CL-9026: stop writing duplicate prompt snapshot when identical to live turns

src/session/optimized-context-store.test.ts

@@ -960,3 +960,162 @@ describe("createOptimizedContextStore unpublished rewrite", () => {
     expect(turnTexts((await store.load()).turns)).toEqual(["era-2"]);
   });
 });
+
+describe("createOptimizedContextStore prompt dedupe (CL-9026)", () => {
+  const PROMPT_FILE = "prompt.jsonl";
+
+  function toolResultTurn(body: string): ConversationTurn {
+    return {
+      role: "user",
+      content: [
+        {
+          type: "tool_result",
+          callId: "call-1",
+          content: [{ type: "text", text: body }],
+        },
+      ],
+      timestamp: 1,
+    };
+  }
+
+  function cloneTurns(turns: ConversationTurn[]): ConversationTurn[] {
+    return turns.map((t) => JSON.parse(JSON.stringify(t)) as ConversationTurn);
+  }
+
+  test("identical writePrompt writes no prompt segment", async () => {
+    const dir = tempDir();
+    const store = await createOptimizedContextStore(dir);
+    const turns = [turn("a"), turn("b")];
+    await store.writeTurns([...turns]);
+    await store.writePrompt([...turns]);
+    await store.writePrompt(cloneTurns(turns));
+
+    expect(fs.existsSync(path.join(dir, PROMPT_FILE))).toBe(false);
+    expect(await listSegmentFiles(dir, PROMPT_FILE)).toEqual([]);
+
+    await store.writeMetadata(EMPTY_CHECKPOINT_METADATA);
+    await store.commit({ message: "identical prompt is a no-op" });
+    expect(await gitLsTree(dir)).not.toContain(PROMPT_FILE);
+
+    // A fresh instance compares against on-disk live turns and also skips.
+    const fresh = await createOptimizedContextStore(dir);
+    await fresh.writePrompt(cloneTurns(turns));
+    expect(fs.existsSync(path.join(dir, PROMPT_FILE))).toBe(false);
+  });
+
+  test("differing writePrompt still writes the prompt snapshot", async () => {
+    const dir = tempDir();
+    const store = await createOptimizedContextStore(dir);
+    const live = [turn("a"), toolResultTurn("x".repeat(1000))];
+    await store.writeTurns([...live]);
+
+    const shrunk = [turn("a"), toolResultTurn("summarized")];
+    await store.writePrompt(shrunk);
+    expect(fs.existsSync(path.join(dir, PROMPT_FILE))).toBe(true);
+    expect(fs.readFileSync(path.join(dir, PROMPT_FILE), "utf-8")).toBe(
+      jsonl(shrunk),
+    );
+
+    // An appended ephemeral turn also changes the prompt and still writes.
+    const ephemeralDir = tempDir();
+    const ephemeralStore = await createOptimizedContextStore(ephemeralDir);
+    await ephemeralStore.writeTurns([...live]);
+    const withEphemeral = [...cloneTurns(live), turn("ephemeral")];
+    await ephemeralStore.writePrompt(withEphemeral);
+    expect(fs.readFileSync(path.join(ephemeralDir, PROMPT_FILE), "utf-8")).toBe(
+      jsonl(withEphemeral),
+    );
+  });
+
+  test("turns-only store loads live turns and commits no prompt file", async () => {
+    const dir = tempDir();
+    const store = await createOptimizedContextStore(dir);
+    await store.writeTurns([turn("a"), turn("b")]);
+    await store.writeMetadata(EMPTY_CHECKPOINT_METADATA);
+    const first = await store.commit({ message: "turns only" });
+
+    expect(turnTexts((await store.load()).turns)).toEqual(["a", "b"]);
+
+    const fresh = await createOptimizedContextStore(dir);
+    expect(turnTexts((await fresh.load()).turns)).toEqual(["a", "b"]);
+    const second = await fresh.commit({ message: "turns-only no-op" });
+    expect(second.hash).toBe(first.hash);
