Rebuild the trailer when missing pages item

pdfbox/pom.xml

@@ -697,6 +697,19 @@
                             <sha512>5ae7f232c47c13ed31997eb2c368e7deb1013c1321d70bf79369f8d709b33406191d94c21a5d27b4c4bb48241bafd9328a0a6d2d093d4e540d5044e9503bd099</sha512>
                         </configuration>
                     </execution>
+                    <execution>
+                        <id>PDFBOX-5026</id>
+                        <phase>generate-test-resources</phase>
+                        <goals>
+                            <goal>wget</goal>
+                        </goals>
+                        <configuration>
+                            <url>https://issues.apache.org/jira/secure/attachment/13015945/issue9418.pdf</url>
+                            <outputDirectory>${project.build.directory}/pdfs</outputDirectory>
+                            <outputFileName>PDFBOX-5026.pdf</outputFileName>
+                            <sha512>1e47caa4246752bc392e596803cb95556bde578b687352a7434d9b77f92892539f810f046e76e4b2300d859ab9a8755a4212378ae099a825367f154919524d7c</sha512>
+                        </configuration>
+                    </execution>
                 </executions>
             </plugin>
         </plugins>

pdfbox/src/main/java/org/apache/pdfbox/pdfparser/COSParser.java

@@ -272,7 +272,7 @@ protected COSDictionary retrieveTrailer() throws IOException
             }
         }
         // check if the trailer contains a Root object
-        if (trailer != null && trailer.getItem(COSName.ROOT) == null)
+        if (trailer != null && !isValidTrailer(trailer))
         {
             rebuildTrailer = isLenient();
         }
@@ -292,6 +292,30 @@ protected COSDictionary retrieveTrailer() throws IOException
         return trailer;
     }
 
+    /**
+     * Check that the trailer contains a Root object and that the Root
+     * contains a Pages object.
+     *
+     * @param trailer the trailer to validate
+     * @return whether or not the trailer is valid.
+     * @throws IOException if an error occurs
+     */
+    private boolean isValidTrailer(COSDictionary trailer) throws IOException
+    {
+        COSObject root = trailer.getCOSObject(COSName.ROOT);
+        if (root == null)
+        {
+            return false;
+        }
+        COSBase base = parseObjectDynamically(root, false);
+        if (!(base instanceof COSDictionary))
+        {
+            return false;
+        }
+        COSDictionary rootDict = (COSDictionary) base;
+        return rootDict.getCOSObject(COSName.PAGES) != null;
+    }
+
     /**
      * Parses cross reference tables.
      * 

pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFParser.java

@@ -25,7 +25,6 @@
 import org.apache.pdfbox.cos.COSDictionary;
 import org.apache.pdfbox.cos.COSDocument;
 import org.apache.pdfbox.cos.COSName;
-import org.apache.pdfbox.cos.COSNull;
 import org.apache.pdfbox.cos.COSObject;
 import org.apache.pdfbox.io.IOUtils;
 import org.apache.pdfbox.io.RandomAccessRead;
@@ -164,14 +163,14 @@ public PDDocument getPDDocument() throws IOException
      * The initial parse will first parse only the trailer, the xrefstart and all xref tables to have a pointer (offset)
      * to all the pdf's objects. It can handle linearized pdfs, which will have an xref at the end pointing to an xref
      * at the beginning of the file. Last the root object is parsed.
-     * 
+     *
      * @throws InvalidPasswordException If the password is incorrect.
      * @throws IOException If something went wrong.
      */
     protected void initialParse() throws IOException
     {
         COSDictionary trailer = retrieveTrailer();
-    
+
         COSBase base = parseTrailerValuesDynamically(trailer);
         if (!(base instanceof COSDictionary))
         {

pdfbox/src/test/java/org/apache/pdfbox/pdfparser/TestPDFParser.java

@@ -363,6 +363,19 @@ public void testPDFBox4490() throws IOException
         doc.close();
     }
 
+    /**
+     * Test that PDFBOX-5026 has pages tree.
+     *
+     * @throws IOException
+     */
+    @Test
+    public void testPDFBox5026() throws IOException
+    {
+        PDDocument doc = PDDocument.load(new File(TARGETPDFDIR, "PDFBOX-5026.pdf"));
+        assertEquals(1, doc.getNumberOfPages());
+        doc.close();
+    }
+
     private void executeParserTest(RandomAccessRead source, MemoryUsageSetting memUsageSetting) throws IOException
     {
         ScratchFile scratchFile = new ScratchFile(memUsageSetting);