v1.1.0

AimDB v1.1.0 Release Notes

Stage profiling, no_std transforms, context-aware connectors and a safer MCP server!

Typed dataplane for distributed systems

---

🎯 Highlights

This release delivers three major capabilities across the full platform stack: automatic per-stage performance profiling (with LLM-powered bottleneck detection via MCP), no_std support for the complete Transform API (.transform() and .transform_join() now run on Embassy and WASM), and context-aware serializers/deserializers that give connector callbacks access to the runtime clock and logger. The MCP server gains a public/read-only mode with SSRF protection for internet-facing deployments, and a first community contribution ships in the form of a hello-mailbox example.

---

✨ Major Features

📊 Automatic Stage Profiling (Issue #58, RFC 014)

Zero-instrumentation wall-clock timing for every .source(), .tap(), and .link() stage!

Enable with --features profiling — zero overhead when the feature is off.

// Name stages for readable profiling output
builder.configure::<Temperature>(AppKey::Temp, |reg| {
    reg.buffer(BufferCfg::SingleLatest)
       .source(produce_temperature)
       .with_name("temperature_sensor")      // ← names the source stage
       .tap(log_temperature)
       .with_name("temperature_logger")      // ← names the tap stage
});

# Query via MCP
mcp_aimdb-local_get_stage_profiling(record_key: "Temp")
# → bottleneck: "slow_status_processor" (avg 103 ms)
# → recommendation: "Consider offloading or batching"

Details:

• Measures interval between successive produce() calls (source) and successive recv() yields (tap/link) — no per-value handler changes required
• Per-stage counters: call_count, avg_time_ns, min_time_ns, max_time_ns
• RecordRegistrar::with_name("...") always callable — no-op when profiling is disabled
• Works on no_std + alloc via portable-atomic with fallback + critical-section for 64-bit atomics on thumbv7em-none-eabihf
• Two new MCP tools: get_stage_profiling (bottleneck detection + human-readable recommendation) and reset_stage_profiling
• remote-access-demo updated: SystemStatus.slow_status_processor sleeps 100 ms deliberately so the bottleneck is immediately visible when you run the demo

🔄 no_std Transform API (Design 027)

.transform() and .transform_join() now work on every runtime — Tokio, Embassy, and WASM!

Multi-input join fan-in is no longer hardcoded to tokio::sync::mpsc. The new JoinFanInRuntime trait in aimdb-executor is implemented per adapter:

Adapter	Backend	Capacity
Tokio	tokio::sync::mpsc::channel	64
Embassy	embassy_sync::Channel<CriticalSectionRawMutex, T, 8>	8
WASM	futures_channel::mpsc::channel	64

New task-model join handler — replaces the old callback form with a long-running task that owns the event loop:

// New (v1.1.0) — task model, no per-event heap allocation
builder.configure::<DewPoint>(AppKey::DewPoint, |reg| {
    reg.buffer(BufferCfg::SingleLatest)
       .transform_join(|join| {
           join.input::<Temperature>(AppKey::Temp)
               .input::<Humidity>(AppKey::Humidity)
               .on_triggers(dew_point_task)
       })
});

async fn dew_point_task(
    mut rx: JoinEventRx,
    producer: Producer<DewPoint, TokioAdapter>,
) {
    let mut temp = None;
    let mut humidity = None;
    while let Ok(trigger) = rx.recv().await {
        match trigger.index {
            0 => temp = Some(trigger.value::<Temperature>()),
            1 => humidity = Some(trigger.value::<Humidity>()),
            _ => {}
        }
        if let (Some(t), Some(h)) = (&temp, &humidity) {
            producer.produce(calculate_dew_point(t, h)).await;
        }
    }
}

Breaking change vs. the old with_state().on_trigger(...) callback form — see Migration Guide below.

The weather-mesh demo stations now derive a DewPoint record from Temperature and Humidity via transform_join on both Tokio and Embassy.

🔌 Context-Aware Connectors (Design 026)

Deserializers and serializers can now access the runtime clock and logger!

// New: context-aware deserializer
connector.link_from("mqtt://broker/sensor/temperature")
    .with_deserializer(|ctx, bytes| {
        let ts = ctx.now();   // platform-independent timestamp
        ctx.log_debug(&format!("Received {} bytes at {:?}", bytes.len(), ts));
        serde_json::from_slice::<Temperature>(bytes).map_err(|e| e.to_string())
    });

// Old (still available for simple cases)
connector.link_from("mqtt://broker/sensor/temperature")
    .with_deserializer_raw(|bytes| {
        serde_json::from_slice::<Temperature>(bytes).map_err(|e| e.to_string())
    });

Symmetric .with_serializer(|ctx, value| ...) and .with_serializer_raw(|value| ...) APIs for outbound connectors. All existing connectors (MQTT, KNX, WebSocket) updated to propagate context.

✅ Writer-Exclusivity Validation (Issue #89)

.source(), .transform(), and .link_from() are now mutually exclusive — combining any two panics at configuration time with a clear error message instead of silently racing on the buffer:

thread 'main' panicked at 'Record "sensor::Temperature" already has a
source/transform and cannot also have a link_from "mqtt://...". A record
can only have one writer.'

Fan-in with multiple .link_from() connectors remains allowed.

---

🔧 Other Changes

MCP Server

• Public mode (--public flag): Restricts the server to three read-only tools (discover_instances, list_records, get_record) for safe internet-facing deployments. Strips client-supplied socket_path arguments to prevent SSRF. Suppresses resources and prompts capabilities.
• Default socket (--socket <PATH>): Sets a default socket path at server startup. Resolution order: explicit argument → --socket flag → AIMDB_SOCKET env var. Reduces repetition when working with a single instance.

New Examples

• hello-mailbox — minimal example demonstrating Mailbox (latest-wins) semantics using the synchronous API. First community contribution, by @ggmaldo.
• Weather-mesh DewPoint — all weather stations now derive DewPoint from Temperature + Humidity via transform_join, end-to-end on both Tokio and Embassy.

Dependency Updates

• rand 0.8 → 0.10.1 across workspace (aimdb-data-contracts, aimdb-embassy-adapter, examples). Migrated API: gen → random, SmallRng seed size 16 → 32, RngExt imports added.

---

📦 Published Crates

Updated

Crate	Previous	New
aimdb-core	1.0.1	1.1.0
aimdb-executor	0.1.0	0.2.0
aimdb-tokio-adapter	0.5.0	0.6.0
aimdb-embassy-adapter	0.5.0	0.6.0
aimdb-wasm-adapter	0.1.1	0.2.0
aimdb-client	0.5.0	0.6.0
aimdb-codegen	0.1.0	0.2.0
aimdb-websocket-connector	0.1.0	0.2.0
aimdb-mqtt-connector	0.5.1	0.6.0
aimdb-knx-connector	0.3.1	0.4.0
aimdb-mcp	0.7.0	0.8.0
aimdb-data-contracts	0.1.0	0.1.1
aimdb-persistence	0.1.0	0.1.1
aimdb-persistence-sqlite	0.1.0	0.1.1

Unchanged

• aimdb-cli@0.6.0
• aimdb-sync@0.5.0
• aimdb-ws-protocol@0.1.0
• aimdb-derive@0.1.0

---

⚠️ Breaking Changes

Join handler API (aimdb-core, aimdb-codegen)

JoinBuilder::with_state(...).on_trigger(Fn(...) -> Pin<Box<dyn Future>>) is replaced by the task-model:

JoinBuilder::on_triggers(FnOnce(JoinEventRx, Producer<Output, R>) -> impl Future)

The handler now owns the full event loop, eliminating per-event heap allocation and allowing state to borrow across .await points. Generated code from aimdb-codegen is updated automatically.

