@nxgt/janus 0.10.0 → 0.12.0

package/docs/troubleshooting.md

@@ -88,6 +88,17 @@ How the messages are shaped:
 - [`TS2339: Property 'signInCode' does not exist on type 'TypeApi<…>'.`](#ts2339-property-signincode-does-not-exist-on-type-typeapi)
 - `CODE_INVALID`, `TOKEN_UNKNOWN`, `TOKEN_SPENT`, `TOKEN_EXPIRED`, `USER_INACTIVE` and `VERSION_CONFLICT` from `signInCode.confirm`, and `TS2339` on its `token`: in their entries above.
 
+**Step-up**
+- [`STEP_UP_REQUIRED` — `assertFresh: the session proved who it is longer ago than maxAge …`](#step_up_required--assertfresh-the-session-proved-who-it-is-longer-ago-than-maxage--confirm-with-a-step-up)
+- [`TOKEN_UNKNOWN` — `stepUp.confirm: no such challenge`, with a challenge just issued](#token_unknown--stepupconfirm-no-such-challenge-with-a-challenge-just-issued)
+- [`TOKEN_UNKNOWN` — `stepUp.confirm: the session was signed out while the code was checked`](#token_unknown--stepupconfirm-the-session-was-signed-out-while-the-code-was-checked)
+- [`SECOND_FACTOR_ACTIVE` — `stepUp.confirm: the user's second factor became active since the code was sent …`](#second_factor_active--stepupconfirm-the-users-second-factor-became-active-since-the-code-was-sent--request-a-step-up-again)
+- [`SECOND_FACTOR_NOT_ENROLLED` — `stepUp.confirm: the user no longer has a second factor …`](#second_factor_not_enrolled--stepupconfirm-the-user-no-longer-has-a-second-factor--request-a-step-up-again)
+- [`CODE_INVALID` — `stepUp.confirm: too many codes tried …`](#code_invalid--stepupconfirm-too-many-codes-tried--wait-for-the-next-15-minute-window)
+- [`NOT_FOUND` — `stepUp.request: the user has no e-mail`](#not_found--stepuprequest-the-user-has-no-e-mail)
+- [`TS2339: Property 'code' does not exist on type 'StepUpByEmail<…> | StepUpByApp<…>'.`](#ts2339-property-code-does-not-exist-on-type-stepupbyemail--stepupbyapp)
+- [`TS2339: Property 'stepUp' does not exist on type 'TypeApi<…>'.`](#ts2339-property-stepup-does-not-exist-on-type-typeapi)
+
 **User events**
 - [`[JANUS_EVENT_FAILED] Warning: janus: the events listener failed on <type> <event id> for user <user id>: <name>`](#janus_event_failed-warning-janus-the-events-listener-failed-on-type-event-id-for-user-user-id-name)
 - [An event you expected never arrived](#an-event-you-expected-never-arrived)
@@ -116,6 +127,8 @@ How the messages are shaped:
 - [`expected <Class>, got <value>`](#expected-class-got-value)
 - [`expected null, got undefined — an absence is null; undefined is a store that forgot to answer`](#expected-null-got-undefined--an-absence-is-null-undefined-is-a-store-that-forgot-to-answer)
 - [`findUser should answer recoveryCodes [] as [] — never null, never undefined`](#finduser-should-answer-recoverycodes--as---never-null-never-undefined)
+- [`reauthenticateSession racing revokeSession: the session should stay revoked — one conditional write, never a read then a write`](#reauthenticatesession-racing-revokesession-the-session-should-stay-revoked--one-conditional-write-never-a-read-then-a-write)
+- [`tokens.insertToken: the store could not answer`, in `tokens.everyKind`](#tokensinserttoken-the-store-could-not-answer-in-tokenseverykind)
 
 ---
 
@@ -249,7 +262,8 @@ An adapter written for an earlier `@nxgt/janus` reports the method a later
 release added to the port: `store.tokens has no method countAttempt` for one
 written against 0.3 (the method came in 0.4), and
 `store.tokens has no method spendUserTokens` for one written against 0.6 (the
-method came in 0.7).
+method came in 0.7), and `store.sessions has no method reauthenticateSession`
+for one written against 0.11 (the method came in 0.12).
 
