close #166 Parse documents with negative byte offset for inUse xref e…

…ntry
torakiki · May 2, 2024 · c6a5ca6 · c6a5ca6
1 parent d382bf5
commit c6a5ca6
Show file tree

Hide file tree

Showing 2 changed files with 22 additions and 1 deletion.
diff --git a/src/main/java/org/sejda/sambox/input/XrefParser.java b/src/main/java/org/sejda/sambox/input/XrefParser.java
@@ -88,7 +88,16 @@ void onTrailerFound(COSDictionary found)
             @Override
             void onEntryFound(XrefEntry entry)
             {
-                parser().provider().addEntryIfAbsent(entry);
+                if(entry.getByteOffset() < XrefEntry.UNKNOWN_OFFSET)
+                {
+                    LOG.warn("Ignoring xrefEntry inUse with byteOffset: {}", entry);
+                }
+                else
+                {
+                    parser().provider().addEntryIfAbsent(entry);
+                }
+
+
             }
         };
     }

diff --git a/src/test/java/org/sejda/sambox/input/PDFParserTest.java b/src/test/java/org/sejda/sambox/input/PDFParserTest.java
@@ -512,6 +512,18 @@ public void testPDFBOX4372() throws IOException
         }
     }
 
+    @Test
+    public void testNegativeOffsetXrefEntry() throws IOException
+    {
+        try (PDDocument doc = PDFParser.parse(SeekableSources.inMemorySeekableSourceFrom(
+                getClass().getResourceAsStream("/sambox/test_multiple_xref_tables_negative_offset.pdf"))))
+        {
+
+            PDFTextStripper textStripper = new PDFTextStripper();
+            textStripper.getText(doc);
+        }
+    }
+
     //https://lists.apache.org/thread/rmvn7t4bj4d8wsln722kdk39k4fc0ff7
     @Test
     public void noLoopOrOOM() throws IOException