTheme;
 import org.apache.pdfbox.pdmodel.PDDocument;
 import org.apache.pdfbox.pdmodel.PDDocumentInformation;
 
-import net.sf.jabref.model.entry.BibEntry;
-import net.sf.jabref.pdfimport.PdfImporter;
-import net.sf.jabref.pdfimport.PdfImporter.ImportPdfFilesResult;
 import net.sf.jabref.JabRef;
 import net.sf.jabref.external.ExternalFileType;
 import net.sf.jabref.external.ExternalFileTypes;
+import net.sf.jabref.gui.IconTheme;
 import net.sf.jabref.logic.xmp.XMPUtil;
+import net.sf.jabref.model.entry.BibEntry;
+import net.sf.jabref.pdfimport.PdfImporter;
+import net.sf.jabref.pdfimport.PdfImporter.ImportPdfFilesResult;
 
 /**
  * Uses XMPUtils to get one BibEntry for a PDF-File.
@@ -41,11 +41,7 @@ private static ExternalFileType getPDFExternalFileType() {
         return pdfFileType.get();
     }
 
-    /*
-     * (non-Javadoc)
-     *
-     * @see net.sf.jabref.imports.EntryFromFileCreator#accept(java.io.File)
-     *
+    /**
      * Accepts all Files having as suffix ".PDF" (in ignore case mode).
      */
     @Override
@@ -114,7 +110,7 @@ private void addEntryDataFromPDDocumentInformation(File pdfFile, BibEntry entry)
     }
 
     /**
-     * Adds all data Found in all the entrys of this XMP file to the given
+     * Adds all data Found in all the entries of this XMP file to the given
      * entry. This was implemented without having much knowledge of the XMP
      * format.
      *

diff --git a/src/main/java/net/sf/jabref/importer/fileformat/PdfContentImporter.java b/src/main/java/net/sf/jabref/importer/fileformat/PdfContentImporter.java
@@ -4,6 +4,8 @@
 import net.sf.jabref.importer.OutputPrinter;
 import net.sf.jabref.importer.fetcher.DOItoBibTeXFetcher;
 import net.sf.jabref.logic.util.DOI;
+import net.sf.jabref.logic.xmp.EncryptedPdfsNotSupportedException;
+import net.sf.jabref.logic.xmp.XMPUtil;
 import net.sf.jabref.model.entry.BibEntry;
 import net.sf.jabref.model.entry.BibtexEntryTypes;
 import net.sf.jabref.model.entry.EntryType;
@@ -19,6 +21,7 @@
 import java.io.InputStream;
 import java.io.StringWriter;
 import java.util.ArrayList;
+import java.util.Collections;
 import java.util.List;
 import java.util.Optional;
 import java.util.regex.Matcher;
@@ -184,12 +187,7 @@ public boolean isRecognizedFormat(InputStream in) throws IOException {
     public List<BibEntry> importEntries(InputStream in, OutputPrinter status) throws IOException {
         final ArrayList<BibEntry> result = new ArrayList<>(1);
 
-        try (PDDocument document = PDDocument.load(in)) {
-            if (document.isEncrypted()) {
-                LOGGER.info("Encrypted documents are not supported");
-                return result;
-            }
-
+        try (PDDocument document = XMPUtil.loadWithAutomaticDecryption(in)) {
             String firstPageContents = getFirstPageContents(document);
 
             Optional<DOI> doi = DOI.findInText(firstPageContents);
@@ -481,6 +479,9 @@ public void addEntry(BibEntry entry) {
             }
 
             result.add(entry);
+        } catch (EncryptedPdfsNotSupportedException e) {
+            LOGGER.info("Decryption not supported");
+            return Collections.EMPTY_LIST;
         }
         return result;
     }

diff --git a/.../xmp/EncryptionNotSupportedException.java → ...p/EncryptedPdfsNotSupportedException.java b/.../xmp/EncryptionNotSupportedException.java → ...p/EncryptedPdfsNotSupportedException.java
@@ -1,4 +1,4 @@
-/*  Copyright (C) 2003-2011 JabRef contributors.
+/*  Copyright (C) 2003-2016 JabRef contributors.
     This program is free software; you can redistribute it and/or modify
     it under the terms of the GNU General Public License as published by
     the Free Software Foundation; either version 2 of the License, or
@@ -17,8 +17,6 @@
 
 import java.io.IOException;
 
