Merge master into codex/jsonl-storage-identity

packages/session-persistence/session-persistence-jsonl/README.md

@@ -11,7 +11,8 @@ The JSONL durable session-persistence backend — a concrete `SessionPersistence
     <encoded-id>.jsonl           # only with compression: 'none'
 ```
 
-- The first logical line is the immutable `SessionHeader` tagged `{ type: 'session', version, id, cwd?, createdAt, parentSession?, seedLength?, delegationDepth }`. `delegationDepth` is required on disk and is `0` for a top-level session; a missing or invalid value rejects the log. Every subsequent line is one `SessionEvent` JSON, **verbatim including `assistant/chunk`** so `seq` stays contiguous (`events[i].seq === i`).
+- The first logical line is the immutable `SessionHeader` tagged `{ type: 'session', version, id, cwd?, createdAt, parentSession?, seedLength?, delegationDepth }`. `delegationDepth` is required on disk and is `0` for a top-level session; a missing or invalid value rejects the log. Every subsequent logical line is one storage record; `assistant/chunk` events are never dropped, and `seq` stays contiguous across the decoded log (`events[i].seq === i`).
+- A storage record is a `SessionEvent` JSON verbatim, or — written only under `packChunks` — a **packed chunk row** (`text-chunks` / `reasoning-chunks` / `tool-call-chunks`; bare slash-less tags like the header's `session`, so row tags cannot be confused with event types): one line holding a run of ≥3 consecutive same-block `assistant/chunk` delta events, `seq0`/`time0` plus per-member `dt` gaps reconstructing every member's `seq`/`time` exactly. The lossless codec lives in `@deepseek-ai/dsh-session` (`packChunkRuns`/`decodeStorageRecord`) and whitelists exact shapes — anything unrecognized stores verbatim. Reading is layout-blind: `load` always decodes rows, so packed, unpacked, and mixed files load identically.
 - Session ids are unvalidated branded strings, so they are injectively escaped to a single safe path segment before use (no traversal, no collision).
 
 ## Config
@@ -19,6 +20,7 @@ The JSONL durable session-persistence backend — a concrete `SessionPersistence
 | Key | Type | Notes |
 |---|---|---|
 | `root` | `string` (required) | Root directory for all session files. **No default** — a `process.cwd()` default would scatter files as the process's cwd changes (bash calls, subprocesses). An existing root must be a readable directory; an absent root is created on first materialization. |
+| `packChunks` | `boolean` (default `false`) | Write delta-chunk runs as packed rows (~60% smaller logical logs measured on a real coding session). Off, the written logical layout is byte-identical to the pre-packing format; reading packed rows works regardless of this switch. Off by default while the snapshot goldens stay one-event-per-line — recording with packing on rewrites every fixture `session.jsonl`. |
 | `compression` | `'zstd' \| 'none'` | Defaults to `'zstd'`; `'none'` retains newline-delimited UTF-8 text. |
 
 `locate(meta)` returns `{ kind: 'jsonl', path }` using the resolved absolute root and the same cwd-bucket/id encoding as materialization. It performs no filesystem I/O: the target can be returned before the file exists, and an existing file contains only the last flushed prefix.
@@ -33,7 +35,7 @@ A root belongs to one encoding. Startup discovery and targeted lookup reject the
 
 - **Bound storage identity.** Lookup requires one matching encoded filename across the cwd buckets, then verifies that the header id equals the requested id and that the header's id/cwd derive the selected path. Listing applies the same path check and rejects duplicate ids. Identity failures occur before repair or append.
 - **Lazy materialization.** `create(meta)` writes nothing; on the first `append`, the backend writes and `fsync`s the encoded header and first batch in a temporary file. POSIX publishes it without overwrite via a hard link and `fsync`s the parent directory. Windows publishes it without overwrite via `MoveFileExW(..., MOVEFILE_WRITE_THROUGH)` and creates missing directories through the same write-through pattern. A created-but-never-appended session leaves nothing on disk and is absent from `list`.
-- **Append-only.** Committed events (at or below a flushed `turn/end`) are never rewritten. Subsequent raw batches append lines; compressed batches append one frame. Both paths `fsync`, and a caught write or sync failure rolls the file back to its prior byte length.
