Add 1.0 release scope

docs/release-scope.md

@@ -0,0 +1,167 @@
+# Valkey-Swift 1.0 Release Scope
+
+This page outlines the complete scope, feature requirements, and readiness checklist for the **Valkey-Swift 1.0** release.
+
+---
+
+## Contents
+- [Commands](#commands)
+- [Pipelines](#pipelines)
+- [Transactions](#transactions)
+- [Error Handling](#error-handling)
+- [Security](#security)
+- [Timeouts and Retries](#timeouts-and-retries)
+- [Cluster Mode Support](#cluster-mode-support)
+- [Debug-ability](#debug-ability)
+- [Load Balancing](#load-balancing)
+- [Circuit Breaking](#circuit-breaking)
+- [Additional Items](#additional-items)
+- [1.0 Release Readiness Checklist](#10-release-readiness-checklist)
+
+---
+
+# Commands
+
+## Command Coverage
+Valkey-Swift uses a code generator based on JSON definitions from the Valkey server repository.
+
+- Full support for all standard commands  
+- Support for JSON and Bloom module commands  
+
+## Command Return Types
+- Avoid returning raw `RESPToken` or `ByteBuffer` unless they map directly to primitives  
+- Provide typed responses for complex structures  
+- Reduce decoding burden for client applications  
+
+---
+
+# Pipelines
+
+- Support for **node-level** and **cluster-level** pipelines  
+- Pipelines return standardized Swift types  
+- `ASK` and `MOVED` redirections handled transparently  
+
+---
+
+# Transactions
+
+- Support for node-level and cluster-level transactions  
+- Transactions return typed responses  
+- Automatic handling of redirections (`ASK`, `MOVED`)  
+
+---
+
+# Error Handling
+
+- All errors are typed using **`ValkeyClientError`**  
+- Errors should capture the call site metadata including file, function, line, reason   
+
+### Notes
+- File/line capture not implemented yet  
+- Some NIO limits affect error propagation  
+- Backward compatibility must be maintained  
+
+---
+
+# Security
+
+## TLS / mTLS
+- TLS supported  
+- mTLS supported via SwiftNIO TLS  
+
+## Authentication
+- Username/password authentication supported  
+
+---
+
+# Timeouts and Retries
+
+## Retries
+- Retry logic is **not** part of 1.0
+
+## Timeouts
+- Command timeout  
+- Blocking command timeout  
+
+---
+
+# Cluster Mode Support
+
+## Redirections
+Must handle:
+- `ASK`  
+- `MOVED`  
+
+For:
+- Commands  
+- Pipelines  
+- Transactions  
+
+## Read From Replicas
+- `READONLY` supported  
+- Graceful shutdown pending upstream fix  
+
+## Topology Changes
+Client should:
+- Handle failovers + resharding  
+- Refresh topology after failures  
+- Refresh immediately on `MOVED`  
+
+---
+
+# Debug-ability
+
+## Logging
+- Logging at crucial code paths  
+- Defaults to debug/trace  
+- Application handles top-level error logs  
+
+## Observability
+Out of scope for 1.0:
+- Latency metrics  
+- Throughput  
+- Failure rate  
+- Command profiling  
+
+---
+
+# Load Balancing
+
+Supported:
+- Round-robin (in progress)  
+- Random strategy  
+
+Not in 1.0:
+- Weighted LB  
+- Latency-aware LB  
+
+---
+
+# Circuit Breaking
+
+Not part of 1.0. Future plans:
+- Detect unhealthy nodes  
+- Health-check integration  
+
+---
+
+# Additional Items
+
+- Resolve remaining GitHub issues  
+- Document API behaviors and design choices  
+
+---
+
+# 1.0 Release Readiness Checklist
+
+- [ ] [All standard commands implemented with typed return values](#command-coverage)
+- [x] [Node + cluster pipelines tested and validated](#pipelines)
+- [x] [Cluster transaction support functional](#transactions)
+- [ ] [Typed errors with file/line capture](#error-handling)
+- [x] [TLS/mTLS validated](#security)
+- [x] [ASK/MOVED + READONLY + topology refresh](#cluster-mode-support)
+- [ ] [Debug/trace logs added across critical paths](#debug-ability)
+- [x] [Command timeout support](#timeouts-and-retries)
+- [ ] [Resolve critical GitHub issues](#additional-items)
+- [ ] API docs + changelog finalized
+- [ ] Publish 1.0 tag + changelog