FileDeck – Document Library & Download Manager

修改日誌

2.112.1

• Pro: when scanned-PDF OCR can’t finish a very long scan in time, it now keeps the pages it read, which are searchable, instead of failing and retrying. The Search Index tab says how many documents were read only in part.
• Pro: scanned PDFs are no longer sent for OCR under their own file names, and on FileDeck’s OCR service a licensed site uses its license even if an older OCR key is still saved.
• Security hardening for scanned-PDF OCR.

2.112.0

• Pro: the in-dashboard Help panel keeps your conversation for 90 days, so it is there in another tab or when you come back, and it lets you know when a person has replied to you by email. The Help button shows when a reply is waiting, and New conversation ends it.

2.111.0

• AI tier: on OpenAI, Ask this library and AI summary suggestions now default to gpt-6-luna, ahead of OpenAI retiring the old default (gpt-4o-mini) on 14 January 2027, and work with OpenAI’s newer reasoning models.
• AI tier: on Anthropic, Ask this library and AI summary suggestions now default to Claude Haiku 5.5, at a tenth of the per-token price of the previous default on your own API key. If you chose a model yourself, it is kept. Both also work with the newest Claude models, which think before they answer.
• AI tier: when an Ask this library answer or an AI summary suggestion fails, the Ask AI settings tab now shows the provider’s own error until a request succeeds again.
• AI tier: after your first scan, the accessibility report screens new and replaced PDFs automatically, and it says a scan is in progress only while one is scheduled.
• Pro: a Help button for administrators on FileDeck’s screens in wp-admin. An AI assistant answers questions from the FileDeck documentation, and passes a question to a person by email when the documentation doesn’t cover it.
• Deactivating FileDeck because it wasn’t what you were looking for? The feedback form now lets you pick what you were after in one click, and says so when FileDeck already does it.
• Access-rule improvements in wp-admin.

2.110.3

• Pro: on sites where WordPress’s scheduled tasks don’t run, documents you add or edit are now indexed for content search as you save them, instead of waiting for a background task that doesn’t run there. A large batch at once, such as an import, starts a rebuild you can finish from the Search Index tab, and index work left waiting from before is picked up as you use the admin. The Search Index tab says when it is working this way.

2.110.2

• Pro: a search-index rebuild that has stopped moving can finish from your browser. On hosts that switch off WordPress’s scheduled tasks, a rebuild could sit at “0 of N documents processed” indefinitely; the Search Index tab now says when a rebuild has stalled, and “Continue in this browser” runs it from the page with live progress. A file too large or damaged to read in time no longer stops a rebuild: the document is indexed by title and description and listed on the tab. wp filedeck reindex also ends any queued rebuild.

2.110.1

• Dates show on sites whose theme or a plugin hides post dates. A library’s Date and Updated columns, its date filter and year grouping, and a document page’s Published and Updated lines were empty there; they now show each document’s own date. Pro: review reminders, retention and transmittals take the document’s date on those sites too.

2.110.0

• Pro: a client portal. Add a client under Documents → Clients and FileDeck creates their account, a private area and your portal page, and emails them an invitation to choose their own password. Clients sign in on your site rather than the WordPress dashboard, see only their own documents with what is new since their last visit, and manage their own details. The Clients screen shows who has accepted and when each client last visited, and lets you resend invitations, import clients from a spreadsheet and preview the documents a client sees. Add it to any page with the Client portal block or the [filedeck_client_portal] shortcode.
• AI agents can reach your library. On WordPress 6.9 and later, FileDeck’s search, document lookup and category list show up for agents through the MCP Adapter or WordPress’s REST API; before, they did not. Agent searches leave out expired documents and, in Pro, superseded ones.
• Switched from BSK PDF Manager? Its shortcodes already in your pages now show your FileDeck library, document link or viewer once the old plugin is deactivated.
• Documents with a file attached from another plugin’s private folder (Memphis Documents’ uploads/mdocs, for example) now download and preview; before, the server refused them on most hosts.
• Pro: OneDrive and SharePoint connections accept the app’s Directory (tenant) ID, so an app registered for one organization (Entra’s default) can connect, and the SharePoint browser lists every site, not only the first 50.
• Pro: an existing [filedeck_my_documents] page shows signed-out visitors the portal’s sign-in card, with password reset, instead of a link to the WordPress login. Emails about new documents link through your portal page once you have one, and clients can turn them off on their Account tab.
• Pro: Library Health’s “Attach them now” works through large libraries without getting stuck. It takes only the documents it can attach and lists the rest apart, with what to do about them. Its scan also finds local links written with %20 for spaces.
• Columns that keep their width in the library table no longer break a word such as “Download” across two lines.
• Access-rule and reliability improvements in wp-admin and the media library.

