Update to Holochain 0.7 (off develop)#140
Conversation
Bump the runtime, FFI, and Tauri plugins to Holochain 0.7.0-dev.28 and
point the JS client at local checkouts for development.
Rust deps:
- holochain/conductor_api/types -> 0.7.0-dev.{28,27,27}, hdk 0.7.0-dev.19,
hdi 0.8.0-dev.13, holochain_serialized_bytes 0.0.57,
kitsune2_api 0.5.0-dev.3, lair_keystore_api 0.7.1
- crates/runtime: holochain_keystore 0.6.1 -> 0.7.0-dev.18 (the stray 0.6
pin was dragging in a duplicate lair_keystore_api)
- rust-toolchain 1.88 -> 1.95 (wasmer 7.1 / wasmtime 42 need >= 1.91)
- pin the iroh ecosystem to 1.0.0-rc.0 in Cargo.lock; rc.1's iroh-relay
API breaks kitsune2_bootstrap_srv 0.5.0-dev.3
Runtime API changes (crates/runtime/src/runtime.rs):
- Conductor::shutdown() now consumes Arc<Self> -> clone before calling
- AdminResponse::NetworkStatsDumped carries holochain_types
HolochainTransportStats, not kitsune2_api ApiTransportStats
JS (local dev):
- all @holochain/client refs -> file: ../holochain-client-js (v0.21.x,
the Holochain 0.7 client)
- example-client-app tests -> @holochain-open-dev/tryorama 0.20.0-dev.0
(import sites updated to the new package name)
Fixtures / tests:
- repack forum.happ with hc 0.7 zomes; build:zomes now sets
getrandom_backend="custom" since the hdk 0.7 tree pulls getrandom 0.3,
which otherwise fails to build on wasm32-unknown-unknown
- integration test wait_for_ready gates on try_runtime() instead of
holochain(): the plugin state is managed at setup before the async boot
sets the runtime, so holochain() resolves immediately
Both tauri-plugin-holochain integration tests pass against 0.7.
|
❌ Found 0 compliant commit and 1 non-compliant commits in 4eca328. Commit 4eca328 by @zippy is not conform to the conventional commit specification :
Bump the runtime, FFI, and Tauri plugins to Holochain 0.7.0-dev.28 and Rust deps:
Runtime API changes (crates/runtime/src/runtime.rs):
JS (local dev):
Fixtures / tests:
Both tauri-plugin-holochain integration tests pass against 0.7.`
|
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Bumps the runtime, FFI, and Tauri plugins from Holochain 0.6.1 to 0.7.0-dev.28
(the current
develop), and points the JS client at local checkouts fordevelopment. Both
tauri-plugin-holochainintegration tests pass against 0.7.Based on
feat/tauri-direct-auth-and-restart(one commit on top); retarget tomainonce that branch lands.Rust dependencies
holochain/holochain_conductor_api/holochain_types→0.7.0-dev.28/0.7.0-dev.27/0.7.0-dev.27hdk 0.7.0-dev.19,hdi 0.8.0-dev.13,holochain_serialized_bytes 0.0.57,kitsune2_api 0.5.0-dev.3,lair_keystore_api 0.7.1crates/runtime:holochain_keystore 0.6.1 → 0.7.0-dev.18— the stray 0.6 pin was dragging a duplicatelair_keystore_apiinto the graphrust-toolchain 1.88 → 1.95— Holochain 0.7's wasmer 7.1 / wasmtime 42 require rustc ≥ 1.911.0.0-rc.0inCargo.lock; a fresh resolve picksiroh-relay 1.0.0-rc.1, whose API breakskitsune2_bootstrap_srv 0.5.0-dev.3Runtime API changes (
crates/runtime/src/runtime.rs)Conductor::shutdown()now consumesArc<Self>→ clone before callingAdminResponse::NetworkStatsDumpednow carriesholochain_types::HolochainTransportStatsinstead ofkitsune2_api::ApiTransportStats(sametransport_statsfield; only the sole in-crate test consumes it)JS (local dev)
@holochain/clientrefs →file:the local../holochain-client-js(v0.21.x, the Holochain 0.7 client)example-client-apptests →@holochain-open-dev/tryorama 0.20.0-dev.0(import sites updated to the new package name)Fixtures / tests
forum.happwith hc 0.7 zomes;build:zomesnow setsgetrandom_backend="custom"since the hdk 0.7 tree pullsgetrandom 0.3, which otherwise fails to build onwasm32-unknown-unknowntauri-plugin-holochainintegrationwait_for_readygates ontry_runtime()instead ofholochain(): the plugin state ismanage()d at setup before the async boot sets the runtime, soholochain()resolves immediately (the old check raced ahead and the assertions ran before the conductor was up)Testing
Follow-ups (not in this PR)
guillemcordoba/wasmerfix-x86patch is now unused (wasmer 6.0.1 fork vs 0.7's wasmer 7.x) — needs a wasmer-7 fork branch before an x86 Android build@holochain/clientover Tauri IPC) not exercised here