[PDFBOX-5801] StackOverflowError with specific document

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Not A Problem
• Affects Version/s: 2.0.31, 3.0.2 PDFBox
• Labels: stack-overflow
• Environment: Windows 10 Pro - 19045.3208 ; openjdk version "13.0.2" 2020-01-14 ; Memory: -Xmx4G -Xms100m
• Assignee: Unassigned
• Reporter: Rodrigo Schonardt
• Created: 08/Apr/24 12:22
• Updated: 08/Apr/24 20:16
• Resolved: 08/Apr/24 20:16

Description

When i try to merge the attached file with a cover the error happens:

>>java.lang.StackOverflowError
>>    at java.base/sun.nio.cs.UTF_8$Decoder.decodeLoop(UTF_8.java:414)
UTF_8.java:414
>>    at java.base/java.nio.charset.CharsetDecoder.decode(CharsetDecoder.java:578)
CharsetDecoder.java:578
>>    at java.base/java.nio.charset.CharsetDecoder.decode(CharsetDecoder.java:801)
CharsetDecoder.java:801
>>    at org.apache.pdfbox.pdfparser.BaseParser.decodeBuffer(BaseParser.java:885)
BaseParser.java:885
>>    at org.apache.pdfbox.pdfparser.BaseParser.parseCOSName(BaseParser.java:872)
BaseParser.java:872
>>    at org.apache.pdfbox.pdfparser.BaseParser.parseCOSDictionaryNameValuePair(BaseParser.java:368)
BaseParser.java:368
>>    at org.apache.pdfbox.pdfparser.BaseParser.parseCOSDictionary(BaseParser.java:289)
BaseParser.java:289
>>    at org.apache.pdfbox.pdfparser.BaseParser.parseDirObject(BaseParser.java:915)
BaseParser.java:915
>>    at org.apache.pdfbox.pdfparser.COSParser.parseFileObject(COSParser.java:737)
COSParser.java:737
>>    at org.apache.pdfbox.pdfparser.COSParser.parseObjectDynamically(COSParser.java:671)
COSParser.java:671
>>    at org.apache.pdfbox.pdfparser.COSParser.dereferenceCOSObject(COSParser.java:625)
COSParser.java:625
>>    at org.apache.pdfbox.cos.COSObject.getObject(COSObject.java:121)
COSObject.java:121
>>    at org.apache.pdfbox.multipdf.PDFCloneUtility.hasSelfReference(PDFCloneUtility.java:249)
PDFCloneUtility.java:249
>>    at org.apache.pdfbox.multipdf.PDFCloneUtility.cloneCOSDictionary(PDFCloneUtility.java:176)
PDFCloneUtility.java:176
>>    at org.apache.pdfbox.multipdf.PDFCloneUtility.cloneCOSBaseForNewDocument(PDFCloneUtility.java:122)
PDFCloneUtility.java:122
>>    at org.apache.pdfbox.multipdf.PDFCloneUtility.cloneForNewDocument(PDFCloneUtility.java:100)
PDFCloneUtility.java:100
>>    at org.apache.pdfbox.multipdf.PDFCloneUtility.cloneCOSBaseForNewDocument(PDFCloneUtility.java:110)
PDFCloneUtility.java:110
>>    at org.apache.pdfbox.multipdf.PDFCloneUtility.cloneForNewDocument(PDFCloneUtility.java:100)
PDFCloneUtility.java:100{}

The code is the following, tried different configs but none helped:

PDFMergerUtility merger = new PDFMergerUtility();
merger.addSource( cover );
merger.addSource( file );
merger.setDestinationFileName( location );
merger.mergeDocuments(  null ); 

A similar error happens if i try to extract a page of the file and save it in a new file:

 >>java.lang.StackOverflowError
>>    at java.base/java.util.HashMap.putVal(HashMap.java:631)
HashMap.java:631
>>    at java.base/java.util.HashMap.put(HashMap.java:612)
HashMap.java:612
>>    at org.apache.pdfbox.pdfwriter.compress.COSObjectPool.put(COSObjectPool.java:80)
COSObjectPool.java:80
>>    at org.apache.pdfbox.pdfwriter.compress.COSWriterCompressionPool.addObjectToPool(COSWriterCompressionPool.java:141)
COSWriterCompressionPool.java:141
>>    at org.apache.pdfbox.pdfwriter.compress.COSWriterCompressionPool.addStructure(COSWriterCompressionPool.java:179)
COSWriterCompressionPool.java:179
>>    at org.apache.pdfbox.pdfwriter.compress.COSWriterCompressionPool.addElements(COSWriterCompressionPool.java:219)
COSWriterCompressionPool.java:219
>>    at org.apache.pdfbox.pdfwriter.compress.COSWriterCompressionPool.addStructure(COSWriterCompressionPool.java:188)
COSWriterCompressionPool.java:188
>>    at org.apache.pdfbox.pdfwriter.compress.COSWriterCompressionPool.addElements(COSWriterCompressionPool.java:219)

The code is the following:

PDDocument pdf = Loader.loadPDF( file );
       
int pageCount = pdf.getNumberOfPages();
           
for ( int i = 0; i < pageCount; i++) {               
   PDDocument singlePageDocument = new PDDocument();  
   PDPage page = pdf.getPage(i);   
   singlePageDocument.addPage(page); 
   singlePageDocument.save( location + (i + 1) + ".pdf");  
   singlePageDocument.close(); 
}
           
pdf.close(); 

I believe theres is something to do with the specific structure of the file, considering i was able to use way bigger files, both in pages and in size without problems. 
Also tried to use the same file on itext and no problem was found.