+    expect(fs.existsSync(path.join(dir, PROMPT_FILE))).toBe(false);
+    expect(await gitLsTree(dir)).not.toContain(PROMPT_FILE);
+  });
+
+  test("identical writePrompt after a differing one removes stale prompt segments", async () => {
+    const dir = tempDir();
+    const store = await createOptimizedContextStore(dir);
+    const live = [turn("a"), turn("b")];
+    await store.writeTurns([...live]);
+    await store.writePrompt([turn("a")]);
+    expect(fs.existsSync(path.join(dir, PROMPT_FILE))).toBe(true);
+
+    await store.writePrompt(cloneTurns(live));
+    expect(fs.existsSync(path.join(dir, PROMPT_FILE))).toBe(false);
+    expect(await listSegmentFiles(dir, PROMPT_FILE)).toEqual([]);
+
+    await store.writeMetadata(EMPTY_CHECKPOINT_METADATA);
+    await store.commit({ message: "stale prompt removed" });
+    expect(await gitLsTree(dir)).not.toContain(PROMPT_FILE);
+    expect(turnTexts((await store.load()).turns)).toEqual(["a", "b"]);
+  });
+
+  test("committed then identical writePrompt drops prompt.jsonl from HEAD", async () => {
+    const dir = tempDir();
+    const store = await createOptimizedContextStore(dir);
+    const live = [turn("a"), turn("b")];
+    await store.writeTurns([...live]);
+    await store.writePrompt([turn("a")]);
+    await store.writeMetadata(EMPTY_CHECKPOINT_METADATA);
+    await store.commit({ message: "differing prompt" });
+    expect(await gitLsTree(dir)).toContain(PROMPT_FILE);
+
+    await store.writePrompt(cloneTurns(live));
+    await store.writeMetadata(EMPTY_CHECKPOINT_METADATA);
+    const afterIdentical = await store.commit({ message: "identical prompt" });
+    expect(await gitLsTree(dir)).not.toContain(PROMPT_FILE);
+    expect(turnTexts((await store.load()).turns)).toEqual(["a", "b"]);
+
+    await store.writeMetadata(EMPTY_CHECKPOINT_METADATA);
+    const empty = await store.commit({ message: "empty checkpoint" });
+    expect(empty.hash).toBe(afterIdentical.hash);
+  });
+
+  test("identical writePrompt after multi-segment differing prompt drops all prompt files from HEAD", async () => {
+    const dir = tempDir();
+    const store = await createOptimizedContextStore(dir);
+    const live: ConversationTurn[] = [];
+    const big = "x".repeat(20_000);
+    for (let i = 0; i < 18; i++) {
+      live.push(turn(`${i}-${big}`));
+    }
+    await store.writeTurns([...live]);
+
+    const differing: ConversationTurn[] = [];
+    for (let i = 0; i < 18; i++) {
+      differing.push(turn(`p-${i}-${big}`));
+    }
+    await store.writePrompt(differing);
+    const extraPrompt = segmentFileName(PROMPT_FILE, 1);
+    expect(fs.existsSync(path.join(dir, PROMPT_FILE))).toBe(true);
+    expect(fs.existsSync(path.join(dir, extraPrompt))).toBe(true);
+
+    await store.writeMetadata(EMPTY_CHECKPOINT_METADATA);
+    await store.commit({ message: "multi-segment prompt" });
+    const treeAfterDiffering = await gitLsTree(dir);
+    expect(treeAfterDiffering).toContain(PROMPT_FILE);
+    expect(treeAfterDiffering).toContain(extraPrompt);
+
+    await store.writePrompt(cloneTurns(live));
+    await store.writeMetadata(EMPTY_CHECKPOINT_METADATA);
+    await store.commit({ message: "identical drops prompt segments" });
+    const tree = await gitLsTree(dir);
+    expect(tree).not.toContain(PROMPT_FILE);
+    expect(tree).not.toContain(extraPrompt);
+    expect(turnTexts((await store.load()).turns)).toEqual(
+      live.map((t) => (t.content[0] as { text: string }).text),
+    );
+  });
+});

