diff options
| author | Adam Malczewski <[email protected]> | 2026-06-04 21:21:20 +0900 |
|---|---|---|
| committer | Adam Malczewski <[email protected]> | 2026-06-04 21:21:20 +0900 |
| commit | 394f1ed37ce860da6fdc385769bf29f9737105cd (patch) | |
| tree | 4b825dc642cb6eb9a060e54bf8d69288fbee4904 /packages/core/src/chunks | |
| parent | 81a9cdbadf8c9d940d4fe9a2a0de607dee1f5f1a (diff) | |
| download | dispatch-394f1ed37ce860da6fdc385769bf29f9737105cd.tar.gz dispatch-394f1ed37ce860da6fdc385769bf29f9737105cd.zip | |
chore: genesis — remove all files to rebuild from scratch (arch rewrite)
Diffstat (limited to 'packages/core/src/chunks')
| -rw-r--r-- | packages/core/src/chunks/append.ts | 314 | ||||
| -rw-r--r-- | packages/core/src/chunks/transform.ts | 271 |
2 files changed, 0 insertions, 585 deletions
diff --git a/packages/core/src/chunks/append.ts b/packages/core/src/chunks/append.ts deleted file mode 100644 index 4ca6fe1..0000000 --- a/packages/core/src/chunks/append.ts +++ /dev/null @@ -1,314 +0,0 @@ -/** - * Chunk-builder helper. - * - * `appendEventToChunks` is the single source of truth for how a stream of - * `AgentEvent`s collapses into an ordered `Chunk[]` on a message. Both the - * backend (agent + agent-manager) and the frontend store call this helper - * so the wire format stays in lockstep across the boundary. - * - * Open/close rules — see notes/plan-chunk-refactor.md for the full table. - * - * | Chunk | Opens on | Coalesces | - * |---------------|-----------------------------------------------------------------|------------------------------------------------------------| - * | `text` | first `text-delta` after a non-text chunk | consecutive `text-delta` events append to `.text` | - * | `thinking` | first `reasoning-delta` after a non-thinking chunk | consecutive `reasoning-delta` events append to `.text` | - * | | OR after the last thinking chunk was sealed by `reasoning-end` | (only into the most recent UNSEALED thinking chunk) | - * | `tool-batch` | first `tool-call` after a non-tool-batch chunk | consecutive `tool-call` events push a new entry to `.calls`| - * | `error` | every `error` event | NEVER (always single-event) | - * | `system` | every `notice`/`model-changed`/`config-reload`/... | NEVER (two consecutive system events → two chunks) | - * - * Side-effect events (no new chunk): - * - `tool-result` → finds the call by `id` across all `tool-batch` - * chunks (most-recent first) and updates its - * `result` / `isError`. - * - `shell-output` → appends to the most recent entry of the most - * recent `tool-batch` chunk. - * - `reasoning-end` → attaches `metadata` (the AI SDK v6 - * `providerMetadata` blob) to the most recent - * UNSEALED `thinking` chunk. The metadata is also - * the "sealed" marker — subsequent - * `reasoning-delta`s will open a new chunk rather - * than extending this one. Anthropic's signature - * lives inside this blob; round-tripping it on the - * next turn is mandatory for Anthropic to accept - * the conversation. Orphan `reasoning-end` events - * (no unsealed thinking chunk) are dropped. - * - * Ignored events: - * - `status`, `turn-start`, `turn-sealed`, `done`, `usage`, - * `task-list-update`, `tab-created`, `message-queued`, `message-consumed`, - * `message-cancelled` — these are control / lifecycle events, not message - * content. - */ - -import type { - AgentEvent, - ChatMessage, - Chunk, - MessageRole, - SystemChunk, - SystemChunkKind, - ToolBatchChunk, -} from "../types/index.js"; - -/** - * Mutates `chunks` in place based on `event`. - * - * Returns void; the array is the output channel. - */ -export function appendEventToChunks(chunks: Chunk[], event: AgentEvent): void { - switch (event.type) { - case "text-delta": { - // Open or extend the current text chunk. - const last = chunks[chunks.length - 1]; - if (last && last.type === "text") { - last.text += event.delta; - } else { - chunks.push({ type: "text", text: event.delta }); - } - return; - } - - case "reasoning-delta": { - // Open a new thinking chunk if the last chunk is not a thinking - // chunk OR if it's already sealed by metadata. Anthropic emits - // each thinking content block with its own metadata; a fresh - // reasoning-delta after a sealed thinking chunk is the start of - // a new block, not a continuation — extending the sealed chunk - // would corrupt the metadata/text mapping. - const last = chunks[chunks.length - 1]; - if (last && last.type === "thinking" && last.metadata === undefined) { - last.text += event.delta; - } else { - chunks.push({ type: "thinking", text: event.delta }); - } - return; - } - - case "reasoning-end": { - // Attach `providerMetadata` to the most recent unsealed - // thinking chunk. Anthropic's signature lives inside this - // blob; without it on the next request, Anthropic rejects the - // thinking block. The walk-back is a defensive backstop — - // Anthropic's SSE delivers a content block's deltas strictly - // in order and `appendEventToChunks` runs synchronously per - // event, so the most recent thinking chunk is normally the - // last chunk in the array. - if (event.metadata === undefined) return; - for (let i = chunks.length - 1; i >= 0; i--) { - const c = chunks[i]; - if (!c || c.type !== "thinking") continue; - if (c.metadata !== undefined) { - // Already sealed; the orphan metadata has no home. - return; - } - c.metadata = event.metadata; - return; - } - // No thinking chunk found at all — drop silently. - return; - } - - case "tool-call": { - // Open or extend the current tool-batch chunk. - const last = chunks[chunks.length - 1]; - const entry = { - id: event.toolCall.id, - name: event.toolCall.name, - arguments: event.toolCall.arguments, - }; - if (last && last.type === "tool-batch") { - last.calls.push(entry); - } else { - chunks.push({ type: "tool-batch", calls: [entry] }); - } - return; - } - - case "tool-result": { - // Find the matching call (by id) across all tool-batch chunks, - // most-recent first. Tool results can arrive after subsequent - // text-deltas, so we cannot rely on the *last* chunk being the - // tool-batch — we have to search. - for (let i = chunks.length - 1; i >= 0; i--) { - const c = chunks[i]; - if (!c || c.type !== "tool-batch") continue; - const call = c.calls.find((e) => e.id === event.toolResult.toolCallId); - if (call) { - call.result = event.toolResult.result; - call.isError = event.toolResult.isError; - return; - } - } - // Orphan result with no matching call — drop silently. - return; - } - - case "shell-output": { - // Append to the most recent entry of the most recent tool-batch. - // Walk back through chunks to find the latest tool-batch; if there - // are intervening text/thinking/etc chunks (which can happen if - // the model streams text while a shell tool is still running), - // we still want the most recent tool-batch. - for (let i = chunks.length - 1; i >= 0; i--) { - const c = chunks[i]; - if (!c || c.type !== "tool-batch") continue; - const entry = c.calls[c.calls.length - 1]; - if (!entry) return; - const prev = entry.shellOutput ?? { stdout: "", stderr: "" }; - entry.shellOutput = { - stdout: prev.stdout + (event.stream === "stdout" ? event.data : ""), - stderr: prev.stderr + (event.stream === "stderr" ? event.data : ""), - }; - return; - } - // Orphan shell-output with no tool-batch in scope — drop silently. - return; - } - - case "error": { - // Always a fresh single-event chunk — no coalescing. - chunks.push({ - type: "error", - message: event.error, - ...(event.statusCode !