Connector serializer/deserializer APIs (aimdb-core)

Old	New
.with_deserializer(|bytes| ...)	.with_deserializer_raw(|bytes| ...)
.with_serializer(|value| ...)	.with_serializer_raw(|value| ...)

New context-aware forms: .with_deserializer(\|ctx, bytes\| ...) and .with_serializer(\|ctx, value\| ...).

Internal types InboundConnectorLink::deserializer, Router::route(), ConnectorLink::serializer, and OutboundRoute have changed to use the DeserializerKind / SerializerKind enums. Custom connector implementations must update their router dispatch calls to pass db.runtime_any().

TimeOps::duration_as_nanos — new required method (aimdb-executor)

Implementations of the TimeOps trait must now provide duration_as_nanos(duration: Self::Duration) -> u64. Custom runtime adapters outside this workspace need to add this method; implementations should saturate rather than overflow.

---

🚀 Migration Guide

Step 1: Update join handlers

// Before (callback model)
.transform_join(|join| {
    join.input::<Temperature>(AppKey::Temp)
        .input::<Humidity>(AppKey::Humidity)
        .with_state(MyState::default())
        .on_trigger(|trigger, state, producer| Box::pin(async move {
            // ...
        }))
})

// After (task model)
.transform_join(|join| {
    join.input::<Temperature>(AppKey::Temp)
        .input::<Humidity>(AppKey::Humidity)
        .on_triggers(my_join_task)
})

async fn my_join_task(
    mut rx: JoinEventRx,
    producer: Producer<DewPoint, TokioAdapter>,
) {
    // State lives here — no heap allocation per event
    let mut state = MyState::default();
    while let Ok(trigger) = rx.recv().await {
        // handle trigger
        producer.produce(compute(&state)).await;
    }
}

Step 2: Update connector serializer/deserializer signatures

// Before
.with_deserializer(|bytes| serde_json::from_slice(bytes).map_err(|e| e.to_string()))
.w...

v1.0.0

AimDB v1.0.0 Release Notes

Stable core API, real-time browser streaming, and LLM-driven architecture design!

Async in-memory database for MCU → edge → cloud data synchronization

---

🎯 Highlights

This release promotes aimdb-core to 1.0.0 — declaring a stable public API for the core database engine. It also ships four entirely new crates: a WebSocket connector for real-time browser streaming, a WASM adapter for running AimDB in the browser, a code generation library for architecture-to-code workflows, and a shared wire protocol for the WebSocket ecosystem. The MCP server gains a full architecture agent for LLM-driven system design, and the CLI adds code generation and live monitoring commands.

Headline: MCU → Edge → Cloud → Browser — AimDB now spans the full stack.

Extensible Streamable registry — the Streamable type dispatch system has been redesigned from a closed dispatcher to an open, extensible registry pattern. Users now register their own types via .register::<T>() on connector and adapter builders. Concrete contracts (Temperature, Humidity, GpsLocation) have moved to application-level crates.

---

✨ Major Features

🌐 WebSocket Connector (New Crate)

Real-time bidirectional streaming to browsers and between AimDB instances!

use aimdb_tokio_adapter::TokioAdapter;
use aimdb_websocket_connector::WebSocketConnector;

let db = AimDbBuilder::new()
    .runtime(TokioAdapter::new())
    .with_connector(
        WebSocketConnector::new()
            .bind("0.0.0.0:8080")
            .path("/ws")
            .with_late_join(true),
    );

// Push records to browser clients
builder.configure::<Temperature>(AppKey::Temp, |reg| {
    reg.buffer(BufferCfg::SingleLatest)
       .link_to("ws://temperature");  // ← streams to all subscribers
});

Two modes:

• 🖥️ Server mode (Axum-based) — accept incoming WebSocket connections via link_to("ws://topic")
• 🔗 Client mode (tokio-tungstenite) — connect out to remote servers via link_to("ws-client://host/topic") for AimDB-to-AimDB sync without a broker
• 🔐 Authentication via pluggable AuthHandler trait
• 📡 MQTT-style wildcards — # multi-level, * single-level topic matching
• ⏱️ Late-join — new subscribers receive current values immediately
• 📋 StreamableRegistry — extensible type-erased dispatch via .register::<T>() with schema-name collision detection

🕸️ WASM Adapter (New Crate)

Run AimDB in the browser — full dataflow engine via WebAssembly!

import { useRecord, useBridge } from 'aimdb-wasm-adapter/react';

function TemperatureDashboard() {
    const bridge = useBridge("ws://localhost:8080/ws");
    const temp = useRecord<Temperature>(bridge, "sensor.temperature");

    return <div>Current: {temp?.celsius}°C</div>;
}

Features:

• 🏗️ Full aimdb-executor trait implementations (RuntimeAdapter, Spawn, TimeOps, Logger)
• 🪶 Rc<RefCell<…>> buffers — zero-overhead for single-threaded WASM
• 🎯 WasmDb facade via #[wasm_bindgen]: configureRecord, get, set, subscribe
• 🌉 WsBridge — WebSocket bridge connecting browser to remote AimDB server
• ⚛️ React hooks: useRecord<T>, useSetRecord<T>, useBridge
• 📋 SchemaRegistry for type-erased record dispatch with extensible .register::<T>() API
• no_std + alloc compatible (wasm32-unknown-unknown target)

📋 Data Contracts — Pure Trait Crate

aimdb-data-contracts refocused as a pure trait-definition crate (version reset to 0.1.0).

Trait definitions for self-describing data schemas that work identically across MCU, edge, and cloud:

• SchemaType — unique identity and versioning
• Streamable — capability marker for types crossing serialization boundaries
• Linkable — wire format for connector transport
• Simulatable — test data generation
• Observable — signal extraction for monitoring
• Migratable — schema evolution with MigrationChain and MigrationStep

Concrete contracts (Temperature, Humidity, GpsLocation) and the closed StreamableVisitor dispatcher have been removed — see Breaking Changes below.

🏗️ Code Generation (New Crate)

Architecture-to-code: generate Rust source and Mermaid diagrams from .aimdb/state.toml!

# Generate Mermaid architecture diagram
aimdb generate mermaid

# Generate Rust schema code
aimdb generate schema

# Scaffold a new common crate
aimdb generate common-crate

# Scaffold a hub binary crate
aimdb generate hub-crate

aimdb-codegen features:

• 📄 ArchitectureState type for reading .aimdb/state.toml decision records
• 📊 Mermaid diagram generation from architecture state
• 🦀 Rust source generation — value structs, key enums, SchemaType/Linkable impls, configure_schema() functions
• 🏗️ Common crate, hub crate, and binary crate scaffolding
• ✅ State validation module for architecture integrity checks

🤖 Architecture Agent (MCP)

LLM-driven system design — propose, validate, and apply architecture changes through conversation!

The MCP server now includes a full architecture agent with a session state machine:

Idle → Gathering → Proposing → Resolve

16+ new MCP tools:

Tool	Description
propose_add_record	Add a new record to the architecture
propose_add_connector	Add a connector (MQTT, WebSocket, etc.)
propose_modify_buffer	Change buffer type/capacity
propose_modify_fields	Modify record fields
propose_modify_key_variants	Change key variants
remove_record	Remove a record
rename_record	Rename a record
reset_session	Reset agent session state
resolve_proposal	Apply or reject pending proposals
save_memory	Persist architecture decisions
validate_against_instance	Validate architecture against running instance
get_architecture	Get current architecture state
get_buffer_metrics	Get buffer performance metrics

Architecture MCP resources provide Mermaid diagrams and validation results as live resources.