+- **Append-only.** Flushed events are never rewritten. Subsequent raw batches append lines; compressed batches append one frame. Both paths `fsync`, and a caught write or sync failure rolls the file back to its prior byte length.
 - **Crash recovery — preserve valid tail work.** `load` validates every complete compressed frame and scans their decompressed JSONL. If the last frame is structurally incomplete, the reader keeps its complete decoded records, truncates from that frame's start, and re-encodes those records with the synthetic tool, step, and turn closers required by the shared [persistence contract](../../../.agents/notes/implemented/architecture/2026-06-14-session-persistence.md). Raw mode truncates from its first incomplete line. A checksum/decompression failure in a complete frame, or a defect at or before the last committed `turn/end`, is corruption and rejects.
 - **Contiguous-seq.** `append` rejects a batch whose first `seq` does not continue the stored log, and rejects non-JSON-serializable `event.data` naming the offending event type.
 
@@ -47,7 +49,7 @@ The plugin buffers frozen session events and drains them on flush or disposal. A
 
 #### What the model sees
 
-JSONL storage contributes no live prompt or schema. Loading restores stored surface history and preserves prior request headers for reconstruction; the new loop composes its current envelope. Each unanswered call in an interrupted tail is balanced with the exact error text `Tool call interrupted by a crash; no result was recorded.` Raw `assistant/chunk` records do not duplicate messages.
+JSONL storage contributes no live prompt or schema. Loading restores stored surface history and preserves prior request headers for reconstruction; the new loop composes its current envelope. Recovery balances an assistant request without a durable call with `TOOL_NOT_STARTED`; a durable call without a result becomes `TOOL_OUTCOME_UNKNOWN`, which tells the model to retry only read-only or idempotent work and to verify possible side effects or ask the user. Raw `assistant/chunk` records do not duplicate messages.
 
 #### Token effect
 

packages/session-persistence/session-persistence-jsonl/src/format.ts

@@ -10,7 +10,8 @@
 
 import { createHash } from 'node:crypto'
 import { join } from 'node:path'
-import type { SessionEvent, SessionHeader, SessionId } from '@deepseek-ai/dsh-session'
+import { decodeStorageRecord, packChunkRuns } from '@deepseek-ai/dsh-session'
+import type { SessionEvent, SessionHeader, SessionId, StorageRecord } from '@deepseek-ai/dsh-session'
 
 /** Physical encoding selected for JSONL session artifacts. */
 export type JsonlCompression = 'zstd' | 'none'
@@ -151,17 +152,26 @@ export function logPath(
 }
 
 /**
- * Serialize one event as a JSONL line (no trailing newline).
- * @param event - the event to serialize verbatim.
- * @returns the event's single-line JSON text; the writer adds the newline.
+ * Serialize an event batch as JSONL lines (no trailing newline). With
+ * `packChunks` on, delta-chunk runs pack into `text-chunks` /
+ * `reasoning-chunks` / `tool-call-chunks` storage rows; off writes one event
+ * per line, byte-identical to the pre-packing layout. Reading is layout-blind
+ * either way ({@link scanLog} always decodes rows), so the switch only shapes
+ * NEW bytes.
+ * @param events - the batch to serialize, in log order.
+ * @param packChunks - whether to pack delta runs into storage rows.
+ * @returns the batch's JSONL text; the writer adds the final newline.
  */
-export function eventLine(event: SessionEvent): string {
-  return JSON.stringify(event)
+export function eventLines(events: readonly SessionEvent[], packChunks: boolean): string {
+  const records: readonly StorageRecord[] = packChunks ? packChunkRuns(events) : events
+  return records.map(record => JSON.stringify(record)).join('\n')
 }
 
 /**
  * Parse a JSONL log buffer into its preserved event prefix (the header is line
- * 0). Fully written events in an interrupted final turn remain part of the
+ * 0). Event lines pass through verbatim; packed chunk rows expand back into
+ * their events, so callers see one contiguous event list regardless of layout.
+ * Fully written events in an interrupted final turn remain part of the
  * prefix. The first unparsable record or seq gap after the last `turn/end`
  * marks a tolerated torn tail; the same hole in the committed region rejects.
  *
@@ -200,46 +210,60 @@ export function scanLog(buffer: Buffer): { meta: SessionHeader; events: SessionE
   }
   const headerLine = parsedHeader
 
-  // Parse every complete record first so the last valid `turn/end` determines
-  // whether an earlier hole is committed corruption or an uncommitted tail.