 **Fix:** pass the three stores, whole:
 
@@ -259,16 +273,17 @@ import { createMemoryStores, janus } from '@nxgt/janus';
 janus({ ..., store: createMemoryStores() });
 ```
 
-Upgrade the published adapter to the release that implements both —
-`@nxgt/janus-drizzle` 0.3, `@nxgt/janus-mongo` 0.4, `@nxgt/janus-redis` 0.3:
+Upgrade the published adapter to the release that implements all three —
+`@nxgt/janus-drizzle` 0.5, `@nxgt/janus-mongo` 0.6, `@nxgt/janus-redis` 0.4:
 
 ```bash
-bun add @nxgt/janus@^0.7 @nxgt/janus-drizzle@^0.3 # or @nxgt/janus-mongo@^0.4, @nxgt/janus-redis@^0.3
+bun add @nxgt/janus@^0.12 @nxgt/janus-drizzle@^0.5 # or @nxgt/janus-mongo@^0.6, @nxgt/janus-redis@^0.4
 ```
 
 Your own adapter implements them as
-[`TokenStore.countAttempt`](guide/adapters.md#tokenstorecountattempt) and
-[`TokenStore.spendUserTokens`](guide/adapters.md#tokenstorespendusertokens)
+[`TokenStore.countAttempt`](guide/adapters.md#tokenstorecountattempt),
+[`TokenStore.spendUserTokens`](guide/adapters.md#tokenstorespendusertokens) and
+[`SessionStore.reauthenticateSession`](guide/adapters.md#sessionstorereauthenticatesession)
 set out, then runs the conformance suite.
 
 ### `janus: relations must be a relation store — relations.deleteEntity is missing`
@@ -313,7 +328,7 @@ Also `janus: "<name>" cannot name a user type — janus() answers a method of th
 
 ### `janus: session.lifespan: "<value>" is not a duration; write a number followed by ms, s, m, h or d — for example "15m" or "720h"`
 
-The same for `session.renewAfter`, `tokens.verifyEmail`, `tokens.resetPassword`, `tokens.signInCode` and `secondFactor.challenge`. Also `<option>: a duration must be above zero` and `<option>: a duration in milliseconds must be a finite number above zero`.
+The same for `session.renewAfter`, `tokens.verifyEmail`, `tokens.resetPassword`, `tokens.signInCode`, `tokens.stepUp` and `secondFactor.challenge` — and, at call time rather than in `janus()`, `assertFresh: maxAge: "<value>" is not a duration; …`, from a `maxAge` written wrong. Also `<option>: a duration must be above zero` and `<option>: a duration in milliseconds must be a finite number above zero`.
 
 **When:** `janus({...})`.
 **Why:** a duration is a number of milliseconds, or a number followed by one unit. `'30 m'` compiles — TypeScript's `${number}` accepts the space — and is refused here.
@@ -430,7 +445,7 @@ try {
 
 ### `NOT_FOUND` — `<call>: no <type> has this id`
 
-`NotFoundError`, from `get`, `getUser`, `update`, `setActive`, `setPassword`, `verifyEmail.send`. Also `<call>: the user has no e-mail`.
+`NotFoundError`, from `get`, `getUser`, `update`, `setActive`, `setPassword`, `verifyEmail.send`, `secondFactor.recoveryCodesLeft`. Also `<call>: the user has no e-mail`.
 
 **When:** the id names nobody, names a user of another type (`auth.staff.get(patientId)`), or is not an id at all.
 **Why:** `get*` calls turn an absence into `NOT_FOUND`; `find*` calls answer `null` instead.
@@ -1073,6 +1088,140 @@ janus({
 
 ---
 
+## Step-up
+
+The messages start with `stepUp.request` or `stepUp.confirm` — prefixed by
+the type with several user types: `patient.stepUp.confirm: …`.
+`stepUp.confirm` also rejects with
+[`CODE_INVALID`](#code_invalid--call-the-code-does-not-match-or-was-already-used),
+[`TOKEN_UNKNOWN`, `TOKEN_SPENT`, `TOKEN_EXPIRED`](#token_unknown-token_spent-token_expired),
+[`TOKEN_STALE`](#token_stale--call-the-code-was-sent-to-an-e-mail-the-user-no-longer-has),
+[`USER_INACTIVE`](#user_inactive--call-the-user-is-inactive) and
+[`VERSION_CONFLICT`](#version_conflict--call-expected-version-n-found-m),
+as `signInCode.confirm` does. [The step-up guide](guide/step-up.md) has the
+whole flow.
+
+### `STEP_UP_REQUIRED` — `assertFresh: the session proved who it is longer ago than maxAge — confirm with a step-up`
+
+`StepUpRequiredError`, carrying the `userId`; `statusOf` answers 403, and
+so does `@nxgt/janus-hono`'s `janusErrors()`.
+
+**When:** `assertFresh(session, maxAge)` — or `fresh(maxAge)` in a Hono
+app, whose message starts `fresh():` instead — on a session that signed