2.109.2

• Pro: a CSV import recognises a Media Library file whose name contains spaces and attaches it, so the document shows its size and type. Before, such rows were kept as links. Documents → Settings → Health → Attach them now attaches the ones an earlier import left as links. One whose file is in another plugin’s private folder stays a link: move those documents to the Trash first, then import those rows again, and a file in your Media Library is attached.

2.109.1

• Access-rule and reliability improvements in wp-admin. If your authors or contributors attach files to restricted documents, those files no longer appear in their Media Library or media picker; each one stays attached and is shown in its document’s editor.
• The opt-in usage ping dates a library imported from another plugin from its import, or from the day FileDeck was installed if that is later, not from the original items’ dates.

2.109.0

• Pro: when a one-click import can’t bring an item across, the Migrate screen names it and says why (its file is missing, it is in the old plugin’s trash, and so on), with an Import again button. Documents → Settings → Health shows the same list. Stopping an import keeps what it had done so far, that list included.
• The “Added in the last…” filter now works on large libraries too.
• The Elementor Library widget offers your custom-field and taxonomy columns.
• Pro: share buttons switch off with the rest of Pro when a licence lapses.
• Pro: duplicating an imported document no longer carries over its link to the old plugin’s item.
• The opt-in usage ping (sent only if you opted in) now carries the install, first-upload and first-publish dates (day only), whether last week’s import had failures, a bucketed count of documents whose text recognition stopped on an error, and a random identifier for the site. The External services FAQ lists every field.
• Updated Freemius SDK to the latest version.
• The opt-in and pricing screens show the FileDeck icon.
• Access-rule and reliability improvements in wp-admin.

2.108.0

• New: the Library builder (Documents → Library builder) designs a library with a live preview on your own site, in your theme, as a visitor or as you. Create the page in one click or copy its shortcode; paste an existing shortcode to keep editing it. The Document Library block opens the same builder, and its preview in the editor now uses FileDeck’s styles.
• After a one-click import from Download Monitor, WordPress Download Manager, Simple Download Monitor, Shared Files or Simple File List, the old plugin’s download links, item pages and shortcodes (where it had them) keep working once you switch it off — in the free plugin.
• Document pages list a document’s reference, published and updated dates, categories and tags above the Download button; choose which under Documents → Settings → Display → Document pages, including an optional download count.
• Pro: importing from Download Monitor brings older versions into each document’s version history; importing from WordPress Download Manager brings every file in a multi-file package. Run an import again to fill in what an earlier one left out.
• Pro: client-area emails reach the people who can open a new document (up to 200 per document) — named people, group members and email-domain matches — including batch uploads, block-editor publishing, scheduled documents and sub-folders. Imports stay quiet.
• The default-columns setting keeps the Reference and Views columns on save.
• Security and reliability improvements.
• The wordpress.org listing shows the full description and the external-services disclosure; both had been cut short.

2.107.2

• The plugin’s description on wordpress.org fits the directory’s length limit again, so the whole of it shows; it had been cut short. Older changelog entries moved to changelog.txt in the plugin folder. This release changes the listing only; the plugin works as before.

2.107.1

• The plugin’s description on wordpress.org now says what FileDeck is used for: safety data sheets and product documentation, meeting minutes and newsletters by year, school and staff policies, client file sharing and public reports. This release changes the description; the plugin works as before.

2.107.0

• Pro: a “Test connection” button on the Scanned OCR settings tab checks that the saved Service URL reaches an OCR service, and says why when it doesn’t.
• Pro: pasting an email redirect link as the OCR Service URL (the long google.com/url?q= or Outlook Safe Links form an email client can turn a link into) now saves the address inside it. A Service URL with a query string, or one that isn’t https or points at a private network, is refused with a message saying why. Until now such a URL was either saved as-is, sending scans to the wrong address, or silently not saved.

2.106.1

• Instant search finds titles written in your site’s language. On a Danish site, “møde”, “bestyrelsesmøde” and “ø” now find “Bestyrelsesmøde”; on a German site, “prüfung” and “pruefung” both find “Prüfungsordnung”, and “strasse” finds “Straße”. Letters such as ø, æ, å, ß and umlauts could make a search miss, because the library and the search box spelled them differently.
• Instant search finds titles in Cyrillic, Greek, Korean and Japanese as you type them. Letters such as й and ё, accented Greek vowels, Korean syllables, Japanese kana with voicing marks, and capital Cyrillic and Greek letters could make a search miss.

