Version 1.20 (January 20, 2022)

Applications

• added support for PDF/A-4 including Level F (file attachments) and Level E (engineering)
• added more informative logs in batch processing
• added new parameter to specify the default validation profile in case of missing standard identification in XMP Metadata

PDF Model

• extended the model to support PDF/A-4 rules

PDF Parser

• more robust handling of malicisious PDF documents
• improved parsing of PostScript and CFF fonts

Validation

• allow empty Lang Alt arrays in XMP metadata
• excluded All and None colorants from the PDF/A-2 and PDF/A-3 requirement to have the same tintTransform and alternateSpace
• disabled JPEG2000 colr box checks in case of explicitly defined ColorSpace in the Image dictionary
• fixed validation of predefined XMP value types if they are redefined in the extension schema
• validate XMP URL type as Text
• fixed CIDSet and CharSet validation for PDF/A-1
• fixed validation of the permissions dictionary in PDF/A-2 and PDF/A-3
• added validation of Lang against RFC 1766 regular expression
• fixed validaiton of permitted transfer functions in Halftone dictionaries

Corpus

• added test corpus of ~600 new atomic documents covering PDF/A-4 specification
• extended PDF/A-2u tests on ToUnicode mapping and character encodings in simple fonts

Core library

• added support of Java versions from 11 to 16
• fixed validation of documents with non-PDF extension in multi-process mode