-  interface Parsed { ok: boolean; event?: SessionEvent; endByte: number }
+  // Parse and decode every complete line first so the last valid `turn/end`
+  // determines whether an earlier hole is committed corruption or an
+  // uncommitted tail. One line yields one event, or a whole run for a packed
+  // chunk row; a row-tagged line that fails row validation is a hole, exactly
+  // like unparsable JSON.
+  interface Parsed { ok: boolean; events?: SessionEvent[]; endByte: number }
   const parsed: Parsed[] = eventEntries.map((entry) => {
     try {
-      return { ok: true, event: JSON.parse(entry.text) as SessionEvent, endByte: entry.endByte }
+      return { ok: true, events: decodeStorageRecord(JSON.parse(entry.text)), endByte: entry.endByte }
     } catch {
       return { ok: false, endByte: entry.endByte }
     }
   })
 
-  // The last index (into eventEntries) that is a valid `turn/end` — the last
-  // fully-committed boundary (the loop flushes only at turn/end).
+  // The last index (into eventEntries) that ends in a valid `turn/end` — the
+  // last fully-committed boundary (the loop flushes only at turn/end). A packed
+  // row never stores a turn/end, so only single-event lines can match.
   let lastTurnEnd = -1
   for (let i = parsed.length - 1; i >= 0; i--) {
     const p = parsed[i]
-    if (p?.ok && p.event?.type === 'turn/end') { lastTurnEnd = i; break }
+    if (p?.ok && p.events?.some(e => e.type === 'turn/end')) { lastTurnEnd = i; break }
   }
 
   // Preserve the contiguous prefix, including a complete interrupted turn;
   // holes through the last committed boundary throw, while later holes stop.
+  // Contiguity is a cursor over seqs (not the line index): a packed row
+  // advances the cursor by its whole run.
   const preserved: SessionEvent[] = []
-  for (let i = 0; i < parsed.length; i++) {
+  let lastPreservedLine = -1
+  scan: for (let i = 0; i < parsed.length; i++) {
     const p = parsed[i]
-    if (!p?.ok || p.event === undefined) {
+    if (!p?.ok || p.events === undefined) {
       if (i <= lastTurnEnd) throw new Error(`corrupt session log: unparsable committed event at line ${i + 1}`)
       break // torn tail fragment after the last turn/end — stop, tolerate
     }
-    if (p.event.seq !== i) {
-      if (i <= lastTurnEnd) throw new Error(`corrupt session log: seq gap in committed region at line ${i + 1} (expected ${i}, got ${p.event.seq})`)
-      break // gap after the last turn/end — torn tail, stop
+    for (const event of p.events) {
+      if (event.seq !== preserved.length) {
+        if (i <= lastTurnEnd) {
+          throw new Error(`corrupt session log: seq gap in committed region at line ${i + 1} (expected ${preserved.length}, got ${event.seq})`)
+        }
+        break scan // gap after the last turn/end — torn tail, stop
+      }
+      preserved.push(event)
     }
-    preserved.push(p.event)
+    lastPreservedLine = i
   }
 
-  // committedBytes = end of the last PRESERVED line (header if none): the next
-  // append truncates any torn bytes past this point before writing the
-  // synthetic closers + new events.
-  const lastPreserved = parsed[preserved.length - 1]
-  const committedBytes = preserved.length > 0 && lastPreserved ? lastPreserved.endByte : headerEntry.endByte
+  // committedBytes = end of the last FULLY preserved line (header if none): the
+  // next append truncates any torn bytes past this point before writing the
+  // synthetic closers + new events. A line is preserved whole or not at all —
+  // a mid-row seq gap discards the whole row, keeping the truncation offset on
+  // a line boundary.