📡 Wire Protocol (New Crate)

Shared wire protocol types for the WebSocket ecosystem.

aimdb-ws-protocol provides ServerMessage and ClientMessage enums used by both the WebSocket connector (server side) and the WASM adapter (browser side). JSON-encoded with "type" discriminant tag.

---

🔧 Other Changes

aimdb-core

• Added ws:// and wss:// URL scheme support in ConnectorUrl for WebSocket connectors
• ConnectorUrl::default_port() now handles WebSocket schemes
• ConnectorUrl::is_secure() includes wss://

aimdb-cli

• New aimdb generate subcommand for code generation via aimdb-codegen
• New aimdb watch subcommand for live record monitoring

Dependency Updates

• aimdb-knx-connector 0.3.1 — Updated Embassy dependency versions (executor 0.10.0, time 0.5.1, sync 0.8.0, futures 0.1.2, net 0.9.0)
• aimdb-mqtt-connector 0.5.1 — Updated Embassy dependency versions (executor 0.10.0, time 0.5.1, sync 0.8.0, net 0.9.0)
• Embassy submodules updated to latest upstream commits

---

📦 Published Crates

New Crates

• 🆕 aimdb-websocket-connector@0.1.0 — real-time bidirectional WebSocket streaming
• 🆕 aimdb-ws-protocol@0.1.0 — shared wire protocol types
• 🆕 aimdb-wasm-adapter@0.1.1 — WebAssembly runtime adapter with React hooks
• 🆕 aimdb-codegen@0.1.0 — architecture-to-code generation

Updated

Crate	Previous	New
aimdb-core	0.5.0	1.0.0
aimdb-data-contracts	0.5.0	0.1.0 (reset — pure trait crate)
aimdb-cli	0.5.0	0.6.0
aimdb-mcp	0.5.0	0.6.0
aimdb-mqtt-connector	0.5.0	0.5.1
aimdb-knx-connector	0.3.0	0.3.1

Unchanged

• aimdb-tokio-adapter@0.5.0
• aimdb-embassy-adapter@0.5.0
• aimdb-client@0.5.0
• aimdb-sync@0.5.0
• aimdb-persistence@0.1.0
• aimdb-persistence-sqlite@0.1.0
• aimdb-derive@0.1.0
• aimdb-executor@0.1.0

---

⚠️ Breaking Changes

• aimdb-data-contracts: Concrete contracts removed from this crate. If you depended on Temperature, Humidity, or GpsLocation from aimdb-data-contracts, define them in your own application crate or shared common crate (see examples/weather-mesh-demo/weather-mesh-common for a reference).
• aimdb-data-contracts: for_each_streamable() and StreamableVisitor removed. Use .register::<T>() on connector/adapter builders instead.
• aimdb-data-contracts: ts feature removed (ts-rs dependency dropped).
• aimdb-data-contracts: Version reset from 1.0.0 to 0.1.0 to reflect the reduced, stabilizing scope as a pure trait crate.

---

🖥️ New CLI Commands

# Code generation
aimdb generate mermaid           # Generate Mermaid architecture diagram
aimdb generate schema            # Generate Rust schema code
aimdb generate common-crate      # Scaffold a common crate
aimdb generate hub-crate         # Scaffold a hub binary crate

# Live monitoring
aimdb watch <record>             # Watch live record updates

---

🚀 Migration Guide

Step 1: Update dependencies

[dependencies]
aimdb-core = "1.0"
aimdb-data-contracts = "0.1"  # reset — now a pure trait crate
aimdb-tokio-adapter = "0.5"   # unchanged
# Optional: add WebSocket streaming
aimdb-websocket-connector = "0.1"
# Optional: add WASM support
aimdb-wasm-adapter = "0.1"

Step 2: Move concrete contracts to your crate

If you used Temperature, Humidity, or GpsLocation from aimdb-data-contracts, copy them into your own application or shared common crate. See examples/weather-mesh-demo/weather-mesh-common for a working example.

Step 3: Replace StreamableVisitor with .register::<T>()

// Before (closed dispatcher):
// for_each_streamable(visitor);

// After (extensible registry):
let connector = WebSocketConnector::new()
    .register::<Temperature>()
    .r...

v0.5.0 - Transforms, Persistence, Graph Introspection & Dynamic Routing

AimDB v0.5.0 Release Notes

Transforms, persistence, graph introspection, and dynamic routing!

---

🎯 What's New in v0.5.0

This release introduces reactive data transformations, a pluggable persistence layer, a dependency graph introspection API, and dynamic topic/address routing for connectors. It also ships two new crates (aimdb-persistence, aimdb-persistence-sqlite).

---

✨ Major Features

🔄 Transform API (Design 020)

Reactive data transformations between records — computed values that update automatically!

use aimdb_core::{AimDbBuilder, RecordKey};

// Single-input transform: derive Fahrenheit from Celsius
builder.configure::<Celsius>(AppKey::TempCelsius, |reg| {
    reg.buffer(BufferCfg::SingleLatest);
});

builder.configure::<Fahrenheit>(AppKey::TempFahrenheit, |reg| {
    reg.transform_raw(AppKey::TempCelsius, |c: Celsius| {
        Fahrenheit { value: c.value * 9.0 / 5.0 + 32.0 }
    });
});

// Multi-input join: combine humidity + temperature into comfort index
builder.configure::<ComfortIndex>(AppKey::Comfort, |reg| {
    reg.transform_join_raw(|join| {
        join.input::<Celsius>(AppKey::TempCelsius)
            .input::<Humidity>(AppKey::Humidity)
            .on_trigger(|trigger, state| {
                // Called whenever any input changes
                Some(ComfortIndex::compute(state))
            })
    });
});

Features:

• 🔗 Single-input transform_raw() for simple derivations
• 🔗 Multi-input transform_join_raw() with JoinTrigger event dispatch
• 🔒 Mutual exclusion: a record cannot have both .source() and .transform()
• 🚀 Automatic spawning: transforms run as async tasks during AimDb::build()
• 🔍 Tracing integration: full lifecycle event logging

📊 Graph Introspection API (Design 021)

Visualize the dependency graph of your AimDB instance!

// Get full dependency graph
let nodes = db.graph_nodes();
let edges = db.graph_edges();
let order = db.graph_topo_order();

// Nodes show origin type and buffer config
for node in &nodes {
    println!("{}: {:?} ({:?})", node.key, node.origin, node.buffer_info);
}

// Edges show data flow
for edge in &edges {
    println!("{} → {} ({:?})", edge.from, edge.to, edge.edge_type);
}

RecordOrigin variants:

• Source — direct producer writes
• Link — connector-bridged external data
• Transform — single-input reactive derivation
• TransformJoin — multi-input reactive join
• Passive — no producer (consumer-only)

Also available via aimdb-client and aimdb-mcp tools.

