Opensolr Changelog — Recent Updates & Improvements

Search Mar 17, 2026

Fix Fixed "Did you mean?" spellcheck suggestions — previously only corrected a single word in the query (e.g. "downloda invoicesr" → "download"). Now corrects all misspelled words and shows the full corrected query (e.g. "download invoices"). Works across all Search pages including hybrid vector search.

Fix Fixed a bug where forked worker processes did not inherit the --uri flag from the parent Drush process. Without --uri, Drush defaults to http://default as the base URL, causing URL-dependent tokens like [site:url-brief] in Solr index field configurations to resolve incorrectly. Affected fields (e.g. ss_index_source, site, ss_url) would contain the wrong hostname. All spawned workers now correctly inherit the site URI. Update to v2.5.1 and reindex affected indexes.

Fix Fixed an issue where visiting any admin page without being logged in showed a raw error message instead of redirecting to the login page. Non-authenticated users are now properly redirected.

Fix Fixed Click Analytics and No Results tabs returning 403 errors when accessing Query Analytics via the shared search link. AJAX data calls now use the correct endpoints based on how the page is accessed.
Fix Fixed Query Analytics tabs (No Results, Click Analytics) not loading for indexes without Elevation Rules enabled. The tab content and all interactive features now work correctly regardless of whether elevation is configured on the index.

Fix Fixed a bug where the internal rtf flag was not always stripped before sending documents to Solr, causing an "unknown field rtf" error that failed the entire batch. The flag is now reliably removed regardless of how the value was encoded (boolean, string, or integer).
Fix The Data Ingestion API now correctly sets the content_type field for every document. If you don't provide one, it defaults to text/html so your documents appear in the web search tab instead of being classified as media. When using rtf:true, the MIME type is auto-detected from the actual file content — PDFs get application/pdf, Word documents get their proper DOCX type, and so on.

Fix Fixed a critical bug where pinning multiple documents to the same query only applied the last one. Solr was receiving the document IDs in the wrong format — now all pinned results correctly appear at the top in the order you set them.

Fix Error Log now catches all errors for your index, even when multiple different issues occur at the same time. Previously, errors with the same top-level message but different root causes were silently merged, and some Solr errors logged without the index name prefix were missed entirely. The dashboard now shows each distinct error with its actual root cause and a compact, readable stack trace.

Fix Document indexing now works reliably for all major office formats. DOCX, DOC, XLSX, XLS, and PPTX files are fully supported with proper text extraction — including tables, headers, footers, and speaker notes. Previously, many documents were indexed with empty text due to format misdetection and encoding issues.

Fix Fixed tag field generation during crawl. The tags and title_tags fields used for autocomplete and spellcheck were being stored with raw special characters intact, which could produce noisy or broken suggestions. They are now properly cleaned — special characters stripped, whitespace normalized — so autocomplete and spellcheck results are cleaner.

Fix Search results no longer show Jan 01 1970 as the date when a document has no real creation date. The date badge is now hidden entirely when no valid date is available.
Fix Fixed title and description highlighting on complex search queries. When a query contained boolean operators like AND or OR, only the full-text body was being highlighted — titles and short descriptions were left plain. Highlighting now works consistently across all result fields, regardless of how complex your query is.

Fix Automatic cleanup of stale crawler lock files. If a previous crawl crashed or was interrupted, leftover lock files could silently prevent the next run from starting — the crawler would launch but do nothing. Resume now detects and removes stale lock files before starting, so scheduled cron runs and manual resumes always work reliably.

Fix Fixed search results showing blank descriptions on some indexes. When the Solr highlighter could not find matching terms in the description field, results appeared with no text at all — even though the document had a perfectly good description. The fallback logic now correctly shows the original description when no highlighted snippet is available.

Fix Fixed the Flush to Solr button in Web Crawler always reporting buffer is empty even when documents were actually flushed. It now correctly reports the number of documents flushed, and automatically commits after flushing so your documents become searchable immediately — no more waiting for the next auto-commit cycle.