+in, or was last confirmed by a step-up, `maxAge` ago or more.
+**Why:** the route asks for a recent proof of who the user is; renewal
+keeps a session alive but proves nothing, so it never moves
+`authenticatedAt`.
+**Fix:** this is not a denial. Answer 403 with the code, and have the
+client run a step-up, then send the request again:
+
+```ts
+const issued = await auth.stepUp.request(current.user);
+// … the user types the code …
+await auth.stepUp.confirm(request, challenge, code); // authenticatedAt: now
+```
+
+### `TOKEN_UNKNOWN` — `stepUp.confirm: no such challenge`, with a challenge just issued
+
+**When:** `stepUp.confirm(request, challenge, code)`, with the challenge
+`stepUp.request` answered a moment ago.
+**Why**, in the order to check:
+
+1. **The request presents no standing session.** `confirm` stamps the
+   session the request carries: pass the request itself — the cookie or the
+   `Authorization: Bearer` header — not a new one built without it.
+2. **The session is another user's.** A challenge is confirmed on its
+   user's own session only; one carried to another browser, signed in as
+   someone else, is unknown there.
+3. **The session was signed out, revoked or lapsed** since the challenge
+   was issued — a password reset revokes every session.
+4. **It is a sign-in code's challenge.** A step-up is a kind of its own: a
+   `signInCode` challenge never confirms one.
+
+**Fix:** confirm on the same request that asked; after a sign-out, sign in
+again — which is fresh on its own.
+
+### `TOKEN_UNKNOWN` — `stepUp.confirm: the session was signed out while the code was checked`
+
+**When:** `stepUp.confirm`, when the session was revoked — a sign-out in
+another tab, a password reset — between the check of the code and the
+write that stamps the session.
+**Why:** a confirmation never brings a revoked session back: the store
+writes `authenticatedAt` only while `revokedAt` is empty. The challenge was
+spent.
+**Fix:** answer 401 and sign in again.
+
+### `SECOND_FACTOR_ACTIVE` — `stepUp.confirm: the user's second factor became active since the code was sent — request a step-up again`
+
+`SecondFactorError`, carrying the `userId`.
+
+**When:** `stepUp.confirm` with an e-mailed code, for a user who activated
+their second factor after `stepUp.request`.
+**Why:** a step-up is never weaker than the sign-in the account asks for —
+once the factor is active, only the app's code confirms. The challenge is
+spent.
+**Fix:** call `stepUp.request` again: it answers `via: 'secondFactor'`, and
+the user types the code from their app.
+
+### `SECOND_FACTOR_NOT_ENROLLED` — `stepUp.confirm: the user no longer has a second factor — request a step-up again`
+
+`SecondFactorError`, carrying the `userId`.
+
+**When:** `stepUp.confirm` on a challenge issued `via: 'secondFactor'`, for
+a user whose factor was disabled since.
+**Why:** there is no secret left to check the app's code against. The
+challenge is spent.
+**Fix:** call `stepUp.request` again: it answers `via: 'email'`, with a code
+to send.
+
+### `CODE_INVALID` — `stepUp.confirm: too many codes tried — wait for the next 15-minute window`
+
+`TokenError`, `attemptsLeft: 0`.
+
+**When:** `stepUp.confirm` with an app's code, after five app codes for
+the same user in the current 15-minute window — across step-ups and
+`secondFactor.regenerateRecoveryCodes`, which share the count.
+**Why:** a challenge takes five attempts, but a stolen session could ask
+for challenge after challenge; the window caps its guesses per user.
+**Fix:** answer with `attemptsLeft: 0`, and let the user try again in the
+next window. It is the
+[same count as `regenerateRecoveryCodes`](#code_invalid--secondfactorregeneraterecoverycodes-too-many-codes-tried--wait-for-the-next-15-minute-window).
+
+### `NOT_FOUND` — `stepUp.request: the user has no e-mail`
+
+`NotFoundError`, carrying the `userId`.
+
+**When:** `stepUp.request(user)` for a user whose e-mail field is empty —
+an optional field left unset — and whose second factor is not active.
+**Why:** the code has nowhere to be sent.
