* Use a `StoreHandle` to init OlmMachine
This will be faster if we need to prepare the store.
* Include "needsBackup" flag in inbound group session batches
* On startup, import data from libolm cryptostore
* ISessionExtended -> SessionExtended
* Move existing crypto integ tests into a subdirectory
* Factor out some common bits from `crypto.spec.ts`
* Integration test for device verification
* Ignore generated file in prettier
Fixes https://github.com/vector-im/element-web/issues/14947
Much like element-web's Jitsi wrapper build steps, this downloads the emoji JSON at build time to ensure it gets reasonably updated. In the future, the spec might want to consider publishing a dedicated i18n package on npm for this, however this is fine for now. We download rather than copy/paste to ensure we always have an updated copy.
Most `npm` operations are replaced with `yarn`, which generally has better
behavior. However, steps like publish that write to the NPM registry are left to
`npm`, which currently handles these tasks best.
jsdoc can't read our raw source, because of our dangling commas in function
calls. On the other hand, running on /lib means that a lot of the useful
information about exports is lost and you end up having to jump through hoops
to get jsdoc to generate the right thing.
This uses a separate run of babel (with all the presets turned off) to generate
source which is almost identical to the input, but lacks trailing commas.
(https://babeljs.io/blog/2015/10/31/setting-up-babel-6 says 'Babel 6 ships
without any default transforms, so when you run Babel on a file it will just
print it back out to you without changing anything.' - however, that is,
empirically, not entirely true.)
Richard van der Hoff