-public class EncryptionNotSupportedException extends IOException {
-    public EncryptionNotSupportedException(String string) {
-        super(string);
-    }
+public class EncryptedPdfsNotSupportedException extends IOException {
+    // no additional information needed
 }
diff --git a/src/main/java/net/sf/jabref/logic/xmp/XMPUtil.java b/src/main/java/net/sf/jabref/logic/xmp/XMPUtil.java
@@ -46,10 +46,13 @@
 import org.apache.pdfbox.cos.COSDictionary;
 import org.apache.pdfbox.cos.COSName;
 import org.apache.pdfbox.exceptions.COSVisitorException;
+import org.apache.pdfbox.exceptions.CryptographyException;
 import org.apache.pdfbox.pdmodel.PDDocument;
 import org.apache.pdfbox.pdmodel.PDDocumentCatalog;
 import org.apache.pdfbox.pdmodel.PDDocumentInformation;
 import org.apache.pdfbox.pdmodel.common.PDMetadata;
+import org.apache.pdfbox.pdmodel.encryption.BadSecurityHandlerException;
+import org.apache.pdfbox.pdmodel.encryption.StandardDecryptionMaterial;
 import org.w3c.dom.Document;
 
 /**
@@ -121,6 +124,39 @@ public static List<BibEntry> readXMP(File file) throws IOException {
         return result;
     }
 
+    public static PDDocument loadWithAutomaticDecryption(InputStream inputStream)
+            throws IOException, EncryptedPdfsNotSupportedException {
+        PDDocument doc = PDDocument.load(inputStream);
+        if (doc.isEncrypted()) {
+            // try the empty string as user password
+            StandardDecryptionMaterial sdm = new StandardDecryptionMaterial("");
+            try {
+                doc.openProtection(sdm);
+            } catch (BadSecurityHandlerException | CryptographyException e) {
+                LOGGER.error("Cannot handle encrypted PDF: " + e.getMessage());
+                throw new EncryptedPdfsNotSupportedException();
+            } catch (NoClassDefFoundError e) {
+                // This is to avoid following exception:
+                // Exception in thread "JabRef CachedThreadPool" java.lang.NoClassDefFoundError: org/bouncycastle/jce/provider/BouncyCastleProvider
+                // at org.apache.pdfbox.pdmodel.PDDocument.openProtection(PDDocument.java:1611)
+                // at net.sf.jabref.logic.xmp.XMPUtil.loadWithAutomaticDecryption(XMPUtil.java:133)
+                // This exception occurs if JabRef is compiled without 'org.bouncycastle:bcprov-jdk15on' (meaning, without the BouncyCastle library), which may happen in some countries not allowing cryptography.
+                // See for instance http://www.bouncycastle.org/wiki/display/JA1/Frequently+Asked+Questions#FrequentlyAskedQuestions-11.WhatisBouncyCastle%27sexportclassificationintheUnitedStatesofAmerica?
+                // See also https://sourceforge.net/p/jabref/bugs/1257/ and http://stackoverflow.com/a/2929228/873282
+                if (e.getMessage().equals("org/bouncycastle/jce/provider/BouncyCastleProvider")) {
+                    LOGGER.warn(
+                            "Java Bouncy Castle library not found. This might have been removed due redistribution restrictions. Please download and install it. For more information see http://www.bouncycastle.org/.");
+                    // We convert it to a EncryptionNotSupportedException as this is handled properly by the caller
+                    throw new EncryptedPdfsNotSupportedException();
+                } else {
+                    // we really cannot deal with it
+                    throw e;
+                }
+            }
+        }
+        return doc;
+    }
+
     /**
      * Try to read the given BibTexEntry from the XMP-stream of the given
      * inputstream containing a PDF-file.
@@ -139,11 +175,7 @@ public static List<BibEntry> readXMP(InputStream inputStream)
 
         List<BibEntry> result = new LinkedList<>();
 
