Enable ability to create named in memory stores and in memory stores with shared caches.
Include ability to create encrypted connections.
Update `Store` to take an `Arc<Conn>` so references can be shared.
Update FFI to use `Stores` instead of `Store`.
Add `store_open_named_in_memory_store` to open a named in-memory store over FFI (useful for tests).
* Add iOS SDK build and test to rust 1.25.0 version of travis CI build
* Address review comments
* Move iOS testing and document generation into post test jobs
Sync needs to operate over a "mentat transaction", not just a "db transaction".
This shuffle allows internal mentat crates to consume InProgress, which models
the concept of a "mentat transaction".
This was a work-around for Tolstoy, which couldn't gracefully handle
syncing a store with a bootstrap transaction. Tolstoy now handles
that single transaction, so this is no longer necessary.
This also patches our CI test script to only run "--feature sqlcipher"
tests on sub-crates which expose this feature (i.e. themselves rely on rusqlite).
For some reason, the converted doc test fails on Rust 1.25.0, while
working with other Rust versions. For simplicity, just convert it into
a regular test.
This is ready for Android Rust-y components: it no longer references Mentat.
The standalone toolchains are installed into
$ANDROID_NDK_TOOLCHAIN_DIR/arch-$ANDROID_NDK_API_VERSION.
This leverages JNA to test the Android SDK on the host machine using
Robolectric, which is significantly faster and easier to debug than
the equivalent on-device instrumentation tests.
We'll still want instrumentation smoke tests, but they won't need to
cover the entire range of the Android SDK.
Locally, I witnessed very slow tests. Profiling with Visual VM
revealed a lot of time spent in `wait`.
Digging in, we were trying to be clever, with a `wait(1000)/notify`
mechanism. However, there were never multiple threads in play, so the
waiter wasn't waiting when `notify` was invoked. That means we always
timed out. I think this never worked and using bare `wait()` would
have revealed that.
Anyway, `CountDownLatch` maintains the one bit of state (was I
notified) and generalizes smoothly to when we have threads.