== undefined ? { statusCode: event.statusCode } : {}), - }); - return; - } - - case "notice": { - chunks.push({ type: "system", kind: "notice", text: event.message }); - return; - } - - case "model-changed": { - chunks.push({ - type: "system", - kind: "model-changed", - text: `Switched to ${event.modelId} (${event.keyId})`, - }); - return; - } - - case "config-reload": { - chunks.push({ - type: "system", - kind: "config-reload", - text: "Configuration reloaded", - }); - return; - } - - // Lifecycle / control events — no chunk emitted. - case "status": - case "turn-start": - case "turn-sealed": - case "done": - case "usage": - case "task-list-update": - case "tab-created": - case "message-queued": - case "message-consumed": - case "message-cancelled": - case "compaction-started": - case "compaction-complete": - case "compaction-error": - return; - - default: { - // Exhaustiveness check — if a new event variant is added to - // AgentEvent, TypeScript will complain here. - const _exhaustive: never = event; - void _exhaustive; - return; - } - } -} - -// ─── System event routing across messages ──────────────────────── - -/** - * Minimal shape needed by `applySystemEvent`. - * - * The caller (agent-manager / persistence layer) typically tracks message - * id alongside the wire-format `ChatMessage`. This generic constraint - * lets us keep core `ChatMessage` clean while still letting downstream - * pass anything with an `id`. - */ -export interface IdentifiedMessage { - id: string; - role: MessageRole; - chunks: Chunk[]; -} - -/** - * Describes the system event in caller-controlled terms. We let the caller - * decide both the `kind` (so the same helper can record cancellations, - * notices, model swaps, etc.) and the `text` (so the caller controls - * formatting / localization). - */ -export interface SystemEventLike { - kind: SystemChunkKind; - text: string; -} - -/** - * Routes a system event to the right message when *no assistant turn is - * in flight*. (When a turn IS in flight, the caller should instead use - * `appendEventToChunks` against the in-flight message's chunks directly.) - * - * Routing rules (from notes/plan-chunk-refactor.md): - * - * 1. Most recent message is `role: "system"` → append a `system` chunk - * to it. (Note: a second consecutive system event creates a second - * system chunk inside the same system message — chunks themselves - * never coalesce.) - * 2. Otherwise → create a fresh `role: "system"` message containing one - * `system` chunk and push it. - * - * Returns the `messageId` that was used (either the existing system - * message's id or the newly-created one) so the caller can persist / - * emit a diff to subscribers. - * - * `idFactory` defaults to `crypto.randomUUID()`; tests inject a - * deterministic factory. - */ -export function applySystemEvent<M extends IdentifiedMessage>( - messages: M[], - event: SystemEventLike, - idFactory: () => string = defaultIdFactory, -): { messageId: string } { - const chunk: SystemChunk = { type: "system", kind: event.kind, text: event.text }; - - const last = messages[messages.length - 1]; - if (last && last.role === "system") { - last.chunks.push(chunk); - return { messageId: last.id }; - } - - const id = idFactory(); - // We can't fabricate the full `M` shape without knowing its extra - // fields, but `IdentifiedMessage` is the minimum we need to push. - // Callers that extend the shape with extra fields are responsible for - // initializing them via post-hoc patching, or by passing in their own - // message-creation logic. In practice callers either: - // (a) use `ChatMessage` itself (no extra fields beyond IdentifiedMessage), or - // (b) construct messages and look them up by id after this call returns. - const newMessage = { id, role: "system" as const, chunks: [chunk] } as unknown as M; - messages.push(newMessage); - return { messageId: id }; -} - -function defaultIdFactory(): string { - // In Node 19+ / modern browsers, `crypto.randomUUID` is available globally. - if (typeof crypto !