💾 Persistence Layer (New Crates)

Long-term record history with pluggable backends!

use aimdb_persistence::AimDbBuilderPersistExt;
use aimdb_persistence_sqlite::SqliteBackend;

// Set up SQLite persistence with 7-day retention
let backend = SqliteBackend::new("./aimdb_history.db")?;
builder.with_persistence(backend, Duration::from_secs(7 * 24 * 3600));

// Mark records to persist
builder.configure::<Temperature>(AppKey::Temp, |reg| {
    reg.buffer(BufferCfg::SingleLatest)
       .persist("sensor.temperature");  // ← stored to SQLite
});

let db = builder.build().await?;

// Query historical data
let last_100 = db.query_latest::<Temperature>("sensor.*", Some(100)).await;
let this_week = db.query_range::<Temperature>(
    "sensor.*",
    week_start_ms,
    week_end_ms,
    None,  // no per-record limit
).await;

aimdb-persistence features:

• PersistenceBackend trait — implement your own storage
• Automatic retention cleanup task (24-hour interval)
• query_latest, query_range, query_raw APIs

aimdb-persistence-sqlite features:

• WAL journal mode for concurrent reads
• Window-function queries for efficient top-N per record
• * wildcard pattern matching
• Actor-model writer thread; Clone = O(1) handle copy

🌐 Dynamic Topic/Address Routing (Design 018)

Resolve MQTT topics or KNX group addresses at runtime based on data values!

// Outbound: per-message topic from payload
builder.configure::<SensorReading>(AppKey::Reading, |reg| {
    reg.link_to("mqtt://broker/sensors/default")
       .with_topic_provider(|reading: &SensorReading| {
           format!("sensors/{}/data", reading.sensor_id)
       });
});

// Inbound: late-binding topic from config/discovery
builder.configure::<Command>(AppKey::Command, |reg| {
    reg.link_from("mqtt://broker/")
       .with_topic_resolver(|| {
           // Called once at connector startup
           format!("commands/{}", load_device_id())
       });
});

Works in both std and no_std + alloc environments.

📡 Record Drain API (Design 019)

Non-blocking batch pull for accumulated history — perfect for LLM analysis!

// Via AimDbClient (remote access)
let response = client.drain_record("sensor.temperature").await?;
println!("Drained {} values", response.count);
// First call is always empty (cold start — creates the reader)
// Subsequent calls return everything since last drain

Cold-start semantics: the first drain call creates a reader and returns empty. Subsequent calls return all values accumulated since the previous drain. This enables stateful batch analysis without missing data.

---

💥 Breaking Changes

1. .with_serialization() → .with_remote_access()

// Before (v0.4.x)
reg.buffer(...).with_serialization();

// After (v0.5.0)
reg.buffer(...).with_remote_access();

2. RecordId::new() requires RecordOrigin

This affects custom buffer/record implementations only:

// Before (v0.4.x)
RecordId::new(type_id, idx)

// After (v0.5.0)
RecordId::new(type_id, idx, RecordOrigin::Source)

3. MCP subscription tools removed

The subscribe_record, unsubscribe_record, list_subscriptions, and get_notification_directory MCP tools have been replaced by drain_record. Update any LLM prompts or MCP tool configurations accordingly.

---

📦 Published Crates

New Crates

• 🆕 aimdb-persistence@0.1.0 — pluggable persistence layer
• 🆕 aimdb-persistence-sqlite@0.1.0 — SQLite backend for persistence

Updated

Crate	Version
aimdb-core	0.5.0
aimdb-tokio-adapter	0.5.0
aimdb-embassy-adapter	0.5.0
aimdb-client	0.5.0
aimdb-sync	0.5.0
aimdb-mqtt-connector	0.5.0
aimdb-knx-connector	0.3.0
aimdb-cli	0.5.0
aimdb-mcp	0.5.0

Unchanged

• aimdb-derive@0.1.0
• aimdb-executor@0.1.0

---

🔧 New MCP Tools

The aimdb-mcp server now provides these tools:

Tool	Description
discover_instances	Find running AimDB servers
list_records	List all records with metadata
get_record	Get current value
set_record	Set writable record value
query_schema	Infer JSON Schema from value
get_instance_info	Server version and capabilities
drain_record	Batch pull accumulated history
graph_nodes	All graph nodes with origin/buffer info
graph_edges	Directed data-flow edges
graph_topo_order	Topological record ordering

---

🖥️ New CLI Commands (aimdb graph)

# List all graph nodes (color-coded by origin)
aimdb graph nodes

# Show directed data-flow edges
aimdb graph edges

# Show topological (spawn) order
aimdb graph order

# Export to Graphviz DOT format
aimdb graph dot > pipeline.dot
dot -Tsvg pipeline.dot -o pipeline.svg

---

🚀 Migration Guide

Step 1: Update dependencies

[dependencies]
aimdb-core = "0.5.0"
aimdb-tokio-adapter = "0.5.0"
# Optional: add persistence
aimdb-persistence = "0.1.0"
aimdb-persistence-sqlite = "0.1.0"

Step 2: Rename .with_serialization() → .with_remote_access()

# Quick find & replace
grep -rn "with_serialization" src/
# Replace with:
#   .with_remote_access()

Step 3: Update RecordId::new() calls (custom buffer implementations only)

// Add RecordOrigin argument
RecordId::new(type_id, idx, RecordOrigin::Source)

Step 4: Update MCP tool usage

Replace subscribe_record / unsubscribe_record workflows with drain_record polling:

# Old: subscribe → wait → unsubscribe
# New: call drain_record periodically (first call always empty)
drain_record(socket_path: "...", record_name: "sensor.*")

---

📚 Examples

All examples updated for v0.5.0:

git clone https://github.com/aimdb-dev/aimdb.git && cd aimdb

# MQTT connector demo
cargo run -p tokio-mqtt-connector-demo

# KNX connector demo
cargo run -p tokio-knx-connector-demo

# Sync API demo
cargo run -p sync-api-demo

# Weather mesh demo (multi-station)
cargo run -p weather-hub

# Embedded (cross-compile)
cd examples/embassy-mqtt-connector-demo
cargo build --release --target thumbv7em-none-eabihf

---

🤝 Contributing

git clone https://github.com/aimdb-dev/aimdb.git
cd aimdb
make check  # fmt + clippy + test + embedded cross-compile

---

📄 License

Apache License 2.0 — see LICENSE.

---

v0.4.0

AimDB v0.4.0 Release Notes

Compile-time safe record keys with #[derive(RecordKey)] macro!

---

🎯 What's New in v0.4.0

This release introduces compile-time safe record keys via a new derive macro, transforming RecordKey from a struct to a trait. This enables user-defined enum keys with automatic string representation, eliminating runtime key typos and improving embedded system efficiency.

---

✨ Major Features

🔑 RecordKey Trait + Derive Macro

New crate aimdb-derive provides #[derive(RecordKey)] for compile-time checked keys!

Instead of error-prone string literals, define type-safe enum keys:

use aimdb_core::RecordKey;

#[derive(RecordKey, Clone, Copy, PartialEq, Eq, Debug)]
#[key_prefix = "sensor"]
pub enum SensorKey {
    #[key = "temp.indoor"]
    TempIndoor,
    
    #[key = "temp.outdoor"]
    TempOutdoor,
    
    #[key = "humidity"]
    #[link_address = "zigbee/sensors/humidity"]  // MQTT topic
    Humidity,
}

// Compile-time typo detection!
let producer = db.producer::<Temperature>(SensorKey::TempIndoor)?;
// vs runtime error with string: db.producer::<Temperature>("sensor.temp.indor")?;

Benefits:

• 🛡️ Compile-time safety: Typos caught at build time
• 🚀 Zero-allocation: Enum variants are Copy, no heap allocation
• 📍 Connector metadata: #[link_address = "..."] for MQTT topics, KNX addresses
• 🔧 IDE support: Autocomplete and refactoring work correctly
• 📦 no_std compatible: Works on embedded targets

📝 StringKey Type

The previous RecordKey struct is now StringKey with improved memory model:

use aimdb_core::StringKey;

// Static keys (zero allocation)
let key: StringKey = "sensors.temp".into();

// Dynamic keys (interned via Box::leak for O(1) Copy/Clone)
let key = StringKey::intern(dynamic_string);

Memory Model:

• Static(&'static str) - Zero-allocation for string literals
• Interned(&'static str) - Uses Box::leak for O(1) cloning
• Designed for startup-time registration (<1000 keys)
• Debug warning if >1000 interned keys

🐛 MQTT Connector Fix

Fixed initialization deadlock when subscribing to >10 MQTT topics (Issue #63)

The fix implements:

1. Spawn-before-subscribe: Event loop spawned before topic subscriptions
2. Dynamic channel capacity: Scales with topic count (topics + 10)
3. Proper task yielding: Ensures scheduler runs between operations

Also upgraded rumqttc from 0.24 to 0.25.