2.106.0

• Pro: one-click imports bring across a password set with WordPress’s own Password protected visibility. The imported document asks for the same password before its file can be downloaded, and FileDeck protects its own copy of the file. For documents imported before this release, the Migrate screen lists any that don’t yet ask for their item’s password, with a button to add the same passwords.
• Pro: Settings → Health lists documents that use the same file as a restricted document without the same restrictions, with a link to edit each, so you can decide whether the file should stay private.
• Security and reliability improvements across the library, document pages, imports, exports and embeds, in the free plugin and Pro. If your site uses a page cache, clear it once after updating.

2.105.1

• Pro: on sites that run WordPress on SQLite (for example WordPress Playground, Studio by WordPress.com, or the SQLite Database Integration plugin), search could match words a document no longer contains. Updating fixes this, re-indexing in the background where needed.
• Pro: on a server missing a PHP extension FileDeck needs to read PDFs (zlib, iconv or mbstring), PDFs are no longer counted as scans or sent for OCR, including PDFs an earlier version had already counted: there, a PDF with a text layer can’t be told apart from a scan. The Search Index tab names the missing extensions. On every server, each PDF is now checked again just before it is sent, and one that turns out to have a text layer stays on your site.
• Pro: a document moved to a byte-identical copy of its file keeps its OCR text instead of dropping it.
• Pro: switching scanned-PDF OCR back on no longer sends scans that were still waiting for room in the queue when it was switched off, including when it was switched off while the license had lapsed.

2.105.0

• Pro: Settings → Health lists restricted documents whose imported original is still public. A one-click import gives a restricted document its own protected copy and leaves the original where the old plugin keeps it, so anyone with a direct link to that original can still open it. Each row names the file and offers to delete it, or to move it into protected storage if you’d rather keep it. Deleting is offered only when the file isn’t used by another document and isn’t linked from a post, page or menu (in its content, a custom menu link, Elementor’s page data or as its featured image), and isn’t your site icon or logo; links kept anywhere else aren’t checked, so look before you delete. Either way, the old plugin’s link to that file stops working.
• Pro: Settings → Health also lists restricted documents whose file could not be moved into protected storage, and imported documents that arrived without their file. These checks include drafts, and the weekly Health email reports them. An import that leaves a restricted file unprotected says so in its summary, with a link to Health.
• Pro: running a WP Customer Area import again finishes a document that an interrupted run left without its file. The document stays a draft for you to publish.
• Pro: re-running a WP Customer Area import leaves the client areas it made earlier, and the people you’ve added to them, as they are.
• Pro: a Download Monitor download whose file is in Download Monitor’s protected dlm_uploads folder gets its own copy in FileDeck, so its Download button works. It used to link the file where Download Monitor keeps it, which the server refuses to serve.
• Security and reliability improvements to the one-click importers.

2.104.1

• Pro: signed-in members get the restricted documents they’re allowed to have when lead capture or terms acceptance is switched on. Leaving their details or accepting the terms used to end in an error instead of the file.
• Pro: a member’s zip includes the restricted documents they selected, and searching inside documents finds text in restricted documents they’re allowed to read. Both used to leave those documents out.
• Pro: daily download limits, terms acceptance, lead capture and the form and bot checks apply to restricted documents’ Download buttons the same way they apply to every other document. Share links you create yourself (single-use links and transmittals) still skip them. If you set daily download limits, downloads of restricted documents now count towards them, whether by members or by visitors who unlocked a password-protected category.
• Pro: the [filedeck_download] button and an embedded document’s download link show the lead form, terms or check when one is switched on, instead of an error.
• AI plan: a view-only restricted document opens for the people it’s shared with, from the library and from its own page. A view-only document’s own page shows the lead form, terms or check when one is switched on, and with lead capture on, a view-only document opens after the form instead of showing an error.
• Security and reliability improvements.

2.104.0

• Pro: with scanned-PDF OCR switched on, each scanned PDF you publish is now sent for OCR automatically, so it becomes searchable without anyone going back to the Search Index tab. Before, a scan added after OCR was switched on waited until someone pressed Send again. Scans that were already in your library when you switched OCR on, and scans added on the Import & Export screen (bulk add, including files dropped there, a CSV import or a migration), still wait for you to press Send. New scans go one at a time in the background, within the same 100-a-day fair-use limit as before.
• Pro: replacing a scanned document’s file with a new scan now makes the new file searchable. Before, the old scan’s OCR text stayed in the search index for the new file, and the new scan wasn’t sent. This also holds when the file is replaced while the old scan is still being read.
• Pro: a scanned PDF that you unpublish or move to the trash before its turn in the OCR queue is no longer sent.

Older releases are listed in changelog.txt.