fix(observer): uuid-dedup in parseLines (C-1 root fix for quirk #101)

Bug: Claude Code's transcript JSONL file accumulates duplicated context- rebuild snapshots — the same entry re-printed with the SAME `uuid`. Without dedup, session_turn / task_size / events double-count, and session_turn becomes non-monotonic across episodes parsed at different file-growth states. Live evidence: episodes-2026-05.jsonl lines 14/15/16 of the same session showed session_turn 139 → 140 → 91 (backwards in time). Probe on transcript 553717ec: 22400 entries, only 6074 unique uuid (68% dup rate); real user prompts 264 total vs 92 unique-uuid. Fix: parseLines now tracks a `seenUuid` Set and skips entries whose uuid has already been encountered (keep-first). Entries without `uuid` (synthetic test fixtures) pass through unchanged. All downstream functions (findTurnStart, extractEnvironment, extractTaskSize, etc.) operate on the deduped entries array, so the fix is single-point and total. Tests: new `parseTranscript — uuid-dedup` describe block covers (1) duplicated-uuid prompts collapse → session_turn counts once, (2) distinct-uuid entries preserved (no over-dedup), (3) no-uuid entries pass through (synthetic-fixture safety), (4) duplicated-uuid assistant turns → tool_calls / files_touched counted once. 110/110 parser tests green (was 106). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 11:00:50 +03:00
parent 1ba8b6e590
commit 94f831f7d1
2 changed files with 117 additions and 1 deletions
@@ -670,6 +670,108 @@ describe('parseTranscript — AskUserQuestion in-turn choice', () => {
  });
 });

+describe('parseTranscript — uuid-dedup for duplicated snapshots (quirk #101)', () => {
+  // Claude Code's transcript file accumulates duplicated context-rebuild
+  // snapshots; the same entry is re-printed with the SAME `uuid`. Without
+  // dedup, session_turn / task_size / events double-count and become
+  // non-monotonic across episodes parsed at different file-growth states.
+  // Root fix: dedup by uuid in parseLines.
+
+  it('collapses duplicated-uuid user prompts (session_turn counts once)', () => {
+    const dup = 'aaaa-bbbb-cccc';
+    const lines = [
+      JSON.stringify({
+        uuid: dup,
+        type: 'user',
+        message: { role: 'user', content: 'hi' },
+        timestamp: '2026-05-19T10:00:00Z',
+        sessionId: 's1',
+      }),
+      JSON.stringify({
+        uuid: dup,
+        type: 'user',
+        message: { role: 'user', content: 'hi' },
+        timestamp: '2026-05-19T10:00:00Z',
+        sessionId: 's1',
+      }),
+    ].join('\n');
+    const ep = parseTranscript(lines, 'fallback');
+    expect(ep.environment.session_turn).toBe(1);
+  });
+
+  it('preserves distinct-uuid entries (no over-dedup)', () => {
+    const lines = [
+      JSON.stringify({
+        uuid: 'one',
+        type: 'user',
+        message: { role: 'user', content: 'first' },
+        timestamp: '2026-05-19T09:00:00Z',
+        sessionId: 's1',
+      }),
+      JSON.stringify({
+        uuid: 'two',
+        type: 'user',
+        message: { role: 'user', content: 'second' },
+        timestamp: '2026-05-19T10:00:00Z',
+        sessionId: 's1',
+      }),
+    ].join('\n');
+    const ep = parseTranscript(lines, 'fallback');
+    expect(ep.environment.session_turn).toBe(2);
+  });
+
+  it('entries without uuid pass through unchanged (synthetic fixtures unaffected)', () => {
+    // Existing tests build entries without `uuid` — must still work.
+    const t = jsonl([
+      userPrompt('one', '2026-05-19T09:00:00Z'),
+      userPrompt('two', '2026-05-19T10:00:00Z'),
+    ]);
+    expect(parseTranscript(t).environment.session_turn).toBe(2);
+  });
+
+  it('collapses duplicated-uuid assistant turns (task_size counted once)', () => {
+    const lines = [
+      JSON.stringify({
+        uuid: 'u-prompt',
+        type: 'user',
+        message: { role: 'user', content: 'go' },
+        timestamp: '2026-05-19T10:00:00Z',
+        sessionId: 's1',
+      }),
+      JSON.stringify({
+        uuid: 'u-asst',
+        type: 'assistant',
+        message: {
+          role: 'assistant',
+          content: [
+            { type: 'tool_use', id: 't1', name: 'Read', input: { file_path: '/x.js' } },
+            { type: 'tool_use', id: 't2', name: 'Edit', input: { file_path: '/x.js' } },
+          ],
+        },
+        timestamp: '2026-05-19T10:01:00Z',
+        sessionId: 's1',
+      }),
+      // EXACT duplicate of the assistant entry — snapshot rebuild artifact
+      JSON.stringify({
+        uuid: 'u-asst',
+        type: 'assistant',
+        message: {
+          role: 'assistant',
+          content: [
+            { type: 'tool_use', id: 't1', name: 'Read', input: { file_path: '/x.js' } },
+            { type: 'tool_use', id: 't2', name: 'Edit', input: { file_path: '/x.js' } },
+          ],
+        },
+        timestamp: '2026-05-19T10:01:00Z',
+        sessionId: 's1',
+      }),
+    ].join('\n');
+    const ep = parseTranscript(lines, 'fallback');
+    expect(ep.task_size.tool_calls).toBe(2);
+    expect(ep.task_size.files_touched).toBe(1);
+  });
+});
+
 describe('extractEnvironment — parallel_session narrowed to tool_result evidence', () => {
  // Collision detection scans ONLY tool_result content (real command output).
  // Prose mentions in user prompts / assistant text — including analysis text