-        try (PDDocument document = PDDocument.load(inputStream)) {
-            if (document.isEncrypted()) {
-                throw new EncryptionNotSupportedException("Error: Cannot read metadata from encrypted document.");
-            }
-
+        try (PDDocument document = loadWithAutomaticDecryption(inputStream)) {
             Optional<XMPMetadata> meta = XMPUtil.getXMPMetadata(document);
 
             if (meta.isPresent()) {
@@ -508,13 +540,8 @@ public static String toXMP(Collection<BibEntry> bibtexEntries,
      * @return The XMPMetadata object found in the file
      */
     private static Optional<XMPMetadata> readRawXMP(InputStream inputStream) throws IOException {
-        try (PDDocument document = PDDocument.load(inputStream)) {
-            if (document.isEncrypted()) {
-                throw new EncryptionNotSupportedException("Error: Cannot read metadata from encrypted document.");
-            }
-
+        try (PDDocument document = loadWithAutomaticDecryption(inputStream)) {
             return XMPUtil.getXMPMetadata(document);
-
         }
     }
 
@@ -1036,8 +1063,7 @@ public static void writeXMP(File file,
 
         try (PDDocument document = PDDocument.load(file.getAbsoluteFile())) {
             if (document.isEncrypted()) {
-                throw new EncryptionNotSupportedException(
-                        "Error: Cannot add metadata to encrypted document.");
+                throw new EncryptedPdfsNotSupportedException();
             }
 
             if (writePDFInfo && (resolvedEntries.size() == 1)) {
@@ -1083,10 +1109,9 @@ public static void writeXMP(File file,
             try {
                 document.save(file.getAbsolutePath());
             } catch (COSVisitorException e) {
-                throw new TransformerException("Could not write XMP-metadata: "
-                        + e.getLocalizedMessage());
+                LOGGER.debug("Could not write XMP metadata", e);
+                throw new TransformerException("Could not write XMP metadata: " + e.getLocalizedMessage(), e);
             }
-
         }
     }
 
@@ -1265,7 +1290,7 @@ public static boolean hasMetadata(InputStream inputStream) {
         try {
             List<BibEntry> bibEntries = XMPUtil.readXMP(inputStream);
             return !bibEntries.isEmpty();
-        } catch (EncryptionNotSupportedException ex) {
+        } catch (EncryptedPdfsNotSupportedException ex) {
             LOGGER.info("Encryption not supported by XMPUtil");
             return false;
         } catch (IOException e) {

diff --git a/src/test/java/net/sf/jabref/importer/fileformat/PdfContentImporterTestFiles.java b/src/test/java/net/sf/jabref/importer/fileformat/PdfContentImporterTestFiles.java
@@ -34,7 +34,11 @@ public static Collection<Object[]> fileNames() {
         // The test folder contains pairs of PDFs and BibTeX files. We check each pair.
         // This method returns the basenames of the available pairs
 
-        Object[][] data = new Object[][] {{"LNCS-minimal"}};
+        Object[][] data = new Object[][] {
+                // minimal PDF, not encrypted
+                {"LNCS-minimal"},
+                // minimal PDF, write-protected, thus encrypted
+                {"LNCS-minimal-protected"}};
         return Arrays.asList(data);
     }
 

diff --git a/src/test/java/net/sf/jabref/importer/fileformat/PdfXmpImporterTest.java b/src/test/java/net/sf/jabref/importer/fileformat/PdfXmpImporterTest.java
@@ -3,12 +3,10 @@
 import net.sf.jabref.Globals;
 import net.sf.jabref.JabRefPreferences;
 import net.sf.jabref.importer.OutputPrinterToNull;
+import net.sf.jabref.logic.xmp.EncryptedPdfsNotSupportedException;
 import net.sf.jabref.model.entry.BibEntry;
 import org.junit.Before;
-import org.junit.Rule;
 import org.junit.Test;
-import org.junit.rules.ExpectedException;
-
 import java.io.IOException;
 import java.io.InputStream;
 import java.util.Arrays;
@@ -18,8 +16,6 @@
 
 public class PdfXmpImporterTest {
 
-    @Rule
-    public ExpectedException thrown = ExpectedException.none();
     private PdfXmpImporter importer;
 
 
@@ -34,11 +30,9 @@ public void testGetFormatName() {
         assertEquals("XMP-annotated PDF", importer.getFormatName());
     }
 
