mirror of
https://github.com/panatgithub/AnkiHeadingSync.git
synced 2026-07-22 17:10:28 +00:00
fbb289850a
删除已失效的 legacy sync use case、syncRegistry 持久化与旧 card pipeline，运行入口保持 ManualSyncService 主链。 Remove the obsolete legacy sync use cases, syncRegistry persistence, and old card pipeline while keeping the runtime on ManualSyncService.
65 lines
No EOL
2.8 KiB
Markdown
65 lines
No EOL
2.8 KiB
Markdown
# Module 2 Gap Report
|
|
|
|
> Archived on 2026-04-21.
|
|
> This document describes the removed legacy syncRegistry-based sync chain.
|
|
> The active runtime path is now the ManualSyncService manual-sync pipeline.
|
|
|
|
## Current State
|
|
|
|
- Card extraction only derives legacy card identity from heading structure and source position.
|
|
- `CardDraft` has no `embeddedNoteId` field.
|
|
- `SourceLocation` has no block-end marker positioning data beyond `blockEndLine`.
|
|
- `SyncRecord` only stores legacy `cardKey -> noteId` mappings.
|
|
- `SyncPlanningService` only matches cards by legacy `cardKey`.
|
|
- `ExecuteSyncPlanUseCase` only supports add/update driven by plan entries and never writes back into Markdown.
|
|
- `VaultGateway` cannot replace Markdown content with optimistic concurrency.
|
|
- `AnkiGateway` cannot batch-check note existence/model via `noteId`.
|
|
|
|
## Already Correct
|
|
|
|
- Card rendering is separate from extraction and can stay identity-agnostic.
|
|
- Sync registry persistence is already centralized in the shared plugin data snapshot.
|
|
- Execution already validates field mappings before side effects.
|
|
- Source scanning and sync execution are already split into `scan/plan` and `execute` phases.
|
|
|
|
## Missing For Module 2
|
|
|
|
### Extraction
|
|
|
|
- Parse block-end `<!-- AHS:<noteId> -->` markers.
|
|
- Reject malformed or duplicated markers in the same heading block.
|
|
- Exclude marker lines from `bodyMarkdown`.
|
|
- Populate `embeddedNoteId` and marker positioning metadata.
|
|
|
|
### Identity And Registry
|
|
|
|
- Extend `SyncRecord` with `identityMode` and optional `legacyCardKey`.
|
|
- Keep legacy `cardKey` lookup while adding embedded `noteId` lookup.
|
|
- Preserve pending write-back state when Anki note creation succeeds but Markdown write-back fails.
|
|
|
|
### Planning
|
|
|
|
- Prefer embedded `noteId` over legacy `cardKey` when available.
|
|
- Allow embedded-note-id cards to update even if the local registry entry is missing.
|
|
- Detect orphans independently for embedded-note-id and legacy-card-key records.
|
|
|
|
### Execution
|
|
|
|
- Validate embedded `noteId` existence/model before update.
|
|
- Recreate missing embedded-note-id notes and replace the block-end marker.
|
|
- Write back a new marker for newly created notes.
|
|
- Reject basic/cloze type switches for embedded-note-id cards.
|
|
- Retry pending marker write-back on later legacy hits.
|
|
|
|
### Infrastructure
|
|
|
|
- Add `VaultGateway.replaceMarkdownFile(path, expectedContent, nextContent)`.
|
|
- Implement block-end marker replacement in the Obsidian gateway.
|
|
- Add `AnkiGateway.getNoteSummaries(noteIds)` and AnkiConnect implementation.
|
|
|
|
### Tests Missing
|
|
|
|
- Extraction tests for valid, duplicate, and malformed AHS markers.
|
|
- Planning tests for embedded `noteId` precedence and mixed orphan detection.
|
|
- Execution tests for write-back success/failure, recreation, and type-switch rejection.
|
|
- Gateway tests for `getNoteSummaries` and optimistic Markdown replacement. |