+  const lastPreserved = parsed[lastPreservedLine]
+  const committedBytes = lastPreserved !== undefined ? lastPreserved.endByte : headerEntry.endByte
   return { meta: fromHeaderLine(headerLine), events: preserved, committedBytes }
 }
 

packages/session-persistence/session-persistence-jsonl/src/index.ts

@@ -18,7 +18,7 @@ import {
 } from '@deepseek-ai/dsh-session-persistence'
 import type { SessionEvent, SessionId, SessionHeader } from '@deepseek-ai/dsh-session'
 import {
-  encodeSegment, eventLine, logPath, logSuffix, parseHeaderMeta, scanLog, sessionDir, toHeaderLine,
+  encodeSegment, eventLines, logPath, logSuffix, parseHeaderMeta, scanLog, sessionDir, toHeaderLine,
   type JsonlCompression,
 } from './format.ts'
 import { compressZstdFrame, decompressZstdFrame, scanZstdFrames } from './zstd.ts'
@@ -34,7 +34,7 @@ export const JsonlCompressionSchema: z<JsonlCompression> = z.union([
   z.const('none'),
 ]).default(DEFAULT_COMPRESSION)
 
-/** Plugin config: where the JSONL backend keeps its session logs (`root` is required — no default). */
+/** Plugin config: where the JSONL backend keeps its session logs, and the packed-row write switch. */
 export interface Config {
   /**
    * Root directory for all session files. Required (no default): a default of
@@ -44,6 +44,15 @@ export interface Config {
    * first materialization.
    */
   root: string
+  /**
+   * Write runs of consecutive `assistant/chunk` delta events as packed
+   * `text-chunks`/`reasoning-chunks`/`tool-call-chunks` rows (lossless,
+   * ~60% smaller logs measured on a real session). Off by default while
+   * snapshot fixtures stay in the one-event-per-line layout: recording with
+   * packing on rewrites every golden `session.jsonl`. READING packed rows is
+   * unconditional — a log's layout never depends on this switch.
+   */
+  packChunks?: boolean
   /** Physical encoding; defaults to checksummed Zstandard frames. */
   compression?: JsonlCompression
 }
@@ -70,6 +79,7 @@ export class SessionPersistenceJsonl extends SessionPersistence implements Persi
 
   static Config: z<Config> = z.object({
     root: z.string().required(),
+    packChunks: z.boolean().default(false),
     compression: JsonlCompressionSchema,
   })
 
@@ -81,6 +91,7 @@ export class SessionPersistenceJsonl extends SessionPersistence implements Persi
   override readonly name = 'session-persistence-jsonl'
 
   private root: string
+  private packChunks: boolean
   private compression: JsonlCompression
   private coordinator: PersistenceCoordinator<JsonlTornMarker>
   private rootEncodingCheck: Promise<void> | undefined
@@ -89,6 +100,9 @@ export class SessionPersistenceJsonl extends SessionPersistence implements Persi
     super(ctx)
     // Resolve once so later process.cwd() changes cannot split one backend across roots.
     this.root = resolve(config.root)
+    // schemastery (static Config) applied the default before construction;
+    // the cast records that runtime fact for exactOptionalPropertyTypes.
+    this.packChunks = (config as Required<Config>).packChunks
     this.compression = config.compression ?? DEFAULT_COMPRESSION
     this.assertUsableRoot()
     this.coordinator = new PersistenceCoordinator<JsonlTornMarker>(this.ctx, this)
@@ -357,7 +371,7 @@ export class SessionPersistenceJsonl extends SessionPersistence implements Persi
   /** Encode the header and first batch without combining their frame boundaries. */
   private async encodeMaterialization(meta: SessionHeader, events: readonly SessionEvent[]): Promise<Buffer | string> {
     const header = JSON.stringify(toHeaderLine(meta)) + '\n'
-    const body = events.map(eventLine).join('\n') + '\n'
+    const body = eventLines(events, this.packChunks) + '\n'
     if (this.compression === 'none') return header + body
     const headerFrame = await compressZstdFrame(header)
     const eventFrame = await compressZstdFrame(body)
@@ -366,7 +380,7 @@ export class SessionPersistenceJsonl extends SessionPersistence implements Persi
 
   /** Encode one durable append batch in the configured physical representation. */
   private async encodeEventBatch(events: readonly SessionEvent[]): Promise<Buffer | string> {
-    const body = events.map(eventLine).join('\n') + '\n'
+    const body = eventLines(events, this.packChunks) + '\n'
     return this.compression === 'zstd' ? compressZstdFrame(body) : body
   }
 

packages/session-persistence/session-persistence-jsonl/tests/jsonl.spec.ts

@@ -6,7 +6,7 @@ import { isAbsolute, join, relative, resolve } from 'node:path'
 import SessionStore, { SessionId } from '@deepseek-ai/dsh-session'
 import type { Session, SessionEvent, SessionHeader } from '@deepseek-ai/dsh-session'
 import SessionPersistenceJsonl from '@deepseek-ai/dsh-session-persistence-jsonl'
-import { encodeSegment, logPath, scanLog, sessionDir, toHeaderLine } from '../src/format.ts'
+import { encodeSegment, eventLines, logPath, scanLog, sessionDir, toHeaderLine } from '../src/format.ts'
 import { runPersistenceContract, meta, oneTurnLog, appendLog } from '../../session-persistence/tests/contract.ts'
 import { runCoordinatorContract, type CoordinatorFixture } from '../../session-persistence/tests/coordinator-contract.ts'
 
@@ -588,6 +588,121 @@ describe('SessionPersistenceJsonl: scanLog unit', () => {
   })
 })
 
+describe('SessionPersistenceJsonl: packed chunk rows (packChunks: true)', () => {
+  let ctx: Context
+  beforeEach(async () => {
+    root = await freshRoot()
+    ctx = new Context()
+    await ctx.plugin(SessionStore)
+    // compression: 'none' — these tests assert the textual storage-record layout
+    // (row tags per line); packing is orthogonal to the physical encoding.