-    @Test
+    @Test(expected = EncryptedPdfsNotSupportedException.class)
     public void importEncryptedFileThrowsException() throws IOException {
         try (InputStream is = PdfXmpImporterTest.class.getResourceAsStream("/pdfs/encrypted.pdf")) {
-            thrown.expect(IOException.class);
-            thrown.expectMessage("Error: Cannot read metadata from encrypted document.");
             importer.importEntries(is, new OutputPrinterToNull());
         }
     }

diff --git a/src/test/java/net/sf/jabref/logic/xmp/XMPUtilTest.java b/src/test/java/net/sf/jabref/logic/xmp/XMPUtilTest.java
@@ -8,6 +8,7 @@
 import net.sf.jabref.model.database.BibDatabaseMode;
 import net.sf.jabref.model.entry.AuthorList;
 import net.sf.jabref.bibtex.BibEntryWriter;
+import net.sf.jabref.bibtex.BibtexEntryAssert;
 import net.sf.jabref.model.entry.IdGenerator;
 import net.sf.jabref.model.entry.BibEntry;
 import net.sf.jabref.model.entry.BibtexEntryTypes;
@@ -860,6 +861,34 @@ public void testWriteMultiple() throws IOException, TransformerException {
         assertEqualsBibtexEntry(t3BibtexEntry(), b);
     }
 
+    /**
+     * Tests whether a edit-protected PDF can be read
+     */
+    @Test
+    public void testReadProtectedPDFHasMetaData() throws Exception {
+        try (InputStream is = XMPUtilTest.class.getResourceAsStream("/pdfs/write-protected.pdf")) {
+            Assert.assertTrue(XMPUtil.hasMetadata(is));
+        }
+    }
+
+    /**
+     * Tests whether a edit-protected PDF can be read
+     */
+    @Test
+    public void testReadProtectedPDFHasCorrectMetaData() throws Exception {
+        try (InputStream is = XMPUtilTest.class.getResourceAsStream("/pdfs/write-protected.pdf")) {
+            List<BibEntry> readEntries = XMPUtil.readXMP(is);
+
+            BibEntry entry = new BibEntry();
+            entry.setType("misc");
+            entry.setField("author", "Firstname Lastname");
+            List<BibEntry> expected = new ArrayList<>(1);
+            expected.add(entry);
+
+            BibtexEntryAssert.assertEquals(expected, readEntries);
+        }
+    }
+
     @Test
     public void testReadWriteDC() throws IOException, TransformerException {
         List<BibEntry> l = new LinkedList<>();
@@ -1272,14 +1301,14 @@ public void testResolveStrings() throws IOException, TransformerException {
                 AuthorList.parse(x.getField("author")));
     }
 
-    @Test(expected = EncryptionNotSupportedException.class)
+    @Test(expected = EncryptedPdfsNotSupportedException.class)
     public void expectedEncryptionNotSupportedExceptionAtRead() throws IOException {
         try (InputStream is = XMPUtilTest.class.getResourceAsStream("/pdfs/encrypted.pdf")) {
             XMPUtil.readXMP(is);
         }
     }
 
-    @Test(expected = EncryptionNotSupportedException.class)
+    @Test(expected = EncryptedPdfsNotSupportedException.class)
     public void expectedEncryptionNotSupportedExceptionAtWrite() throws IOException, TransformerException {
         XMPUtil.writeXMP("src/test/resources/pdfs/encrypted.pdf", t1BibtexEntry(), null);
     }

diff --git a/src/test/resources/net/sf/jabref/importer/fileformat/LNCS-minimal-protected.bib b/src/test/resources/net/sf/jabref/importer/fileformat/LNCS-minimal-protected.bib
@@ -0,0 +1,5 @@
+@inproceedings{,
+  abstract = {Abstract goes here},
+  author = {Firstname Lastname and Firstname Lastname},
+  title = {Paper Title}
+}
diff --git a/src/test/resources/net/sf/jabref/importer/fileformat/LNCS-minimal-protected.pdf b/src/test/resources/net/sf/jabref/importer/fileformat/LNCS-minimal-protected.pdf
diff --git a/src/test/resources/pdfs/write-protected.docx b/src/test/resources/pdfs/write-protected.docx
diff --git a/src/test/resources/pdfs/write-protected.pdf b/src/test/resources/pdfs/write-protected.pdf