== "undefined" && typeof crypto.randomUUID === "function") { - return crypto.randomUUID(); - } - // Fallback: pseudo-random; not cryptographically secure, but adequate for - // in-memory message identifiers when randomUUID is unavailable. - return `sysmsg-${Date.now().toString(36)}-${Math.random().toString(36).slice(2, 10)}`; -} - -// ─── Re-exports for convenience ────────────────────────────────── - -export type { ChatMessage, Chunk, SystemChunk, SystemChunkKind, ToolBatchChunk }; diff --git a/packages/core/src/chunks/transform.ts b/packages/core/src/chunks/transform.ts deleted file mode 100644 index e8f4a18..0000000 --- a/packages/core/src/chunks/transform.ts +++ /dev/null @@ -1,271 +0,0 @@ -// Pure, dependency-free transforms between the render-shaped `Chunk[]` / -// `ChatMessage` model and the flat append-only `ChunkRow` log. Kept free of any -// DB (`bun:sqlite`) import so BOTH the backend persistence layer -// (`db/chunks.ts`) and the browser frontend store can share the exact same -// explode/group logic. - -import type { - Chunk, - ChunkRow, - ChunkRowDraft, - ErrorData, - MessageRole, - SystemData, - TextData, - ThinkingData, - ToolBatchChunk, - ToolCallData, - ToolResultData, -} from "../types/index.js"; - -/** - * A DERIVED message — a grouping of contiguous chunk rows reconstructed for the - * agent's in-memory history and for the frontend's render bubbles. NOT a stored - * shape: the source of truth is the flat chunk log. - */ -export interface MessageRow { - id: string; - tabId: string; - seq: number; - /** turn_id of the chunk rows this message was grouped from. */ - turnId: string; - role: MessageRole; - chunks: Chunk[]; - createdAt: number; -} - -// ─── Explode: in-memory turn → flat chunk-row drafts ───────────── -// -// A turn's render-shaped `Chunk[]` is flattened into append-only rows. -// `tool-batch` chunks are split into SEPARATE `tool_call` (role=assistant) and -// `tool_result` (role=tool) rows linked by `callId`, mapping 1:1 to the -// Anthropic wire format. `step` increments after each tool-batch: every LLM -// round-trip emits text/thinking then (optionally) one tool-batch, so a -// tool-batch boundary is exactly a step boundary. - -/** Explode a single user message's text into one row draft. */ -export function explodeUserText(turnId: string, text: string): ChunkRowDraft[] { - return [{ turnId, step: 0, role: "user", type: "text", data: { text } }]; -} - -/** Explode an assistant turn's accumulated chunks into ordered row drafts. */ -export function explodeTurn(turnId: string, chunks: Chunk[]): ChunkRowDraft[] { - const drafts: ChunkRowDraft[] = []; - let step = 0; - for (const chunk of chunks) { - switch (chunk.type) { - case "text": - drafts.push({ turnId, step, role: "assistant", type: "text", data: { text: chunk.text } }); - break; - case "thinking": - drafts.push({ - turnId, - step, - role: "assistant", - type: "thinking", - data: { - text: chunk.text, - ...(chunk.metadata !== undefined ? { metadata: chunk.metadata } : {}), - }, - }); - break; - case "tool-batch": { - for (const call of chunk.calls) { - drafts.push({ - turnId, - step, - role: "assistant", - type: "tool_call", - data: { callId: call.id, name: call.name, arguments: call.arguments }, - }); - } - for (const call of chunk.calls) { - if (call.result === undefined) continue; - drafts.push({ - turnId, - step, - role: "tool", - type: "tool_result", - data: { - callId: call.id, - name: call.name, - result: call.result, - isError: call.isError ?? false, - ...(call.shellOutput !== undefined ? { shellOutput: call.shellOutput } : {}), - }, - }); - } - // A tool-batch ends the current LLM step; subsequent chunks belong - // to the next round-trip. - step++; - break; - } - case "error": - drafts.push({ - turnId, - step, - role: "assistant", - type: "error", - data: { - message: chunk.message, - ...(chunk.statusCode !