Martin Mirchev created PDFBOX-6334:
--------------------------------------

             Summary: StringIndexOutOfBoundsException in DomXmpParser for a 
truncated xpacket end instruction
                 Key: PDFBOX-6334
                 URL: https://issues.apache.org/jira/browse/PDFBOX-6334
             Project: PDFBox
          Issue Type: Bug
          Components: XmpBox
    Affects Versions: 4.0.0, 3.0.8 PDFBox
            Reporter: Martin Mirchev


{{DomXmpParser.parseEndPacket()}} reads {{charAt(5)}} after checking only that 
the processing instruction data starts with {{end=}}. Metadata that ends with 
{{<?xpacket end=?>}} or {{<?xpacket end='?>}} makes {{DomXmpParser.parse()}} 
throw a {{StringIndexOutOfBoundsException}} instead of the 
{{XmpParsingException}} ({{XpacketBadEnd}}) that the method throws for other 
bad values:

{code}
java.lang.StringIndexOutOfBoundsException: String index out of range: 5
    at java.base/java.lang.StringLatin1.charAt(StringLatin1.java:48)
    at org.apache.xmpbox.xml.DomXmpParser.parseEndPacket(DomXmpParser.java:929)
{code}

To reproduce, parse {{<?xpacket begin="" 
id="W5M0MpCehiHzreSzNTczkc9d"?><x:xmpmeta xmlns:x="adobe:ns:meta/"><rdf:RDF 
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#";></rdf:RDF></x:xmpmeta><?xpacket
 end=?>}} with {{new DomXmpParser().parse(bytes)}}; with {{end='w'}} instead it 
parses. Each result was the same in 3 runs.

A possible change; with it, both truncated forms throw {{XmpParsingException}} 
and {{end='w'}} and {{end="r"}} still parse:

{code}
--- a/xmpbox/src/main/java/org/apache/xmpbox/xml/DomXmpParser.java
+++ b/xmpbox/src/main/java/org/apache/xmpbox/xml/DomXmpParser.java
@@ -926,6 +926,11 @@
         // should be ignored
         if (xpackData.startsWith("end="))
         {
+            if (xpackData.length() < 7)
+            {
+                throw new XmpParsingException(ErrorType.XpacketBadEnd,
+                        "Expected xpacket 'end' attribute with value 'r' or 
'w' ");
+            }
             char end = xpackData.charAt(5);
             // check value (5 for end='X')
             if (end != 'r' && end != 'w')
{code}

The module unit tests pass with the change (io, fontbox, xmpbox, pdfbox; trunk 
fff879b).

This issue was found by an LLM-assisted tool and verified manually.




--
This message was sent by Atlassian Jira
(v10.3.26#10030026)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to