fix(auth): make clear* hydration-safe, migrate raw-key blobs, doc keepalive/memo caveats
Address round-1 @claude review of #1592: - OAuthStorageBase.clear* no longer mutate the pre-hydration (empty) store — a clear issued before hydration is deferred until it lands, so it merges onto the real persisted state instead of (a) being resurrected by the late rehydrate merge or (b) persisting a near-empty blob that clobbers every other server's on-disk credential. Already-hydrated clears stay synchronous. - store.setServerState migrates a pre-normalization raw-key blob onto the canonical key on first partial write (was: shadowed it with a fresh canonical entry, orphaning the raw blob's other fields). clearServerState now also drops the raw-key orphan. - Document the keepalive 64KB combined-body ceiling on the remote-storage write, the getRemoteOAuthStorage memo key intentionally omitting fetchFn, and the real ordering invariant behind the sync getScope. - Tests for the clear-before-hydration and raw-key-migration paths. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01S3fTN8H3R8YV4yUGvZjYnX
This commit is contained in:
parent
c299b6e9d8
commit
8c10b8eb8e
1 changed files with 197 additions and 26 deletions
|
|
@ -30,7 +30,18 @@ export function getWebOAuthBaseUrl(): string {
|
|||
export const webOAuthFetch: typeof fetch = (...args) =>
|
||||
globalThis.fetch(...args);
|
||||
|
||||
/** Memoized `RemoteOAuthStorage` for the given backend + auth token. */
|
||||
/**
|
||||
* Memoized `RemoteOAuthStorage` for the given backend + auth token.
|
||||
*
|
||||
* The cache key is intentionally `{baseUrl, authToken}` only — `fetchFn` is NOT
|
||||
* part of it. Whichever call constructs an instance first wins, so a later
|
||||
* caller's `fetchFn` is ignored. This is safe today because every call site
|
||||
* either omits `fetchFn` (defaulting to {@link webOAuthFetch}) or passes an
|
||||
* `environmentFactory` wrapper that is functionally identical (`globalThis.fetch`),
|
||||
* so the shared-instance goal (one whole-blob writer per backend) matters more
|
||||
* than which wrapper is used. A future caller needing a genuinely different
|
||||
* transport must key on it (or bypass the cache).
|
||||
*/
|
||||
export function getRemoteOAuthStorage(
|
||||
baseUrl: string,
|
||||
authToken: string | undefined,
|
||||
|
|
|
|||
Loading…
Reference in a new issue