== undefined ? { statusCode: chunk.statusCode } : {}), - }, - }); - break; - case "system": - drafts.push({ - turnId, - step, - role: "system", - type: "system", - data: { kind: chunk.kind, text: chunk.text }, - }); - break; - } - } - return drafts; -} - -// ─── Group: flat chunk rows → derived render messages ──────────── -// -// The inverse of explode (best-effort over an arbitrary window, so it tolerates -// orphan tool-results whose tool-call was paged out). `tool_result` rows -// (role=tool) merge back into the preceding assistant message's per-step -// `tool-batch` chunk by `callId` rather than forming their own message. - -export function groupRowsToMessages(rows: ChunkRow[]): MessageRow[] { - const messages: MessageRow[] = []; - - let current: { msg: MessageRow; batches: Map<number, ToolBatchChunk> } | null = null; - const flush = () => { - if (current) { - messages.push(current.msg); - current = null; - } - }; - const ensureAssistant = (row: ChunkRow) => { - if (!current) { - current = { - msg: { - id: row.id, - tabId: row.tabId, - seq: row.seq, - turnId: row.turnId, - role: "assistant", - chunks: [], - createdAt: row.createdAt, - }, - batches: new Map(), - }; - } - return current; - }; - const ensureBatch = (step: number): ToolBatchChunk => { - const c = current; - if (!c) throw new Error("ensureBatch called without an assistant message"); - let batch = c.batches.get(step); - if (!batch) { - batch = { type: "tool-batch", calls: [] }; - c.batches.set(step, batch); - c.msg.chunks.push(batch); - } - return batch; - }; - - for (const row of rows) { - if (row.role === "user") { - flush(); - const d = row.data as TextData; - messages.push({ - id: row.id, - tabId: row.tabId, - seq: row.seq, - turnId: row.turnId, - role: "user", - chunks: [{ type: "text", text: d.text }], - createdAt: row.createdAt, - }); - continue; - } - if (row.role === "system") { - // Coalesce consecutive system rows into one system message (multiple - // system chunks), matching the old applySystemEvent behaviour. - const prev = messages[messages.length - 1]; - const d = row.data as SystemData; - if (current === null && prev && prev.role === "system") { - prev.chunks.push({ type: "system", kind: d.kind, text: d.text }); - continue; - } - flush(); - messages.push({ - id: row.id, - tabId: row.tabId, - seq: row.seq, - turnId: row.turnId, - role: "system", - chunks: [{ type: "system", kind: d.kind, text: d.text }], - createdAt: row.createdAt, - }); - continue; - } - - // Usage rows are an invisible side channel (persisted for the backend - // aggregate only). They're already query-excluded from getChunksForTab, - // so this is defensive insurance: never let one leak into render grouping. - if (row.type === "usage") continue; - - // assistant / tool rows → part of the current assistant message - const c = ensureAssistant(row); - switch (row.type) { - case "text": - c.msg.chunks.push({ type: "text", text: (row.data as TextData).text }); - break; - case "thinking": { - const d = row.data as ThinkingData; - c.msg.chunks.push({ - type: "thinking", - text: d.text, - ...(d.metadata !== undefined ? { metadata: d.metadata } : {}), - }); - break; - } - case "error": { - const d = row.data as ErrorData; - c.msg.chunks.push({ - type: "error", - message: d.message, - ...(d.statusCode !== undefined ? { statusCode: d.statusCode } : {}), - }); - break; - } - case "tool_call": { - const d = row.data as ToolCallData; - ensureBatch(row.step).calls.push({ id: d.callId, name: d.name, arguments: d.arguments }); - break; - } - case "tool_result": { - const d = row.data as ToolResultData; - const batch = ensureBatch(row.step); - const entry = batch.calls.find((e) => e.id === d.callId); - if (entry) { - entry.result = d.result; - entry.isError = d.isError; - if (d.shellOutput !== undefined) entry.shellOutput = d.shellOutput; - } else { - // Orphan result (its tool_call was paged out of this window). - batch.calls.push({ - id: d.callId, - name: d.name, - arguments: {}, - result: d.result, - isError: d.isError, - ...(d.shellOutput !== undefined ? { shellOutput: d.shellOutput } : {}), - }); - } - break; - } - } - } - flush(); - return messages; -} |
