github papra-hq/papra @papra/app@26.7.0

latest release: @papra/lecture@0.5.2
2 hours ago

Minor Changes

  • #1474 aa26524 Thanks @CorentinTh! - API breaking change: The route to trash a document DELETE /api/organizations/:organizationId/documents/:documentId now returns a 204 No Content status code instead of { success: true } JSON body.

  • #1513 1f1a9ad Thanks @CorentinTh! - Added an API endpoint to reprocess existing documents POST /api/organizations/:organizationId/documents/:documentId/reprocess. This endpoint allows re-running content extraction + auto tagging and tagging rules on existing documents.

Patch Changes

  • #1531 4ac1d68 Thanks @CorentinTh! - Added Hungarian translation

  • #1468 4fef955 Thanks @CorentinTh! - Fonts are now bundled with the app, so that it can be used offline, and no longer depends on Bunny Fonts. A step towards making Papra fully air-gapped.

  • #1461 5d3063b Thanks @bartdenhoed! - Fixed missing refresh of the updatedAt db attribute.

  • #1529 1c993e4 Thanks @CorentinTh! - Add support for {{organization.name}} in document storage key patterns to organize files into folders named after your organizations.

  • #1469 347029d Thanks @CorentinTh! - Fixed OCR extraction for images whose rotation is stored in EXIF metadata.

  • #1525 7c9a96f Thanks @CorentinTh! - Added optional storage key synchronization so document renames are reflected in the underlying storage. Enable it with DOCUMENT_STORAGE_KEY_SYNC_ENABLED=true when using pattern-based storage keys.

  • #1519 82778df Thanks @CorentinTh! - Fixed document action menus wrapping longer translations by allowing the menu to expand.

  • #1532 50cbe2c Thanks @CorentinTh! - You can now configure Docling OCR languages and other conversion options with DOCLING_OPTIONS, using a JSON object such as {"ocr_lang":["en","fr"]}.

  • #1503 2993c6e Thanks @CorentinTh! - Security fix: Prevent custom OAuth providers from creating new accounts when registration is disabled, while allowing existing users to sign in.

    Thanks to 白墨 for the responsible disclosure of this issue.

  • #1528 13a6d4a Thanks @CorentinTh! - Added document.date and document.createdAt to storage key patterns, so you can organize files by document date or creation date. Documents without a date use no-date (and can be configured with the default transformer).

  • #1517 956684c Thanks @CorentinTh! - Added a button to reprocess a document in the documents content tab.

  • #1428 76c4501 Thanks @jakubM817! - Add Slovak (sk) translation for the app.

  • #1511 71fdd27 Thanks @CorentinTh! - Fully disabled AI credit accounting by default for self-hosted instances to silence "Failed to register AI credit usage" logs.

  • #1470 a58b4f0 Thanks @CorentinTh! - Silenced the content extraction warning about DPI.

  • #1534 a03dc5a Thanks @CorentinTh! - Fix an issue that prevented reinviting users who left or were removed from an organisation.

  • #1545 193d860 Thanks @CorentinTh! - Added an option to restrict registration to specific email domains using AUTH_ALLOWED_EMAIL_DOMAINS. Forbidden domains still take precedence.

  • #1502 c16661e Thanks @CorentinTh! - Removed the default value for the intake email webhook validation secret to prevent ingestion of documents from untrusted sources when no secret is provided. Users must now explicitly set a secret using INTAKE_EMAILS_WEBHOOK_SECRET.

    Thanks to 白墨 for the responsible disclosure of this issue.

Don't miss a new papra release

NewReleases is sending notifications on new releases.