40 lines
1.4 KiB
Markdown
40 lines
1.4 KiB
Markdown
# Sprint 148 Plan: Generic Text-to-AST Incremental Synchronization Engine
|
|
|
|
## Context
|
|
|
|
Sprint 148 implements shared incremental text-to-AST synchronization logic that
|
|
works across language adapters via explicit adapter hooks.
|
|
|
|
---
|
|
|
|
## Goals
|
|
|
|
1. Implement language-neutral incremental sync engine contracts
|
|
2. Support adapter-provided delta parsing and node identity policies
|
|
3. Emit deterministic diagnostics and confidence scoring across languages
|
|
4. Validate repeated edit streams in at least two languages
|
|
|
|
---
|
|
|
|
## Steps
|
|
|
|
### Step 1719: Generic incremental edit packet schema (10 tests)
|
|
### Step 1720: Adapter sync hook contract model (10 tests)
|
|
### Step 1721: Cross-language node identity preservation policy model (10 tests)
|
|
### Step 1722: Sync diagnostics normalization model (10 tests)
|
|
### Step 1723: `whetstone_sync_text_to_ast_generic` MCP tool (8 tests)
|
|
### Step 1724: `whetstone_get_sync_trace_generic` MCP tool (8 tests)
|
|
### Step 1725: `whetstone_compare_sync_outcomes` MCP tool (8 tests)
|
|
### Step 1726: Cross-language sync confidence packet model (8 tests)
|
|
### Step 1727: Generic sync determinism report artifact (8 tests)
|
|
### Step 1728: Sprint 148 integration summary + regression (8 tests)
|
|
|
|
---
|
|
|
|
## Architecture Gate
|
|
|
|
- Sync engine must be adapter-driven, not language-hardcoded
|
|
- Determinism checks must pass for identical edit streams per language
|
|
- Diagnostics must remain normalized and machine-readable across adapters
|
|
- Max 600 lines per header
|