Apache OpenNLP's DictionaryEntryPersistor class initializes a SAXParserFactory without enabling FEATURE_SECURE_PROCESSING or disabling DTD processing, leaving external entity resolution and DOCTYPE declarations enabled. When the create(InputStream, EntryInserter) method is called, this insecure parser configuration allows an attacker who can supply a crafted dictionary file (such as a stop-word list or domain dictionary) containing a malicious DOCTYPE declaration to exploit XML External Entity (XXE) injection. This can lead to local file disclosure via file:// entity references or server-side request forgery via http:// entity references during SAX parsing, before any dictionary entries are processed. This issue contrasts with other XML parsing paths in the codebase that correctly disable these features. The vulnerability affects Apache OpenNLP versions before 2.5.9 and before 3.0.0-M3. The recommended mitigation is to upgrade to version 2.5.9 or 3.0.0-M3. For users unable to upgrade immediately, it is advised to ensure dictionary files come from trusted sources and to implement input validation to reject XML containing DOCTYPE declarations before parsing.

Successful exploitation allows an unauthenticated attacker who can supply a crafted dictionary XML file to cause local file disclosure or server-side request forgery via XXE during XML parsing. This can lead to confidentiality breaches of local files and potential interaction with internal systems through SSRF. The vulnerability has a CVSS score of 9.1 (critical), indicating high impact on confidentiality and integrity without requiring privileges or user interaction.

A fix is available by upgrading Apache OpenNLP to version 2.5.9 for 2.x users or 3.0.0-M3 for 3.x users. Users who cannot upgrade immediately should ensure all dictionary files are sourced from trusted origins and implement input validation to reject any XML containing DOCTYPE declarations before parsing with the vulnerable Dictionary(InputStream) constructor.