- Start with Shared Prerequisites and How to Run.
- Then pick a category and example that fits your task.

## Quick start
```
docker-compose -f tools/docker-compose.kafka.yml up -d
cd examples/basic-produce-consume
dotnet run
```

## Shared Prerequisites (run once)
- Install the .NET 8 SDK
- Start local Kafka + Schema Registry + ksqlDB
  - `docker-compose -f tools/docker-compose.kafka.yml up -d`

## How to Run (common)
- Example: `basic-produce-consume`
  - `cd examples/basic-produce-consume`
  - `dotnet run`
- Each example includes its own `appsettings.json` (or see its README)

---

## Basics (first touch)
- `basic-produce-consume`: Producer/Consumer fundamentals. Send `BasicMessage` and receive with `ForEachAsync`.
- `hello-world`: Minimal setup - define a POCO, send, wait, and receive (all in `Program.cs`).

## Configuration (settings & attributes)
- `configuration`: Minimal `appsettings.json` and Builder setup (connection/Topic/Consumer/Producer).
- `configuration-mapping`: Switch logging settings between Development and Production.
- `schema-attributes`: Use `[KsqlKey]`, `[KsqlDecimal]`, and `[KsqlTimestamp]`.
- `headers-meta`: Handling message headers and metadata.

## Query Basics (LINQ -> KSQL)
- `query-basics`: Core LINQ -> KSQL flow (introduces View/ToQuery).
- `query-filter`: Filtering with `.Where(...)`.
- `view-toquery`: Fundamentals of View/ToQuery.
- `table-cache-lookup`: Refer to a local cache via `[KsqlTable]`.
- `pull-query`: Issue Pull queries against materialized views.

## Windowing (time windows & aggregation)
- `windowing`: Live aggregation (Push) and a 1-minute -> 5-minute roll-up atop TUMBLING basics.
  - Unified from `examples/tumbling-live-consumer` and `examples/rollup-1m-5m-verify`.
- `continuation-schedule`: Continuation-based scheduling sample (use `Tumbling(..., continuation: true)` and include `WindowStart()` once in `Select`).
- `bar-1m-live-consumer`: Monitor `bar_1m_live` directly with a dedicated POCO.

## Error Handling (operations & reprocessing)
- `error-handling`: Basics of `OnError`/Retry (introduces retry strategies).
- `error-handling-dlq`: Park invalid messages in a DLQ (`.OnError(ErrorAction.DLQ)`).
- `manual-commit`: Manage commits manually (`autoCommit: false`).
- `retry-onerror`: Retry patterns.

## Operations & maintenance
- `streamiz-clear`: Clear Streamiz local state caches (RocksDB). One-off maintenance tool.
  - Run: `dotnet run --project examples/streamiz-clear/StreamizClear.csproj`
- `runtime-events`: Observe runtime/diagnostic events and counters.

## Advanced (verification & application)
- `daily-comparison`: Daily aggregation - import rates, roll up 1/5/60-minute data, then verify daily totals.
- `oss-bars-verify`: Verify OSS around bars.
- `deduprates-producer`: Produce de-duplicated rates.
- `rows-last-assignment`: Deterministic last-value assignment patterns with traces.

---

## Reference Docs (wiki)
- Function/type mapping: [Function-Support](Function-Support)
- KSQL API usage: [KSQL-API](KSQL-API)
- Configuration details: [Configuration-Reference](Configuration-Reference)
- Getting started: [Quick-Start](Quick-Start)