src/session/optimized-context-store.ts

@@ -527,7 +527,7 @@ export async function createSessionStores(
   const pendingBlobFilepaths = new Set<string>();
   const pendingSegmentPaths = new Set<string>();
   let writeTurnsSegmented = createSegmentedJSONLWriter(dir, TURNS_FILE);
-  const writePromptSegmented = createSegmentedJSONLWriter(dir, PROMPT_FILE);
+  let writePromptSegmented = createSegmentedJSONLWriter(dir, PROMPT_FILE);
   let liveTurnRefs: readonly ConversationTurn[] | null = null;
   let unpublishedRewrite: ConversationTurn[] | null = null;
 
@@ -606,6 +606,75 @@ export async function createSessionStores(
     liveTurnRefs = [...turns];
   }
 
+  function promptEqualsLiveTurns(
+    live: readonly ConversationTurn[],
+    prompt: readonly ConversationTurn[],
+  ): boolean {
+    if (live.length !== prompt.length) return false;
+    for (let index = 0; index < prompt.length; index++) {
+      if (live[index] === prompt[index]) continue;
+      if (JSON.stringify(live[index]) !== JSON.stringify(prompt[index]))
+        return false;
+    }
+    return true;
+  }
+
+  // The reactor checkpoints the materialized prompt every cycle, but most
+  // cycles run no transform that changes it — writing an identical snapshot
+  // next to turns.jsonl doubles disk and re-hash cost for zero information.
+  // load() never reads prompt.jsonl (base turns + TURNS_FILE extras only),
+  // so skipping the write leaves resume behavior unchanged; prompts that
+  // actually differ still write exactly as before.
+  async function writePromptIfDiffered(
+    turns: readonly ConversationTurn[],
+  ): Promise<void> {
+    let live: readonly ConversationTurn[] | null = null;
+    if (unpublishedRewrite !== null) live = unpublishedRewrite;
+    else if (liveTurnRefs !== null) live = liveTurnRefs;
+    else {
+      // Fresh instance with no writeTurns yet: recover live turns from disk
+      // the same way writeTurnsLiveOrStage does. Any failure falls through
+      // to a normal write — an unreadable baseline must not drop the snapshot.
+      try {
+        const extraTexts = await readExtraSegmentTexts(dir, TURNS_FILE);
+        let baseTurns: ConversationTurn[];
+        try {
+          baseTurns = (await base.load()).turns;
+        } catch {
+          baseTurns = await readBaseTurnsFromDisk(dir);
+        }
+        live =
+          extraTexts.length === 0
+            ? baseTurns
+            : await loadTurnsWithoutMalformedToolSequence(
+                baseTurns,
+                extraTexts,
+              );
+      } catch {
+        live = null;
+      }
+    }
+    if (live === null || !promptEqualsLiveTurns(live, turns)) {
+      await writeSegmented(writePromptSegmented, turns);
+      return;
+    }
+    // Identical to live turns: converge disk to no prompt segment so a stale
+    // snapshot from an earlier differing write cannot linger. Removals join
+    // pendingSegmentPaths so commit stages them out of the tree.
+    const highest = await highestSegmentIndex(dir, PROMPT_FILE);
+    let removed = false;
+    for (let index = 0; index <= highest; index++) {
+      const name = segmentFileName(PROMPT_FILE, index);
+      if (!(await pathExists(path.join(dir, name)))) continue;
+      await fs.promises.unlink(path.join(dir, name));
+      pendingSegmentPaths.add(name);
+      removed = true;
+    }
+    if (removed) {
+      writePromptSegmented = createSegmentedJSONLWriter(dir, PROMPT_FILE);
+    }
+  }
+
   // Prefer the longest prefix of base + extras whose tool sequence the reactor
   // will accept. Orphan tails left by a fresh-writer compaction rewrite are
   // dropped and unlinked so the next load does not re-poison the session.
@@ -718,7 +787,9 @@ export async function createSessionStores(
       return [...baseTurns, ...parsedExtras.slice(0, keepExtras).flat()];
     },
     readBlob: (key, signal) => base.readBlob(key, signal),
-    writePrompt: (turns) => writeSegmented(writePromptSegmented, turns),
+    // Skipped identical snapshots fall back to live turns in load(), which
+    // never reads prompt.jsonl.
+    writePrompt: (turns) => writePromptIfDiffered(turns),
     writeResponse: (turn, signal) => base.writeResponse(turn, signal),
     writeManifest: (records, signal) => base.writeManifest(records, signal),
     writeTurns: (turns) => writeTurnsLiveOrStage(turns),
@@ -793,9 +864,10 @@ export async function createSessionStores(
           }
 
           const add = extraCommitPaths([...new Set(toAdd)]);
-          const remove = extraCommitPaths([...new Set(toRemove)]).filter(
-            (p) => !add.includes(p),
-          );
+          // extraCommitPaths strips vendor roots because base.commit() git.adds
+          // those that still exist. It does not git.remove missing ones, so an
+          // unlinked prompt.jsonl must stay in `remove`.
+          const remove = [...new Set(toRemove)].filter((p) => !add.includes(p));
           extraPaths = [...new Set([...add, ...remove])];
 
           for (const filepath of add) {