<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=6889170&amp;fmt=gif">
Skip to content
  • There are no suggestions because the search field is empty.

Release Notes - 26.07.00

What's New in v26.07.00 – July 14, 2026

Improvements

The Nebula front end has been upgraded to a current framework generation, delivering improved responsiveness, more performant Document List, a stronger security posture, and a modern foundation for faster delivery of future features.


Resolved Bugs

  • Documents containing line breaks, special characters (including emoji and non-Latin scripts such as Japanese), and heavy punctuation now index dependably and remain fully searchable, resolving cases where such content previously failed to index. (Your original documents are never altered — only the search index is affected.)
  • The indexing pipeline has been reworked for greater stability, with smarter retry handling, clearer per-document error and status reporting, and support for reindexing only the content that's missing — so jobs are less likely to stall or require a full re-run.
  • Corrected a condition where an underlying search-index error could cause indexing to loop indefinitely; jobs now recover gracefully and continue.
  • Keyword categories that could previously remain stuck on "Running" indefinitely now correct themselves automatically, with a nightly cleanup to clear any lingering stuck statuses.
  • Running PII/PHI searches after a collection or documents were deleted no longer produces errors; related analytics data is now cleaned up automatically when items are deleted.
  • Fixed a rare display issue where quickly switching between saved searches on a slow connection could show the wrong search description.
  • Documents imported from third-party sources whose file-extension metadata contained a leading dot now parse successfully in Nebula Spreadsheet Redactions instead of erroring out.
  • Inline images embedded in encrypted or digitally signed (S/MIME) emails are now correctly excluded during processing, consistent with standard emails — producing cleaner, more predictable document sets.
  • Predictive coding jobs that hit an unexpected error now fail cleanly with a clear, recorded reason instead of getting stuck "in progress," reducing manual intervention and speeding up resolution.
  • Fixed a defect where the notification bell and notifications dialog could fall out of sync (e.g., an alert showing with an empty dialog, previously requiring a re-login). Notifications now stay consistent and recover automatically from transient connection errors.