About the organisation page with ordered rich-text sections and admin CRUD
An affiliation on every profile, asked for on the edit page and sent with the profile's JSON
Cookieless analytics (Plausible or Matomo): no consent banner, the script with the request's CSP nonce, never on a page that holds a token nor on the sensitive routes
Release-gated continuous delivery: follows a registry image channel and redeploys the product safely, with backup, health check and rollback
Daily encrypted backups of the database and the file volumes, off the host, with retention, a check of the files the database names and a restore drill
An image captcha the product draws itself, with digits read out for whoever cannot see it, no keys and no third party, as the provider of captcha
Proves an address belongs to its account: a one-time link opening a page with a button that confirms it, never signing in or switching an account on, a page that mails a new link, the link that moves an account to a new address, and the link an account carried over from an older site is mailed at its first sign-in
Contact form with stored messages (public form + admin inbox)
Datasets: metadata, authors, versions, DOI landing pages (with FAIR metadata), BibTeX, QR and downloads. Their files come from hubfile, stacked on top
REST API v1 and v2 of the datasets and their files, out of dataset: fine token scopes, public reads of what is published, tokens limited to some datasets, versioned schemas and the audit trail of every write; v2 with public ids, cursor pages, problem+json, ETags and publication as a resource
Offers of a dataset and its versions to another account (DL-17): an inbox in the account area, a mail with a signed link, one answer whether or not the account exists, offers that lapse, locked answers, one live offer per lineage and a daily limit per account
The page of a hub's search: /explore, /api/v1/search with its declared schema and the header search box, over the public sources the features register in the framework's search registry (a hub's datasets and files, their facets and orders), which the search feature browses with Elasticsearch or in memory. It names no kind of material, so another hub explores its own.
Each person's folder in the browser: list, resumable upload, download as file or zip, preview, folders, rename, move and delete, done as the folder's owner by filesd, a standard-library server; where the folder is comes from whichever feature keeps folders
The shell every hub of the OpenHubs line shares: the application and public layouts (shell('app'), shell('public')), the sidebar drawn from the menu registry, the header with its search slot, the footer, the flashed messages and the 400, 401, 403, 404 and 500 pages, on theme's site object and tokens, with AdminKit's Bootstrap, Inter and the chrome's icons from npm
The files of a dataset, of any type a hub takes: their registry of file types, storage by key, upload staging, view, raw and download, and the file workbench of the dataset page
The icons every page draws: Lucide and brand logos from simple-icons, installed with npm, drawn by the server with icon()
Sign-in provider accounts (ORCID, GitHub…) linked to users, and the one OAuth flow every provider registers on (register_oauth_provider): sign-in, connect, and new accounts through /account/complete-signup, a typed address opening one only through a mailed link. With the identities page, disconnecting, the rule that keeps a product without signup closed, and an optional hint on the login page
A folder with a quota on the lab's shared storage for every approved account, created over SSH by the storage server's helper, with a page for each person and one for administrators
Media library to upload and manage files (the WordPress Media Library analogue)
What each machine is doing, live: an agent per machine posts its CPU, memory, GPU and processes; signed-in people see them next to what is booked
phpMyAdmin database management UI for development
The name and surname of every account, with an edit page and a JSON endpoint that other features extend with fields, sections and keys
Public landing page with the unauthenticated home route
Like and dislike votes on published datasets, counted per version lineage
Product-wide search over whatever the installed features declare searchable, with the owning feature deciding at request time who may see each hit, and a faceted browse of the public sources (words as they are typed, facets with their counts, orders, pages) answered from Elasticsearch when the product has it and from the sources' documents in memory when it does not
Filesystem-backed session storage via Flask-Session
Redis-backed session storage via Flask-Session
Key-value site settings storage used by theme and admin
Hero image slider for the home page (layout.hero hook + admin CRUD)
Hub statistics derived from the records: the /statistics dashboard and the home card, with datasets counted by lineage and by version, distinct authors, a deduplicated registry of use (a visitor a day, no cookies, robots, owners and drafts left out) that holds the daily history carried over from uvlhub 2.12 with its public note, slots the features fill with KPIs, tables, sections and exports, and the CSV/JSON exports at uvlhub 2.12's addresses
Research and SPL tools with a public list grouped by status, detail pages and admin CRUD