fix(tool-fs): CRLF-safe write diff, opaque meta, doc sync

Address the applied-hunk-diffs review:

- CRLF write overwrite emitted bogus every-line-changed hunks: write's
  `before` was LF-normalized but `after` kept the raw model content, so a
  CRLF rewrite of an LF file diffed every line. Normalize write's `after`
  to LF so both sides share the diff basis (edit already did). Regression
  test proves it fails on the raw-after path.
- The tool-private `meta` payload is now typed `unknown` (opaque) at every
  seam instead of `JsonValue`. This drops the `dsh-tools -> dsh-session`
  package edge that existed only to name the type, and removes the
  `FileDiff` index signature that had been widening the type solely for
  JsonValue-assignability. Serializability is still enforced at runtime by
  `Session.append`'s isJsonValue check, which was always the real guard.
- Sync the docs the new result/meta surface left stale: ToolResultView's
  diff card + ToolExecutionResult.meta in tools.md/session.md type-equiv
  blocks, the acp/tools READMEs, and the adding-a-tool cookbook; regenerate
  the cordis catalog and module graph.
This commit is contained in:
Tianyi Cui
2026-07-03 18:00:16 +08:00
parent d8fd3225af
commit dee2dee402
18 changed files with 63 additions and 68 deletions

View File

@@ -1,6 +1,5 @@
import type { Branded } from '@deepseek-ai/dsh-brand'
import type { CallId, ContentBlock, MessageSource, StreamChunk, TokenUsage } from '@deepseek-ai/dsh-llm'
import type { JsonValue } from './json.ts'
/** Identifies one session in the store (and its persistence artifacts). */
export type SessionId = Branded<'SessionId'>
@@ -213,14 +212,14 @@ export interface SessionEventMap {
'tool/call': { turn: number; step: number; callId: CallId; name: string; arguments: string }
/**
* A completed tool call's model-facing result, plus an optional tool-private
* `meta` presentation payload. `meta` is opaque to the core — the producing
* tool owns its shape and reads it back in `presentResult` — and is a
* {@link JsonValue} so it persists in the durable log and reproduces on replay
* (a UI bridge renders the identical card from a loaded session). Absent unless
* the tool attaches one (e.g. `dsh-tool-fs` carries its result-time contextual
* diff here).
* `meta` presentation payload. `meta` is opaque to the core (`unknown` — the
* producing tool owns its shape and reads it back in `presentResult`) but MUST
* be JSON-serializable: `Session.append` runtime-validates all event data with
* `isJsonValue`, so a non-serializable `meta` is rejected at the source, and the
* durable log reproduces the identical card on replay. Absent unless the tool
* attaches one (e.g. `dsh-tool-fs` carries its result-time contextual diff here).
*/
'tool/result': { turn: number; step: number; callId: CallId; content: ContentBlock[]; isError: boolean; error?: { name: string; code: string }; meta?: JsonValue }
'tool/result': { turn: number; step: number; callId: CallId; content: ContentBlock[]; isError: boolean; error?: { name: string; code: string }; meta?: unknown }
/** Steering content injected between steps of a running turn. */
'steering/message': { turn: number; content: ContentBlock[]; source: MessageSource }
/**