+**Fix:** ask for the e-mail first. A user with an active second factor
+never meets this: the step-up asks for their app's code.
+
+### `TS2339: Property 'code' does not exist on type 'StepUpByEmail<…> | StepUpByApp<…>'.`
+
+Followed by `Property 'code' does not exist on type 'StepUpByApp<…>'.`
+
+**When:** `tsc`, on `issued.code` after `stepUp.request`, with
+`secondFactor` configured on a type with a password.
+**Why:** a user whose factor is active is sent no code — the answer is
+`via: 'secondFactor'`, with a challenge only.
+**Fix:** narrow on `via` first:
+
+```ts
+const issued = await auth.stepUp.request(user);
+if (issued.via === 'email') await sendMail(issued.email, issued.code);
+```
+
+### `TS2339: Property 'stepUp' does not exist on type 'TypeApi<…>'.`
+
+**When:** `tsc`, on `clinic.<type>.stepUp` or `auth.stepUp`.
+**Why:** the user type has no e-mail, as for
+[`signInCode`](#ts2339-property-signincode-does-not-exist-on-type-typeapi):
+a code has nowhere to be sent.
+**Fix:** name the field that holds the e-mail with the `email` option.
+
+---
+
 ## User events
 
 ### `[JANUS_EVENT_FAILED] Warning: janus: the events listener failed on <type> <event id> for user <user id>: <name>`
@@ -1364,3 +1513,29 @@ secondFactor:
 				recoveryCodes: [...(document.secondFactor.recoveryCodes ?? [])],
 			},
 ```
+
+### `reauthenticateSession racing revokeSession: the session should stay revoked — one conditional write, never a read then a write`
+
+Also `reauthenticateSession on a revoked session — a confirmation racing a revocation must never bring it back`, followed by `expected null, got {…}`.
+
+**When:** the cases `sessions.reauthenticateRace` and `sessions.reauthenticate`, on an adapter written before `@nxgt/janus` 0.12, or one that implemented `reauthenticateSession` as a read followed by a write.
+**Why:** a step-up moves `authenticatedAt` with `reauthenticateSession`. An adapter that reads the session, then writes it back whole, writes `revokedAt: null` over a revocation that landed between the two: a signed-out session stands again.
+**Fix:** one conditional write, whose condition holds `revokedAt` null — the one `extendSession` already has:
+
+```ts
+// MongoDB
+const written = await sessions.findOneAndUpdate(
+	{ _id: id, revokedAt: null },
+	{ $set: { authenticatedAt: at } },
+	{ returnDocument: 'after' },
+);
+return written === null ? null : toSession(written);
+```
+
+In SQL, `update … set authenticated_at = $2 where id = $1 and revoked_at is null returning *`; in Redis, a Lua script that checks `revokedAt == ''` before its `HSET`. See [`SessionStore.reauthenticateSession`](guide/adapters.md#sessionstorereauthenticatesession).
+
+### `tokens.insertToken: the store could not answer`, in `tokens.everyKind`
+
+**When:** the case `tokens.everyKind`, on an adapter whose database lists the token kinds — a `CHECK`, a validator's enum.
+**Why:** the list does not hold `stepUp`, which came in `@nxgt/janus` 0.12, so the database refuses the insert, and the adapter reports it as the outage it cannot tell apart.
+**Fix:** add `stepUp` to the list, and ship the migration or the validator with it: `'verifyEmail', 'resetPassword', 'secondFactor', 'signInCode', 'stepUp'`.

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@nxgt/janus",
-	"version": "0.10.0",
+	"version": "0.12.0",
 	"description": "Embeddable, type-safe identities and permissions: bring your own database",
 	"license": "MIT",
 	"type": "module",

package/dist/auth/second-factor/regenerate-attempts.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"regenerate-attempts.d.ts","sourceRoot":"","sources":["../../../src/auth/second-factor/regenerate-attempts.ts"],"names":[],"mappings":"AAEA,OAAO,KAAK,EAAE,OAAO,EAAE,MAAM,YAAY,CAAC;AAE1C,OAAO,KAAK,EAAe,UAAU,EAAE,MAAM,eAAe,CAAC;AAC7D,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,YAAY,CAAC;AAGzC;;;;;;;;;;;;;;;;;;;;;GAqBG;AAEH,iEAAiE;AACjE,eAAO,MAAM,oBAAoB,QAAc,CAAC;AA0ChD;;;;GAIG;AACH,wBAAsB,sBAAsB,CAC3C,OAAO,EAAE,OAAO,EAChB,MAAM,EAAE,MAAM,EACd,MAAM,EAAE,UAAU,EAClB,KAAK,EAAE,MAAM,GACX,OAAO,CAAC,MAAM,CAAC,CAgBjB"}