+    await ctx.plugin(SessionPersistenceJsonl, { root, packChunks: true, compression: 'none' })
+  })
+  afterEach(async () => { await ctx.fiber.dispose() })
+
+  /** A one-turn log whose step streams a five-member text-delta run. */
+  function chunkRunLog(): SessionEvent[] {
+    const deltas: SessionEvent[] = Array.from({ length: 5 }, (_, k) => ({
+      type: 'assistant/chunk',
+      seq: 2 + k,
+      time: 3 + k,
+      data: { turn: 1, step: 1, chunk: { type: 'text-delta', index: 0, text: `t${k}` } },
+    }))
+    return [
+      { type: 'turn/start', seq: 0, time: 1, data: { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } } },
+      { type: 'step/start', seq: 1, time: 2, data: { turn: 1, step: 1 } },
+      ...deltas,
+      { type: 'assistant/message', seq: 7, time: 8, data: { turn: 1, step: 1, content: [{ type: 'text', text: 't0t1t2t3t4' }], provenance: { provider: 'mock', model: 'mock' } }, surfaceOp: 'append', sourceEventSeqs: [2, 3, 4, 5, 6] },
+      { type: 'step/end', seq: 8, time: 9, data: { turn: 1, step: 1 } },
+      { type: 'turn/end', seq: 9, time: 10, data: { turn: 1, reason: { kind: 'completed' } } },
+    ]
+  }
+
+  it('writes a delta run as one text-chunks row and loads back identical events', async () => {
+    const m = meta('packed', '/work')
+    const log = chunkRunLog()
+    await ctx.sessionPersistence.create(m)
+    await ctx.sessionPersistence.append(m.id, log)
+
+    const raw = (await readFile(rawLogPath(root, '/work', m.id), 'utf8')).split('\n').filter(Boolean)
+    const tags = raw.slice(1).map(line => (JSON.parse(line) as { type: string }).type)
+    expect(tags).toEqual(['turn/start', 'step/start', 'text-chunks', 'assistant/message', 'step/end', 'turn/end'])
+
+    const loaded = await ctx.sessionPersistence.load(m.id)
+    expect(loaded.events).toEqual(log)
+  })
+
+  it('loads a mixed file: verbatim lines from an unpacked writer, then packed appends', async () => {
+    const m = meta('mixed', '/work')
+    const log = chunkRunLog()
+    // First turn written line-per-event by an unpacked-config writer (an old
+    // file, hand-planted so this packed-config backend adopts it on load).
+    await mkdir(sessionDir(root, '/work'), { recursive: true })
+    await writeFile(rawLogPath(root, '/work', m.id), [
+      JSON.stringify({ type: 'session', version: 0, id: 'mixed', createdAt: 1000, cwd: '/work', delegationDepth: 0 }),
+      ...log.map(e => JSON.stringify(e)),
+    ].join('\n') + '\n')
+    // Adopt the stored log (cursor = stored length), then append a second turn
+    // through THIS packed-config backend.
