Fixes & Improvements
-
Improved history stability for large tables: History archiving, cleanup, and compression now run automatically in the background.
-
Improved record history pagination and reading: Fixed pagination and timeout issues for more stable history viewing.
-
Improved history handling for deletion and import: Fixed history issues during permanent deletion and imported record creation.
-
Fixed App Builder public app login redirect: Public apps no longer redirect to Teable login when login is disabled.
-
Improved billing stability: Charges and refunds are more reliable under high concurrency, reducing duplicates and failures.
-
Fixed Date field display issue: Date-only fields no longer include time values in formulas or copied results.
-
Fixed date filtering issue: Date conditions like “after or equal” are more stable, preventing automation script failures.
-
Fixed attachment preview issue: Onboarding attachments can be previewed more reliably after entering the app.
-
Improved AI streaming responses: AI-generated content now streams more smoothly, reducing frontend waiting and lag.
-
Improved AI content saving reliability: Generated content is better preserved even if streaming is interrupted or fails.
-
Fixed linked record display issue: Linked records in the first column no longer temporarily show as “Untitled” after Number field updates.
-
Improved required field setup prompts: Failed required field saves now show clearer messages about missing configuration.
-
Added localized error prompts: Required field validation messages now follow the user’s language.
-
Fixed App Builder public access issue: Published App Builder apps no longer redirect to Teable login when login is disabled.
-
Improved billing stability: Charges and refunds are more reliable, reducing delays, duplicates, and refund errors.
-
Fixed date field display issue: Date-only fields no longer unexpectedly include time values in formulas or copied results.
-
Fixed date filtering issue: Automation scripts no longer fail when using date conditions like “equal or after.”
-
Fixed onboarding attachment preview issue: Onboarding attachments now preview more reliably in apps, reducing expired-link loading failures.
-
Improved AI app creation: AI now turns user needs into clearer, structured prompts for more accurate App Builder generation.
-
Improved Agent Computer startup speed: Reduced unnecessary startup checks so AI Agent returns the first response faster.
-
Improved connector menu display: The “Connect everything” entry is clearer and less likely to be confused with integrations.
-
Improved Agent file reading: Large files are read more reliably, reducing failures caused by excessive content length.
-
Fixed workflow record retrieval issue: Missing or invalid
create timeno longer causes workflow record retrieval failures. -
Standardized workflow node time format: Node timestamps now use ISO format for more stable validation and service handling.
-
Fixed required single-select field creation issue: Existing records now receive defaults before required constraints are applied.
-
Fixed Enterprise License validation issue: Existing self-hosted License Keys remain valid after upgrades, avoiding false “license is invalid” errors.
-
Fixed duplicate website onboarding attachments: Website-provided attachments no longer appear twice in auto-submitted prompts.
-
Fixed false “unpublished configuration changes” prompt in App Builder: Published apps no longer require redeployment when configurations are unchanged.
-
Improved environment variable change detection: Encryption metadata or timestamp-only changes are no longer treated as configuration changes.
-
Fixed Airtable import failures caused by duplicate select options: Imports are more stable when trimmed option names conflict.
-
Fixed duplicate message sending: Resolved cases where Teable AI could send duplicate messages.
-
Fixed required single-select record creation issue: Records can now be created when required single-select fields have defaults.
-
Improved recovery entry for abnormal tables: Recovery actions remain available when tables enter an error state.
-
Strengthened comment permission controls: Comment details, reactions, updates, deletion, and references are now strictly scoped.
-
Improved Link field data repair: Missing linked records are handled more reliably during cleanup or repair.
-
Improved record history processing stability: Fixed cases where history processing could stop, allowing pending data to continue automatically.
-
Improved large history table processing: Optimized memory usage when processing large history datasets for long periods.
-
Improved temporary file handling for large tables: Reduced temporary disk usage for tables containing large JSON content.
-
Improved data writing and refresh reliability: Better error handling prevents missing data, unreleased resources, and incorrect error attribution.