* Add support for sending user-defined encrypted to-device messages
This is a port of the same change from the robertlong/group-call branch.
* Fix tests
* Expose the method in MatrixClient
* Fix a code smell
* Fix types
* Test the MatrixClient method
* Fix some types in Crypto test suite
* Test the Crypto method
* Fix tests
* Upgrade matrix-mock-request
* Move useRealTimers to afterEach
* Retry to-device messages
This adds a queueToDevice API alongside sendToDevice which is a
much higher-level API that adds the messages to a queue, stored in
persistent storage, and retries them periodically. Also converts
sending of megolm keys to use the new API.
Other uses of sendToDevice are nopt converted in this PR, but could
be later.
Requires https://github.com/matrix-org/matrix-mock-request/pull/17
* Bump matrix-mock-request
* Add more waits to make indexeddb tests pass
* Switch some test expectations to queueToDevice
* Stop straight away if the client has been stopped
Hopefully will fix tests being flakey and logging after tests have
finished.
* Add return types & fix constant usage
* Fix return type
Co-authored-by: Germain <germains@element.io>
* Fix return type
Co-authored-by: Germain <germains@element.io>
* Fix return type
Co-authored-by: Germain <germains@element.io>
* Stop the client in all test cases
Co-authored-by: Germain <germains@element.io>
* Update jest monorepo
* -w
* Fix guest rooms test to use async/await instead of a done callback
The done callback was never being called because it relies on a `process.nextTick()` deep within the mock. For this test we don't get a "next tick" because the event loop is busy, so we instead cargocult some test infrastructure from surrounding tests and verify the expected API call was cleared from the queue.
* Enable github-actions reporter
* Don't override local reporters
* Stop DeviceLists at end of tests
* stop more clients
* Fix tests and DRY typing
* Fix client/crypto stopping in tests
* Fix Buffer c'tor deprecated warnings
* Fix devicelist-integ test being excluded due to poor naming
Co-authored-by: Renovate Bot <bot@renovateapp.com>
Co-authored-by: Travis Ralston <travisr@matrix.org>
Co-authored-by: Michael Telatynski <7t3chguy@gmail.com>
This reworks tracking the Olm sessions a particular task is updating to avoid
deadlocks. By ensuring we synchronously mark all sessions a task cares about as
in progress from the start, we know that no other tasks will own updating a
session in common, which avoids deadlocks across multiple tasks that might be
working on a shared set of devices.
Fixes https://github.com/vector-im/element-web/issues/16194
- allow applications to pre-send decryption keys before the message is sent
- establish new olm sessions with a shorter timeout first, and then re-try in
the background with a longer timeout without blocking message sending
- only exporting account and P2P sessions
- test is halfway done:
- it only prints the export result instead of running assertions on it
- there are no sessions to export
Note: to run only the added test:
node_modules/.bin/jest spec/unit/crypto/algorithms/olm.spec.js --testEnvironment node --testNamePattern OlmDevice
The earlier commit, d3ce0cb82f, has most of the juicy details on this. In addition to d3ce's changes, we also:
* Use `TestClient` in many integration tests due to subtle behaviour changes in imports when switching to ES6. Namely the behaviour where setting the request function is less reliable in the way we did it, but `TestClient` is very reliable.
* We now use the Olm loader more often to avoid having to maintain so much duplicate code. This makes the imports slightly easier to read.
The code to migrate from the `sessionStore` to `cryptoStore` originally appeared
in https://github.com/matrix-org/matrix-js-sdk/pull/584 (2017-12-06). At this
point, it seems safe to assume most sessions that need migrating have already
done so. Removing this code simplifies store handling and removes the
`sessionStore` from most places in JS SDK.
Standardize on importing `olm-loader` rather than pasting the same boilerplate
in different tests. Importantly, `spec/unit/crypto.spec.js` did not include any
loading approach, so it would only find Olm if some other test loaded it first.
Signed-off-by: J. Ryan Stinnett <jryans@gmail.com>