+    expect((await ctx.sessionPersistence.load(m.id)).events).toEqual(log)
+    const secondTurn: SessionEvent[] = JSON.parse(JSON.stringify(log)) as SessionEvent[]
+    for (const [k, e] of secondTurn.entries()) {
+      ;(e as { seq: number }).seq = 10 + k
+      ;(e.data as { turn: number }).turn = 2
+    }
+    await ctx.sessionPersistence.append(m.id, secondTurn)
+
+    const loaded = await ctx.sessionPersistence.load(m.id)
+    expect(loaded.events).toEqual([...log, ...secondTurn])
+    // The packed append really packed: the file's tail carries a text-chunks row.
+    const tags = (await readFile(rawLogPath(root, '/work', m.id), 'utf8')).split('\n').filter(Boolean)
+      .map(line => (JSON.parse(line) as { type: string }).type)
+    expect(tags.filter(t => t === 'text-chunks')).toHaveLength(1)
+    expect(tags.filter(t => t === 'assistant/chunk')).toHaveLength(5)
+  })
+
+  it('scanLog: a packed row advances the seq cursor by its whole run', () => {
+    const logText = [
+      JSON.stringify({ type: 'session', version: 0, id: 'rows', createdAt: 1, delegationDepth: 0 }),
+      JSON.stringify({ type: 'turn/start', seq: 0, time: 1, data: { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } } }),
+      JSON.stringify({ type: 'text-chunks', seq0: 1, time0: 2, data: { turn: 1, step: 1, index: 0, dt: [1, 1], texts: ['a', 'b', 'c'] } }),
+      JSON.stringify({ type: 'turn/end', seq: 4, time: 5, data: { turn: 1, reason: { kind: 'completed' } } }),
+    ].join('\n') + '\n'
+    const { events } = scanLog(Buffer.from(logText))
+    expect(events.map(e => e.seq)).toEqual([0, 1, 2, 3, 4])
+    expect(events[2]).toEqual({ type: 'assistant/chunk', seq: 2, time: 3, data: { turn: 1, step: 1, chunk: { type: 'text-delta', index: 0, text: 'b' } } })
+  })
+
+  it('scanLog: a malformed packed row in the committed region rejects like corrupt JSON', () => {
+    const logText = [
+      JSON.stringify({ type: 'session', version: 0, id: 'bad-row', createdAt: 1, delegationDepth: 0 }),
+      // dt arity mismatch — row validation throws, so the line is a committed hole.
+      JSON.stringify({ type: 'text-chunks', seq0: 0, time0: 1, data: { turn: 1, step: 1, index: 0, dt: [], texts: ['a', 'b'] } }),
+      JSON.stringify({ type: 'turn/end', seq: 2, time: 3, data: { turn: 1, reason: { kind: 'completed' } } }),
+    ].join('\n') + '\n'
+    expect(() => scanLog(Buffer.from(logText))).toThrow(/unparsable committed event/)
+  })
+
+  it('scanLog: a packed row with a mid-run seq gap after the last turn/end drops the whole row', () => {
+    const logText = [
+      JSON.stringify({ type: 'session', version: 0, id: 'row-gap', createdAt: 1, delegationDepth: 0 }),
+      JSON.stringify({ type: 'turn/start', seq: 0, time: 1, data: { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } } }),
+      // seq0 skips 1 — the run's first member is already a gap; no turn/end follows.
+      JSON.stringify({ type: 'text-chunks', seq0: 2, time0: 2, data: { turn: 1, step: 1, index: 0, dt: [1, 1], texts: ['a', 'b', 'c'] } }),
+    ].join('\n') + '\n'
+    const scanned = scanLog(Buffer.from(logText))
+    expect(scanned.events.map(e => e.seq)).toEqual([0])
+    // committedBytes stays on the line boundary BEFORE the dropped row.