---

💥 Breaking Changes

1. RecordKey: Struct → Trait

RecordKey is now a trait, not a struct.

// Before (v0.3.x)
use aimdb_core::RecordKey;
let key: RecordKey = "sensors.temp".into();

// After (v0.4.0)
use aimdb_core::StringKey;
let key: StringKey = "sensors.temp".into();

// Or use derive macro (recommended)
use aimdb_core::RecordKey;  // Now a trait

#[derive(RecordKey, Clone, Copy, PartialEq, Eq)]
pub enum AppKey {
    #[key = "sensors.temp"]
    SensorsTemp,
}

2. RecordKey Trait Bounds

If you have generic code over record keys:

// Before (v0.3.x)
fn process_key(key: RecordKey) { ... }

// After (v0.4.0)
fn process_key<K: RecordKey>(key: K) { ... }
// Or with StringKey specifically:
fn process_key(key: StringKey) { ... }

---

📦 Published Crates

New Crate

• 🆕 aimdb-derive@0.1.0 - #[derive(RecordKey)] macro for compile-time checked keys

Updated to v0.4.0

• ✅ aimdb-core@0.4.0 - RecordKey trait, StringKey type, derive feature
• ✅ aimdb-mqtt-connector@0.4.0 - Deadlock fix + rumqttc 0.25
• ✅ aimdb-tokio-adapter@0.4.0 - Updated for aimdb-core 0.4.0
• ✅ aimdb-embassy-adapter@0.4.0 - Updated for aimdb-core 0.4.0
• ✅ aimdb-client@0.4.0 - Updated for aimdb-core 0.4.0
• ✅ aimdb-sync@0.4.0 - Updated for aimdb-core 0.4.0
• ✅ aimdb-cli@0.4.0 - Updated for aimdb-client 0.4.0
• ✅ aimdb-mcp@0.4.0 - Updated for aimdb-client 0.4.0

Unchanged

• aimdb-executor@0.1.0 - No changes (still compatible)
• aimdb-knx-connector@0.2.0 - No changes (still compatible)

---

🚀 Quick Start

Using Derive Macro (Recommended)

use aimdb_core::{AimDbBuilder, RecordKey, buffer::BufferCfg};
use aimdb_tokio_adapter::TokioAdapter;
use serde::{Serialize, Deserialize};
use std::sync::Arc;

// Define type-safe keys
#[derive(RecordKey, Clone, Copy, PartialEq, Eq, Debug)]
pub enum AppKey {
    #[key = "sensors.temperature"]
    #[link_address = "home/sensors/temp"]  // MQTT topic
    Temperature,
    
    #[key = "sensors.humidity"]
    #[link_address = "home/sensors/humidity"]
    Humidity,
    
    #[key = "config.settings"]
    Settings,
}

#[derive(Clone, Debug, Serialize, Deserialize)]
struct SensorReading {
    value: f32,
    timestamp: u64,
}

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let runtime = Arc::new(TokioAdapter::new()?);
    let mut builder = AimDbBuilder::new().runtime(runtime);

    // Register with type-safe keys
    builder.configure::<SensorReading>(AppKey::Temperature, |reg| {
        reg.buffer(BufferCfg::SingleLatest);
    });

    builder.configure::<SensorReading>(AppKey::Humidity, |reg| {
        reg.buffer(BufferCfg::SingleLatest);
    });

    let db = builder.build().await?;

    // Type-safe producer access
    let temp_producer = db.producer::<SensorReading>(AppKey::Temperature)?;
    let humidity_producer = db.producer::<SensorReading>(AppKey::Humidity)?;

    // Produce data
    temp_producer.produce(SensorReading { value: 22.5, timestamp: 1000 }).await?;
    humidity_producer.produce(SensorReading { value: 65.0, timestamp: 1001 }).await?;

    // Access link address for connector metadata
    println!("Temperature MQTT topic: {:?}", AppKey::Temperature.link_address());
    // Prints: Some("home/sensors/temp")

    Ok(())
}

Using StringKey (Dynamic Keys)

use aimdb_core::{AimDbBuilder, StringKey, buffer::BufferCfg};
use aimdb_tokio_adapter::TokioAdapter;
use std::sync::Arc;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let runtime = Arc::new(TokioAdapter::new()?);
    let mut builder = AimDbBuilder::new().runtime(runtime);

    // Static string literals (zero allocation)
    builder.configure::<String>("app.logs", |reg| {
        reg.buffer(BufferCfg::SpmcRing { capacity: 1000 });
    });

    // Dynamic keys (for runtime-determined keys)
    let tenant_id = "tenant-123";
    let dynamic_key = StringKey::intern(format!("tenants.{}.data", tenant_id));
    builder.configure::<String>(dynamic_key, |reg| {
        reg.buffer(BufferCfg::Mailbox);
    });

    let db = builder.build().await?;
    Ok(())
}

---

📚 Migration Guide

Step 1: Update Dependencies

[dependencies]
aimdb-core = "0.4.0"
aimdb-tokio-adapter = "0.4.0"
# aimdb-mqtt-connector = "0.4.0"  # If using MQTT

# Enable derive macro (included by default)
# aimdb-core = { version = "0.4.0", features = ["derive"] }

Step 2: Choose Key Strategy

Option A: Keep String Literals (Minimal Change)

// Before (v0.3.x)
builder.configure::<Temperature>("sensors.temp", |reg| { ... });

// After (v0.4.0) - No change needed! &'static str implements RecordKey
builder.configure::<Temperature>("sensors.temp", |reg| { ... });

Option B: Adopt Derive Macro (Recommended)

// Define keys once
#[derive(RecordKey, Clone, Copy, PartialEq, Eq)]
pub enum Keys {
    #[key = "sensors.temp"]
    SensorsTemp,
}

// Use everywhere
builder.configure::<Temperature>(Keys::SensorsTemp, |reg| { ... });
let producer = db.producer::<Temperature>(Keys::SensorsTemp)?;

Step 3: Update RecordKey Imports

// Before (v0.3.x)
use aimdb_core::RecordKey;
let key: RecordKey = "name".into();

// After (v0.4.0)
use aimdb_core::StringKey;
let key: StringKey = "name".into();

Step 4: Update Generic Code

// Before (v0.3.x)
fn process(key: RecordKey) { ... }

// After (v0.4.0)
fn process<K: RecordKey>(key: K) { ... }
// Or specifically:
fn process(key: impl RecordKey) { ... }
fn process(key: StringKey) { ... }

---

🎯 Derive Macro Reference

Attributes

Attribute	Level	Required	Description
#[key = "..."]	Variant	Yes	String representation for the key
#[key_prefix = "..."]	Enum	No	Prefix prepended to all variant keys
#[link_address = "..."]	Variant	No	Connector metadata (MQTT topic, KNX address)

Example with All Features

use aimdb_core::RecordKey;

