Data, Files and Jobs

Everything here is opt-in: an app that uses none of it pays nothing.

Tables

// src/db.rs
pub static USERS: Table<User> = Table::saved("users")
    .unique("email", |u: &User| &u.email) // no two rows share it
    .migrate(|row| rename(row, "mail", "email")) // old rows, as they are read
    .live(); // each change is sent on channel "users"

• set(id, v) -> Option<()> replaces a row; clear() takes every row out (ids are not given again).
• unique: by(&email) finds the row without a scan. try_add, try_set and try_update return a 422 on the field (email: is taken), for an action to return with ?. add, set and update panic (a 500) instead. Use try_update to change a unique field.
• On a #[model] field: #[unique] is the same as .unique(..); #[json(default)] or #[json(default = expr)] fills a field old rows lack; #[json(was = "old")] reads it under its old name.
• migrate(|v: &mut Value| ..) runs on each stored row's JSON when it is read (at load, and by WISP_STORE_POLL), before it becomes the type.
• live(): after each change, wisp::channel(name).send("change"), only when someone listens. A page listens with listen('/users/events', invalidate) and a +server.rs of wisp::channel("users").events(). A page that names a live table's static rows refreshes itself through /_wisp/live/<name>.
• Several servers on one store: a Store may answer changes(table, since), and with WISP_STORE_POLL=5 (seconds) every table follows it.
• Tables are whole in memory: the store's size is the RAM bound.

Rules

#[validate(url, one_of = "a b c", pattern = "[a-z]{3}-\\d+", with = ok)] beside len min max min_len max_len email.

• url is an absolute http(s) address.
• pattern matches the whole value: literals, ., [a-z_], [^x], \d \w \s, groups, |, ? * + {n} {n,m} (values up to 1000 characters; too much backtracking does not match).
• with = f calls f(&v) -> Option<String>, the problem.

Jobs

// init
wisp::work("mail", |m: Mail| async move { send(&m).await });
wisp::cron("0 3 * * *", || async { purge().await }); // UTC
// anywhere
wisp::queue("mail").push(&Mail { to, body }); // .later(secs, &job)

• Jobs live in the saved table queue-mail. A failure (Err or a panic) is tried again after 4, 8, 16... seconds (an hour at most), five tries; then it stays dead with its last error (queue.dead(), queue.retry(id)).
• At least once: a job running when the process died runs again a minute later. One at a time per queue, in order.
• cron takes five fields (*, n, a-b, */n, lists; Sunday is 0 or 7).
• On Cloudflare, Vercel and Netlify (wisp build --target ...) the same code runs from the host's cron: wisp build writes each cron schedule (a string literal) into its trigger config, and a trigger also runs the queues' due jobs (each minute if the app has work). Set CRON_SECRET and WISP_STORE there. See /docs/deploy.

Cache

wisp::cache("top", 60, || async { top_posts().await }).await keeps an answer for 60 s per process. wisp::uncache("/posts") forgets cache keys starting with it and the pages const CACHE keeps, on every worker, for /posts and below (/ is all).

Edit This Page