+    const headerAndTurn = logText.split('\n').slice(0, 2).join('\n') + '\n'
+    expect(scanned.committedBytes).toBe(Buffer.byteLength(headerAndTurn, 'utf8'))
+  })
+
+  it('eventLines(packChunks: false) is byte-identical to the pre-packing layout', () => {
+    const log = chunkRunLog()
+    expect(eventLines(log, false)).toBe(log.map(e => JSON.stringify(e)).join('\n'))
+  })
+})
+
 describe('SessionPersistenceJsonl: edge cases', () => {
   let ctx: Context
   beforeEach(async () => {

packages/session-persistence/session-persistence-jsonl/tests/zstd.spec.ts

@@ -7,7 +7,7 @@ import { join } from 'node:path'
 import SessionStore, { SessionId } from '@deepseek-ai/dsh-session'
 import type { SessionEvent } from '@deepseek-ai/dsh-session'
 import SessionPersistenceJsonl from '@deepseek-ai/dsh-session-persistence-jsonl'
-import { eventLine, logPath, scanLog, sessionDir, toHeaderLine, type JsonlCompression } from '../src/format.ts'
+import { logPath, scanLog, sessionDir, toHeaderLine, type JsonlCompression } from '../src/format.ts'
 import { compressZstdFrame, decompressZstdFrame, scanZstdFrames } from '../src/zstd.ts'
 import { runPersistenceContract, meta, oneTurnLog } from '../../session-persistence/tests/contract.ts'
 import { runCoordinatorContract, type CoordinatorFixture } from '../../session-persistence/tests/coordinator-contract.ts'
@@ -217,7 +217,7 @@ describe('SessionPersistenceJsonl: default Zstandard encoding', () => {
     const plaintext = await decodeCompleteFrames(buffer)
     expect(plaintext.toString()).toBe([
       JSON.stringify(toHeaderLine(header)),
-      ...oneTurnLog().map(eventLine),
+      ...oneTurnLog().map(e => JSON.stringify(e)),
       '',
     ].join('\n'))
     expect((await ctx.sessionPersistence.load(header.id)).events).toEqual(oneTurnLog())
@@ -288,7 +288,7 @@ describe('SessionPersistenceJsonl: default Zstandard encoding', () => {
       { type: 'step/start', seq: 7, time: 8, data: { turn: 2, step: 1 } },
       { type: 'assistant/chunk', seq: 8, time: 9, data: { turn: 2, step: 1, chunk: { type: 'text-delta', index: 0, text: deterministicNoise(300_000) } } },
     ] as SessionEvent[]
-    const plaintext = openTurn.map(eventLine).join('\n') + '\n'
+    const plaintext = openTurn.map(e => JSON.stringify(e)).join('\n') + '\n'
     const partial = await tornFrame(plaintext, (decoded) => {
       const newlines = decoded.match(/\n/g)?.length ?? 0
       return newlines >= 2 && !decoded.endsWith('\n')
@@ -334,7 +334,7 @@ describe('SessionPersistenceJsonl: default Zstandard encoding', () => {
       { type: 'turn/start', seq: 6, time: 7, data: { turn: 2, trigger: { kind: 'message', source: { kind: 'user' } } } },
       { type: 'turn/end', seq: 7, time: 8, data: { turn: 2, reason: { kind: 'completed' } } },
     ] as SessionEvent[]
-    const frame = await compressZstdFrame(secondTurn.map(eventLine).join('\n') + '\n')
+    const frame = await compressZstdFrame(secondTurn.map(e => JSON.stringify(e)).join('\n') + '\n')
     await appendFile(path, frame.subarray(0, -1))
 
     const loaded = await ctx.sessionPersistence.load(header.id)
@@ -456,7 +456,7 @@ describe('SessionPersistenceJsonl: encoding selection', () => {
     await mkdir(sessionDir(root, loadHeader.cwd), { recursive: true })
     await writeFile(logPath(root, loadHeader.cwd, loadHeader.id, 'none'), [
       JSON.stringify(toHeaderLine(loadHeader)),
-      ...oneTurnLog().map(eventLine),
+      ...oneTurnLog().map(e => JSON.stringify(e)),
       '',
     ].join('\n'))
     await expect(ctx.sessionPersistence.load(loadHeader.id)).rejects.toThrow(/uses \.jsonl/)
@@ -474,7 +474,7 @@ describe('SessionPersistenceJsonl: encoding selection', () => {
     await mkdir(sessionDir(root, header.cwd), { recursive: true })
     await writeFile(logPath(root, header.cwd, header.id, 'none'), [
       JSON.stringify(toHeaderLine(header)),
-      ...oneTurnLog().map(eventLine),
+      ...oneTurnLog().map(e => JSON.stringify(e)),
       '',
     ].join('\n'))
     await expect(ctx.sessionPersistence.append(header.id, oneTurnLog())).rejects.toThrow(/uses \.jsonl/)