#[derive(RecordKey, Clone, Copy, PartialEq, Eq, Debug)]
#[key_prefix = "home.automation"]  // Applied to all variants
pub enum HomeKey {
    #[key = "lights.living"]
    #[link_address = "1/1/1"]  // KNX group address
    LightsLiving,
    
    #[key = "lights.bedroom"]
    #[link_address = "1/1/2"]
    LightsBedroom,
    
    #[key = "thermostat"]
    #[link_address = "mqtt://home/thermostat"]
    Thermostat,
}

// Generated methods:
// HomeKey::LightsLiving.as_str() -> "home.automation.lights.living"
// HomeKey::LightsLiving.link_address() -> Some("1/1/1")

Compile-Time Validation

The macro validates at compile time:

• ✅ All variants have #[key = "..."] attribute
• ✅ No duplicate keys (including after prefix)
• ✅ Only unit variants (no tuple/struct variants)

// Compile error: duplicate key
#[derive(RecordKey)]
pub enum BadKeys {
    #[key = "same"]
    First,
    #[key = "same"]  // Error: duplicate key "same"
    Second,
}

// Compile error: missing key attribute
#[derive(RecordKey)]
pub enum BadKeys {
    #[key = "valid"]
    First,
    Second,  // Error: missing #[key = "..."] attribute
}

---

🐛 Bug Fixes

MQTT Connector Deadlock (Issue #63)

Problem: When subscribing to more than 10 MQTT topics, the connector would deadlock during initialization.

Root Cause: The internal channel had a fixed capacity of 10, and subscriptions were made before spawning the event loop, causing the channel to fill up and block.

**Solu...

v0.3.0

AimDB v0.3.0 Release Notes

Major update with RecordId/RecordKey architecture and buffer metrics!

---

🎯 What's New in v0.3.0

This release introduces a complete architectural overhaul of AimDB's internal record storage system, enabling multi-instance records (same type, different keys), stable O(1) indexing, and comprehensive buffer metrics. The new RecordId/RecordKey system provides both the performance benefits of numeric indexing and the usability of human-readable keys.

---

✨ Major Features

🔑 RecordId + RecordKey Architecture

Complete rewrite of internal storage for stable record identification!

AimDB now supports multiple records of the same type with unique keys, enabling patterns like:

• Multiple sensors of the same type (Temperature) with different keys ("sensors.indoor", "sensors.outdoor")
• Multi-tenant configurations ("tenant.a.config", "tenant.b.config")
• Regional data streams ("region.us.metrics", "region.eu.metrics")

Key Components:

• RecordId: u32 index wrapper for O(1) Vec-based hot-path access
• RecordKey: Hybrid &'static str / Arc<str> with zero-alloc static keys and flexible dynamic keys
• O(1) key resolution via HashMap<RecordKey, RecordId>
• Type introspection via HashMap<TypeId, Vec<RecordId>>

New API:

use aimdb_core::{AimDbBuilder, buffer::BufferCfg};
use aimdb_tokio_adapter::TokioAdapter;
use serde::{Serialize, Deserialize};
use std::sync::Arc;

#[derive(Clone, Debug, Serialize, Deserialize)]
struct Temperature {
    celsius: f32,
    sensor_id: String,
}

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let runtime = Arc::new(TokioAdapter::new()?);
    
    let mut builder = AimDbBuilder::new().runtime(runtime);

    // Register MULTIPLE records of the same type with different keys
    builder.configure::<Temperature>("sensors.indoor", |reg| {
        reg.buffer(BufferCfg::SingleLatest);
    });

    builder.configure::<Temperature>("sensors.outdoor", |reg| {
        reg.buffer(BufferCfg::SingleLatest);
    });

    let db = builder.build().await?;

    // Key-based access for multi-instance records
    let indoor_producer = db.producer_by_key::<Temperature>("sensors.indoor")?;
    let outdoor_producer = db.producer_by_key::<Temperature>("sensors.outdoor")?;

    indoor_producer.produce(Temperature { celsius: 22.5, sensor_id: "indoor-1".into() }).await?;
    outdoor_producer.produce(Temperature { celsius: 15.2, sensor_id: "outdoor-1".into() }).await?;

    // Introspection
    let temp_ids = db.records_of_type::<Temperature>();  // Returns &[RecordId] with 2 IDs
    let id = db.resolve_key("sensors.indoor");           // Returns Option<RecordId>

    Ok(())
}

Key Naming Conventions:

• Use dot-separated hierarchical names: "sensors.indoor", "config.app"
• Keys must be unique across all records (duplicate keys panic at registration)
• Static string literals ("key") are zero-allocation via &'static str
• Dynamic keys (String::from("key")) use Arc<str> for efficient cloning

📊 Buffer Metrics API (Feature-Gated)

Comprehensive buffer introspection for monitoring and debugging!

Enable buffer metrics with the metrics feature flag:

[dependencies]
aimdb-core = { version = "0.3.0", features = ["metrics"] }
aimdb-tokio-adapter = { version = "0.3.0", features = ["metrics"] }

New Metrics:

use aimdb_core::buffer::BufferMetricsSnapshot;

// Get metrics from any record
let metadata = db.record_metadata::<Temperature>("sensors.indoor")?;

if let Some(metrics) = metadata.buffer_metrics {
    println!("Produced: {}", metrics.produced_count);
    println!("Consumed: {}", metrics.consumed_count);
    println!("Dropped: {}", metrics.dropped_count);
    println!("Occupancy: {}/{}", metrics.occupancy.0, metrics.occupancy.1);
}

Available Metrics:

• produced_count: Total items pushed to the buffer
• consumed_count: Total items consumed across all readers
• dropped_count: Total items dropped due to lag (per-reader semantics documented)
• occupancy: Current buffer fill level as (current, capacity) tuple

Supported Buffers:

• ✅ SPMC Ring Buffer: Full metrics support
• ✅ SingleLatest: Full metrics support
• ✅ Mailbox: Full metrics support

Tokio Adapter: Full implementation with atomic counters
Embassy Adapter: Feature flag present (API consistency), but metrics not functional on embedded targets (requires std)

🔍 Enhanced Introspection API

New methods for exploring records at runtime:

// Find all records of a specific type
let temperature_records = db.records_of_type::<Temperature>();
for record_id in temperature_records {
    let metadata = db.record_metadata_by_id(*record_id)?;
    println!("Temperature record: {} (key: {})", record_id.0, metadata.record_key);
}

// Resolve key to RecordId
if let Some(record_id) = db.resolve_key("sensors.indoor") {
    println!("Found record with ID: {}", record_id.0);
}

// Key-bound producers/consumers
let producer = db.producer_by_key::<Temperature>("sensors.indoor")?;
let consumer = db.consumer_by_key::<Temperature>("sensors.outdoor")?;

println!("Producer key: {}", producer.key());
println!("Consumer key: {}", consumer.key());

🏗️ Internal Architecture Improvements

Optimized storage for sub-50ms latency:

Before (v0.2.0):

BTreeMap<TypeId, Box<dyn AnyRecord>>  // O(log n) lookups

After (v0.3.0):

Vec<Box<dyn AnyRecord>>                      // O(1) hot-path access by RecordId
HashMap<RecordKey, RecordId>                 // O(1) name lookups
HashMap<TypeId, Vec<RecordId>>               // O(1) type introspection

Performance Benefits:

• ✅ O(1) hot-path access (was O(log n))
• ✅ Stable RecordId across application lifetime
• ✅ Zero-allocation static keys
• ✅ Efficient multi-instance type lookups

---

🔨 Breaking Changes

1. Record Registration API

All records now require a key parameter:

// Before (v0.2.x)
builder.configure::<Temperature>(|reg| {
    reg.buffer(BufferCfg::SingleLatest);
});

// After (v0.3.0)
builder.configure::<Temperature>("sensor.temperature", |reg| {
    reg.buffer(BufferCfg::SingleLatest);
});

2. Type-Based Lookup Ambiguity

If you register multiple records of the same type, type-based methods return AmbiguousType error:

// With multiple Temperature records registered...
db.produce(temp).await  // ❌ Returns Err(AmbiguousType { count: 2, ... })

// Use key-based methods instead:
db.produce_by_key("sensors.indoor", temp).await  // ✅ Works correctly

Migration Strategy:

• Single-instance records: Type-based API still works (produce(), subscribe(), etc.)
• Multi-instance records: Use key-based API (produce_by_key(), subscribe_by_key(), etc.)

