[Release] GroupDocs.Redaction for Java 26.7

GroupDocs.Redaction Product Family
, ,
1

Dear GroupDocs users, we’re pleased to announce GroupDocs.Redaction for Java 26.7. This release lets you explicitly specify a document’s type when opening it from a stream or a file, switches HTML and Markdown redaction to the Aspose.Words engine, and improves image metadata handling. The changes are backward compatible — existing code continues to work unchanged, and the new file-type API is opt-in.

Feature — specify the file type when opening a document
You can now tell the engine which format to use instead of relying on auto-detection. LoadOptions accepts a FileType , which is useful when a document comes from a stream without an extension, or from a path whose extension does not match the real content.

import com.groupdocs.redaction.Redactor;
import com.groupdocs.redaction.FileType;
import com.groupdocs.redaction.options.LoadOptions;

// A stream without an extension is loaded as PDF
try (Redactor redactor = new Redactor(stream, new LoadOptions(FileType.getPDF()))) {    
// work with the document
}

When LoadOptions.getFileType() is set to a value other than FileType.getUnknown(), format detection is skipped and the specified type is used. If the type maps to an unsupported format, a DocumentFormatException is thrown.
Scope.

  • Applies when opening from both file paths and streams; a mismatched file extension no longer overrides the specified type.
  • Default behavior is unchanged: with no FileType set (or FileType.getUnknown()), auto-detection works exactly as before.
  • New format recognition was added: PowerPoint templates POTX/POTM and OpenDocument presentation template OTP (Slides), and OpenDocument text template OTT (Words).
  • Backward compatible — a new LoadOptions(FileType) constructor and new FileType constants are added; no existing signatures changed. (REDACTIONNET-572)

Change — HTML and Markdown are now processed with Aspose.Words
HTML and Markdown documents are redacted through the Aspose.Words engine instead of Aspose.Html. Markdown is redacted by traversing the underlying Words document, while HTML is redacted as text and written back preserving its original character encoding.
Scope.

  • No public API changes — Redactor, redactions, and save options are used exactly as before for .html, .htm, and .md files.
  • Redaction of text and regular expressions in HTML/Markdown continues to work; rasterization and preview generation are produced through the Words engine.
  • The Aspose.Html-based markup adapter was removed as part of this change. (REDACTIONNET-775)

Fix — image metadata (EXIF) processing and Aspose.Imaging compatibility
Image metadata handling in ImagesDocument was hardened and aligned with current Aspose.Imaging behavior.

  • EXIF extraction now covers additional properties (IsBigEndian, ISOSpeedValue, MakerNoteData, MakerNoteRawData, MakerNotes, Thumbnail, XResolution, YResolution).
  • Property lookup prefers members declared directly on the type with a safe fallback, and any property that throws while its value is read is now skipped instead of aborting the whole metadata operation.
  • The EXIF tag list was cleaned up to match tags currently surfaced by the library, DICOM images use dedicated save options, and multipage images are saved correctly.

Symptom that is resolved. Reading metadata from certain images previously failed with a reflection error instead of returning the readable metadata. (REDACTIONNET-740, REDACTIONNET-721)

Resources