View File

@@ -26,7 +26,7 @@ Tool registry and execution waterfall. Tool plugins register their schemas and e
- `ToolDefinition``ToolSchema` + `execute(args, exec): Promise<ContentBlock[] | { content: ContentBlock[]; meta? }>` (the bare array is the model-facing content; the object form additionally attaches an opaque, JSON-serializable `meta` presentation payload persisted on the `tool/result` event and handed back to `presentResult`), plus optional `presentCall(args)` / `presentResult(args, result)` for tool-owned UI presentation (see below).
- `ToolExecution` — one pending tool call: `{ callId, name, arguments, agent?, signal? }`.
- `ToolExecutionResult` — outcome: `{ callId, content, isError, error? }`. On failure with a `HarnessError`, `error: { name, code }` carries the structured failure class alongside the model-facing text (the loop forwards it onto the `tool/result` session event for retry/sandbox plugins and replay).
- `ToolExecutionResult` — outcome: `{ callId, content, isError, error?, meta? }`. On failure with a `HarnessError`, `error: { name, code }` carries the structured failure class alongside the model-facing text. `meta` is the tool's opaque presentation payload from a successful `execute` (the object return form); the loop forwards both `error` and `meta` onto the `tool/result` session event (for retry/sandbox plugins, replay, and result-card rendering).
- `ToolCallView` / `ToolResultView` — provider-neutral `card`-tagged render intents a tool returns from `presentCall` / `presentResult` to own how a UI renders ITS calls (see "Tool-owned UI presentation").
### Extension points
@@ -81,7 +81,7 @@ A tool owns how ITS calls render in a UI (an editor's tool-call card, a CLI log
- `{ card: 'terminal', title?, output?, exitCode?, signal? }` — a terminal run's captured `output` and exit status. A capable UI shows an exit-status pill; an incapable UI gets a fenced ` ```console ` fallback the BRIDGE derives from `output` (the tool does not encode the fences).
- `{ card: 'diff', title?, diffs }` — a completed file mutation as an inline diff. `diffs` is `FileDiff[]` — the APPLIED hunks with surrounding context (one entry per changed site), computed from the before/after file content, distinct from the call-time whole-snippet `diff`. Used by `write`/`edit`; a `tool_call_update.content` replaces the call's content, so this supersedes the pending snippet.
Returning `undefined` (or omitting a method) tells a UI to fall back to a generic presentation (title = tool name, raw args as input, raw result content). Both methods must be **pure and side-effect-free**: a UI may call them during live streaming AND during a session-log replay, so they depend only on their arguments. `result.meta` is the tool's own optional presentation payload (`JsonValue`), attached by `execute` (see below) and persisted on the `tool/result` event, so a `presentResult` reading it stays replay-deterministic (the same `meta` is read back from the log). With `defineTool`, `args` is the typed `InferArgs<S>` shape; the helper soft-validates before calling (a malformed/older logged arg shape yields `undefined` rather than throwing, since display must never crash a replay). The views are provider-neutral — the ACP bridge (`dsh-acp`) maps each `card` to ACP `tool_call`/`tool_call_update` wire fields (a `diff` card to a `{ type: 'diff' }` content block, a `terminal` card to the `_meta` terminal convention), and relativizes a file card's title against the session cwd. See the render-intent-union RFC (`docs/rfc/implemented/architecture/2026-07-02-tool-render-intent-union.md`) and the applied-hunk-diffs RFC (`docs/rfc/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.md`); `dsh-tool-bash` (terminal) and `dsh-tool-fs` (diff/generic) are the reference implementations.