3. DynBuffer Implementation

Custom buffer implementations must now explicitly implement DynBuffer<T>:

// Before (v0.2.x) - automatic via blanket impl
impl<T: Clone + Send> Buffer<T> for MyBuffer<T> { ... }
// DynBuffer was automatically implemented

// After (v0.3.0) - explicit implementation required
impl<T: Clone + Send> Buffer<T> for MyBuffer<T> { ... }

impl<T: Clone + Send + 'static> DynBuffer<T> for MyBuffer<T> {
    fn push(&self, value: T) {
        <Self as Buffer<T>>::push(self, value)
    }
    
    fn subscribe_boxed(&self) -> Box<dyn BufferReader<T> + Send> {
        Box::new(self.subscribe())
    }
    
    fn as_any(&self) -> &dyn core::any::Any {
        self
    }
    
    // Optional: implement metrics_snapshot() if you support metrics
    #[cfg(feature = "metrics")]
    fn metrics_snapshot(&self) -> Option<BufferMetricsSnapshot> {
        None // or Some(...) if you track metrics
    }
}

Why this change? Enables adapters to provide metrics_snapshot() when the metrics feature is enabled.

4. RecordMetadata Changes

New fields added to RecordMetadata:

pub struct RecordMetadata {
    pub record_id: u32,           // ← NEW: Stable numeric identifier
    pub record_key: String,       // ← NEW: Human-readable key
    pub type_id: u64,
    pub type_name: String,
    pub buffer_type: String,
    pub buffer_capacity: Option<usize>,
    pub producer_count: usize,
    pub consumer_count: usize,
    pub outbound_connector_count: usize,
    pub inbound_connector_count: usize,
    #[cfg(feature = "metrics")]
    pub buffer_metrics: Option<BufferMetricsSnapshot>,  // ← NEW: Buffer metrics
}

---

📦 Published Crates

Updated to v0.3.0

• ✅ aimdb-core@0.3.0 - RecordId/RecordKey architecture + buffer metrics
• ✅ aimdb-tokio-adapter@0.3.0 - Buffer metrics implementation + multi-instance tests
• ✅ aimdb-embassy-adapter@0.3.0 - Explicit DynBuffer implementation + metrics feature flag
• ✅ aimdb-client@0.3.0 - Updated for new RecordMetadata fields
• ✅ aimdb-sync@0.3.0 - Updated for key-based registration API
• ✅ aimdb-mqtt-connector@0.3.0 - Updated for key-based registration + rumqttc 0.25
• ✅ aimdb-cli@0.3.0 - Updated formatters for RecordId/RecordKey display
• ✅ aimdb-mcp@0.3.0 - Updated tools for RecordId/RecordKey introspection

Updated to v0.2.0

• ✅ aimdb-knx-connector@0.2.0 - Updated for key-based registration (first stable release)

Unchanged

• aimdb-executor@0.1.0 - No changes (still compatible)

---

🚀 Quick Start

Multi-Instance Records Example

use aimdb_core::{AimDbBuilder, buffer::BufferCfg};
use aimdb_tokio_adapter::TokioAdapter;
use serde::{Serialize, Deserialize};
use std::sync::Arc;

#[derive(Clone, Debug, Serialize, Deserialize)]
struct SensorReading {...

v0.2.0

Summary

This release introduces bidirectional connector support, enabling true two-way data synchronization between AimDB and external systems. The new architecture supports simultaneous publishing and subscribing with automatic message routing, working seamlessly across both Tokio (std) and Embassy (no_std) runtimes.

Highlights

• 🔄 Bidirectional Connectors: New .link_to() and .link_from() APIs for clear directional data flows
• 🎯 Type-Erased Router: Automatic routing of incoming messages to correct typed producers
• 🏗️ ConnectorBuilder Pattern: Simplified connector registration with automatic initialization
• 📡 Enhanced MQTT Connector: Complete rewrite supporting simultaneous pub/sub with automatic reconnection
• 🌐 Embassy Network Integration: Connectors can now access Embassy's network stack for network operations
• 📚 Comprehensive Guide: New 1000+ line connector development guide with real-world examples

Breaking Changes

• aimdb-core: .build() is now async; connector registration changed from .with_connector(scheme, instance) to .with_connector(builder)
• aimdb-core: .link() deprecated in favor of .link_to() (outbound) and .link_from() (inbound)
• aimdb-core: Outbound connector architecture refactored to trait-based system:
  • Removed: TypedRecord::spawn_outbound_consumers() method (was called automatically)
  • Added: ConsumerTrait, AnyReader traits for type-erased outbound routing
  • Added: AimDb::collect_outbound_routes() method to gather configured routes
  • Required: Connectors must implement spawn_outbound_publishers() and call it in ConnectorBuilder::build()
• aimdb-mqtt-connector: API changed from MqttConnector::new() to MqttConnectorBuilder::new(); automatic task spawning removes need for manual background task management
• aimdb-mqtt-connector: Added spawn_outbound_publishers() method; must be called in build() for outbound publishing to work

Modified Crates

See individual changelogs for detailed changes:

• aimdb-core: Core connector architecture, router system, bidirectional APIs
• aimdb-tokio-adapter: Connector builder integration
• aimdb-embassy-adapter: Network stack access, connector support
• aimdb-mqtt-connector: Complete bidirectional rewrite for both runtimes
• aimdb-sync: Compatibility with async build

Migration Guide

1. Update connector registration:

// Old (v0.1.0)
let mqtt = MqttConnector::new("mqtt://broker:1883").await?;
builder.with_connector("mqtt", Arc::new(mqtt))

// New (v0.2.0)
builder.with_connector(MqttConnectorBuilder::new("mqtt://broker:1883"))

2. Make build() async:

// Old (v0.1.0)
let db = builder.build()?;

// New (v0.2.0)
let db = builder.build().await?;

3. Use directional link methods:

// Old (v0.1.0)
.link("mqtt://sensors/temp")

// New (v0.2.0)
.link_to("mqtt://sensors/temp")    // For publishing (AimDB → External)
.link_from("mqtt://commands/temp")  // For subscribing (External → AimDB)

4. Remove manual MQTT task spawning (Embassy):

// Old (v0.1.0) - Manual task spawning required
let mqtt_result = MqttConnector::create(...).await?;
spawner.spawn(mqtt_task(mqtt_result.task))?;
builder.with_connector("mqtt", Arc::new(mqtt_result.connector))

// New (v0.2.0) - Automatic task spawning
builder.with_connector(MqttConnectorBuilder::new("mqtt://broker:1883"))
// Tasks spawn automatically during build()

5. Update custom connectors to spawn outbound publishers:

If you've implemented a custom connector, you must add spawn_outbound_publishers() support:

// Old (v0.1.0) - Outbound consumers spawned automatically
impl ConnectorBuilder for MyConnectorBuilder {
    fn build<R>(&self, db: &AimDb<R>) -> DbResult<Arc<dyn Connector>> {
        // ... setup code ...
        Ok(Arc::new(MyConnector { /* fields */ }))
    }
    // Outbound publishing happened automatically via TypedRecord::spawn_outbound_consumers()
}

// New (v0.2.0) - Must explicitly spawn outbound publishers
impl ConnectorBuilder for MyConnectorBuilder {
    fn build<R>(&self, db: &AimDb<R>) -> DbResult<Arc<dyn Connector>> {
        // ... setup code ...
        let connector = MyConnector { /* fields */ };
        
        // REQUIRED: Collect and spawn outbound publishers
        let outbound_routes = db.collect_outbound_routes(self.protocol_name());
        connector.spawn_outbound_publishers(db, outbound_routes)?;
        
        Ok(Arc::new(connector))
    }
}

// REQUIRED: Implement spawn_outbound_publishers method
impl MyConnector {
    fn spawn_outbound_publishers<R: RuntimeAdapter + 'static>(
        &self,
        db: &AimDb<R>,
        routes: Vec<(String, Box<dyn ConsumerTrait>, SerializerFn, Vec<(String, String)>)>,
    ) -> DbResult<()> {
        for (topic, consumer, serializer, _config) in routes {
            let client = self.client.clone();
            let topic_clone = topic.clone();
            
            db.runtime().spawn(async move {
                // Subscribe to record updates using ConsumerTrait
                match consumer.subscribe_any().await {
                    Ok(mut reader) => {
                        loop {
                            match reader.recv_any().await {
                                Ok(value) => {
                                    // Serialize and publish
                                    if let Ok(bytes) = serializer(&*value) {
                                        let _ = client.publish(&topic_clone, bytes).await;
                                    }
                                }
                                Err(_) => break,
                            }
                        }
                    }
                    Err(_) => { /* Log error */ }
                }
            })?;
        }
        Ok(())
    }
}

Why this change? The new trait-based architecture provides:

• ✅ Symmetry with inbound routing (ProducerTrait ↔ ConsumerTrait)
• ✅ Testability (can mock ConsumerTrait without real records)
• ✅ Type safety via factory pattern (type capture at configuration time)
• ✅ Maintainability (connector logic stays in connector crate)

Migration checklist for custom connectors:

• Add spawn_outbound_publishers() method to connector implementation
• Call db.collect_outbound_routes(protocol_name) in ConnectorBuilder::build()
• Call connector.spawn_outbound_publishers(db, routes)? before returning
• Use ConsumerTrait::subscribe_any() to get type-erased readers
• Handle serialization with provided SerializerFn
• Test both inbound (.link_from()) and outbound (.link_to()) data flows

See Connector Development Guide for complete examples.

Documentation

• Added comprehensive Connector Development Guide
• Updated MQTT connector examples for both Tokio and Embassy
• Enhanced API documentation with bidirectional patterns

v0.1.0

Added

Core Database (aimdb-core)

• Initial release of AimDB async in-memory database engine
• Type-safe record system using TypeId-based routing
• Three buffer types for different data flow patterns:
  • SPMC Ring Buffer: High-frequency data streams with bounded memory
  • SingleLatest: State synchronization and configuration updates
  • Mailbox: Commands and one-shot events
• Producer-consumer model with async task spawning
• Runtime adapter abstraction for cross-platform support
• no_std compatibility for embedded targets
• Error handling with comprehensive DbResult<T> and DbError types
• Remote access protocol (AimX v1) for cross-process introspection

Runtime Adapters

• Tokio Adapter (aimdb-tokio-adapter): Full-featured std runtime support
  • Lock-free buffer implementations
  • Configurable buffer capacities
  • Comprehensive async task spawning
• Embassy Adapter (aimdb-embassy-adapter): Embedded no_std runtime support
  • Configurable task pool sizes (8/16/32 concurrent tasks)
  • Optimized for resource-constrained devices
  • Compatible with ARM Cortex-M targets

MQTT Connector (aimdb-mqtt-connector)

• Dual runtime support for both Tokio and Embassy
• Automatic consumer registration via builder pattern
• Topic mapping with QoS and retain configuration
• Pluggable serializers (JSON, MessagePack, Postcard, custom)
• Automatic reconnection handling
• Uses rumqttc for std environments
• Uses mountain-mqtt for embedded environments

Developer Tools

• MCP Server (aimdb-mcp): LLM-powered introspection and debugging
  • Discover running AimDB instances
  • Query record values and schemas
  • Subscribe to real-time updates
  • Set writable record values
  • JSON Schema inference from record values
  • Notification persistence to JSONL files
• CLI Tool (aimdb-cli): Command-line interface (skeleton)
  • Instance discovery and management commands
  • Record inspection capabilities
  • Real-time watch functionality
• Client Library (aimdb-client): Reusable connection and discovery logic
  • Unix domain socket communication
  • AimX v1 protocol implementation
  • Clean error handling

Synchronous API (aimdb-sync)

• Blocking wrapper around async AimDB core
• Thread-safe synchronous record access
• Automatic Tokio runtime management
• Ideal for gradual migration from sync to async

Documentation & Examples

• Comprehensive README with architecture overview
• Individual crate documentation with examples
• 12 detailed design documents in /docs/design
• Working examples:
  • tokio-mqtt-connector-demo: Full Tokio MQTT integration
  • embassy-mqtt-connector-demo: Embedded RP2040 with WiFi MQTT
  • sync-api-demo: Synchronous API usage patterns
  • remote-access-demo: Cross-process introspection server

Build & CI Infrastructure

• Comprehensive Makefile with color-coded output
• GitHub Actions workflows:
  • Continuous integration (format, lint, test)
  • Security audits (weekly schedule + on-demand)
  • Documentation generation
  • Release automation
• Cross-compilation testing for thumbv7em-none-eabihf target
• cargo-deny configuration for license and dependency auditing
• Dev container setup for consistent development environment

Design Goals Achieved

• ✅ Sub-50ms latency for real-time synchronization
• ✅ Lock-free buffer operations
• ✅ Cross-platform support (MCU → edge → cloud)
• ✅ Type safety with zero-cost abstractions
• ✅ Protocol-agnostic connector architecture

Known Limitations

• Kafka and DDS connectors planned for future releases
• CLI tool is currently skeleton implementation
• Performance benchmarks not yet included
• Limited to Unix domain sockets for remote access (no TCP yet)

Dependencies

• Rust 1.75+ required
• Tokio 1.47+ for std environments
• Embassy 0.9+ for embedded environments
• See deny.toml for approved dependency licenses

Breaking Changes

None (initial release)

Migration Guide

Not applicable (initial release)

---

Release Notes

v0.1.0 - "Foundation Release"

AimDB v0.1.0 establishes the foundational architecture for async, in-memory data synchronization across MCU → edge → cloud environments. This release focuses on core functionality, dual runtime support, and developer tooling.

Highlights:

• 🚀 Dual runtime support: Works on both standard library (Tokio) and embedded (Embassy)
• 🔒 Type-safe record system eliminates runtime string lookups
• 📦 Three buffer types cover most real-time data patterns
• 🔌 MQTT connector works in both std and no_std environments
• 🤖 MCP server enables LLM-powered introspection
• ✅ 27+ core tests, comprehensive CI/CD, security auditing

Get Started:

cargo add aimdb-core aimdb-tokio-adapter

See README.md for quickstart guide and examples.

Feedback Welcome:
This is an early release. Please report issues, suggest features, or contribute at:
https://github.com/aimdb-dev/aimdb

---