- now, most of the work is happening on the main thread
- for database in browser, we still have dedicated WebWorker - but it is used only for OPFS access and only for that
- for syn in browser we still offload sync operations to the WebWorker
Intel MAC builds were removed in https://github.com/tursodatabase/turso/
commit/3547bd10931e030a372bedb4968404301c2936c6 but arm builds were
broken after that.
This PR returns back proper ARM builds for Apple
Reviewed-by: Jussi Saurio <jussi.saurio@gmail.com>
Closes#3054
This tries to establish smooth as possible experience for browser
packages in web.
In order to do that this PR do the following tricks:
1. export `./vite` entry-point which should be used like `import {
connect } from "@tursodatabase/database-browser/vite"` for Vite bundler
* This entrypoint has fix for the issue
https://github.com/vitejs/vite/issues/8427 which breaks package in the
dev-server mode
* In order to overcome this we do 2 tricks:
- Inline WASM module in order to avoid loading it from the separate
file
- Use same file as entry-point for main thread and for the web
worker
* Note, that we do these tricks only for `development` build and
produce build will be chunked and optimized as usual with Vite and will
treat worker and WASM modules as separate fiels
3. export `./turbopack` entry-point which should be used like `import {
connect } from @tursodatabase/database-browser/turbopack"` for Turbopack
(Next.js) bundler
* This entrypoint has fix for the issue
https://github.com/vercel/next.js/issues/82520'
* In order to overcome this for now we always inline WASM for Next.js
4. Bundle browser libraries in order to easily consume them without need
for bundlers:
* e.g. `import { connect } from
"https://unpkg.com/@tursodatabase/database-browser/bundle/main.es.js";`
5. We vendor `@napi-rs/wasm-runtime` for now in order to fix runtime
errors due to accesses to `process.env.NODE_DEBUG_NATIVE` env var (not
defined in web)
* This should be very temporary solution because I already fixed
wasm-util dependency which cause this error
(https://github.com/toyobayashi/wasm-util/issues/4) and hope that napi-
rs PR will be merged soon: https://github.com/napi-rs/napi-rs/pull/2921
Reviewed-by: Jussi Saurio <jussi.saurio@gmail.com>
Closes#3017
Particularly we were tracing `ImmutableRecord` / `BTreeKey` which would
then trace the bytes of records. These are super super hot paths and I
think we can probably remove even more to under debug assertions so we
dont eat those atomics/branches all the time.
This PR also introduces the `tracing_release` feature, which turns all
`trace!` and `debug!` macro invocations to noops at compile time, and
makes that feature available for all bindings.
it also removes the unused `lru` dependency, and cleans up the makefile
a bit
Reviewed-by: Jussi Saurio <jussi.saurio@gmail.com>
Closes#2995
The `run_once()` name is just a historical accident. Furthermore, it now
started to appear elsewhere as well, so let's just call it IO::step() as we
should have from the beginning.