Returning `undefined` (or omitting a method) tells a UI to fall back to a generic presentation (title = tool name, raw args as input, raw result content). Both methods must be **pure and side-effect-free**: a UI may call them during live streaming AND during a session-log replay, so they depend only on their arguments. `result.meta` is the tool's own optional presentation payload (opaque `unknown`, JSON-serializable), attached by `execute` (see below) and persisted on the `tool/result` event, so a `presentResult` reading it stays replay-deterministic (the same `meta` is read back from the log). With `defineTool`, `args` is the typed `InferArgs<S>` shape; the helper soft-validates before calling (a malformed/older logged arg shape yields `undefined` rather than throwing, since display must never crash a replay). The views are provider-neutral — the ACP bridge (`dsh-acp`) maps each `card` to ACP `tool_call`/`tool_call_update` wire fields (a `diff` card to a `{ type: 'diff' }` content block, a `terminal` card to the `_meta` terminal convention), and relativizes a file card's title against the session cwd. See the render-intent-union RFC (`docs/rfc/implemented/architecture/2026-07-02-tool-render-intent-union.md`) and the applied-hunk-diffs RFC (`docs/rfc/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.md`); `dsh-tool-bash` (terminal) and `dsh-tool-fs` (diff/generic) are the reference implementations.
```ts
import { defineTool } from '@deepseek-ai/dsh-tools'

View File

@@ -24,14 +24,12 @@
"peerDependencies": {
"@deepseek-ai/dsh-agent": "^0.0.1",
"@deepseek-ai/dsh-llm": "^0.0.1",
"@deepseek-ai/dsh-session": "^0.0.1",
"@deepseek-ai/dsh-system-prompt": "^0.0.1",
"cordis": "^4.0.0-rc.6"
},
"devDependencies": {
"@deepseek-ai/dsh-agent": "workspace:^",
"@deepseek-ai/dsh-llm": "workspace:^",
"@deepseek-ai/dsh-session": "workspace:^",
"@deepseek-ai/dsh-system-prompt": "workspace:^",
"cordis": "^4.0.0-rc.6"
}

View File

@@ -11,7 +11,6 @@ import { Context, Service } from 'cordis'
import type { CallId, ContentBlock, ToolSchema } from '@deepseek-ai/dsh-llm'
import { HarnessError } from '@deepseek-ai/dsh-llm'
import type { Agent } from '@deepseek-ai/dsh-agent'
import type { JsonValue } from '@deepseek-ai/dsh-session'
import type {} from '@deepseek-ai/dsh-system-prompt'
export {
@@ -88,13 +87,6 @@ export interface FileDiff {
oldText: string | null
/** Content after the change. */
newText: string
/**
* Index signature so a `FileDiff` is a valid {@link JsonValue} member — a tool
* persists result-time diffs as `tool/result` `meta`, which must round-trip
* through the session log. Every declared field is already JSON-compatible;
* this only makes the structural compatibility explicit.
*/
[key: string]: string | null
}
/**
@@ -247,12 +239,13 @@ export interface DiffResultView {
/**
* What a tool's `execute` returns. The bare {@link ContentBlock}`[]` form is the
* common case (model-facing content only); the object form additionally attaches
* a tool-private `meta` presentation payload ({@link JsonValue}) that the
* registry threads onto the `tool/result` session event and hands back to the
* tool's `presentResult`. `meta` is opaque to the core — the tool owns its shape
* and validates it on the way out — and persists so replay reproduces the card.
* a tool-private `meta` presentation payload that the registry threads onto the
* `tool/result` session event and hands back to the tool's `presentResult`.
* `meta` is opaque to the core (`unknown` — the tool owns and narrows its shape),
* and MUST be JSON-serializable: it persists on the durable log (the session
* enforces this at `append`), so replay reproduces the card.
*/
export type ToolExecuteReturn = ContentBlock[] | { content: ContentBlock[]; meta?: JsonValue }
export type ToolExecuteReturn = ContentBlock[] | { content: ContentBlock[]; meta?: unknown }
/** A registered tool: its schema plus the execution function. */
export interface ToolDefinition extends ToolSchema {
@@ -286,10 +279,10 @@ export interface ToolResult {
/**
* The tool-private presentation payload the tool attached from `execute` (via
* the object return form), threaded verbatim from the `tool/result` event.
* Opaque {@link JsonValue}; the tool narrows it back to its own shape. Absent
* when the tool attached none.
* Opaque (`unknown`); the tool narrows it back to its own shape. Absent when
* the tool attached none.
*/
meta?: JsonValue
meta?: unknown
}
/** One pending tool call, as it flows through the execution waterfall. */
@@ -336,10 +329,10 @@ export interface ToolExecutionResult {
/**
* The tool-private presentation payload from a successful `execute` (the object
* return form). Threaded onto the `tool/result` session event and back into
* {@link ToolResult} for `presentResult`. Opaque {@link JsonValue}; absent when
* the tool attached none or the call failed.
* {@link ToolResult} for `presentResult`. Opaque (`unknown`); absent when the
* tool attached none or the call failed.
*/
meta?: JsonValue
meta?: unknown
}
/**