feat(providers): migrate codex indexing to adapter + persist, remove legacy indexers (Phase 5c)

Complete the skill-side provider migration: codex now goes through a pure adapter and the shared persist layer, and the two original monolithic indexers are gone. New: - scripts/providers/codex.ts — pure codex adapter. Full-reparse (buffers the whole file) because the event_msg↔response_item dedup needs whole-file, bidirectional knowledge; emits SessionRecord with countMode 'total'. Handles guardian threads (→ delete-session), agent spawns/tool calls (→ tool_call/subagent), token_count (patched onto the message record) and task_complete (→ message-turn-duration). Contract: - SessionRecord.countMode ('total' | 'delta') tells persist whether to replace or accumulate message_count — claude is line-incremental (delta), codex full-reparse (total). SubagentRecord non-key fields are optional; persist merges them column-wise with COALESCE. MessageTurnDurationRecord.turn_duration_ms is nullable. Orchestration: - buildIndex's codex branch parses via the adapter and writes via persist. An unchanged file is skipped but still swept for stale guardian rows (routed through persist as a delete-session), preserving prior behavior. Cleanup: - Remove the now-unused indexJsonl, indexCodexJsonl, deleteCodexThreadRows and upsertCodexSubagent — their semantics now live in the adapters + persist. indexer.mjs drops from ~840 to 428 lines. Codex pure helpers stay exported for codex.ts and the guardian sweep (physical move deferred to the app-side reorg). - Migrate the upsert drift test off indexJsonl to the claude.parse + persist path, keeping the rowid-stability and count-replace regression guards. Tests: tests/codex-parse.test.mjs (record-stream golden: dedup, tools, token patch, turn-duration, guardian→delete) and tests/codex-index.test.mjs (full buildIndex path: fresh build + incremental full-reparse, total-count replace, no duplicates). Verified equivalent on the real ~/.obelisk index: codex messages 82476 and subagents 522 identical before/after, zero guardian leakage; real incremental confirmed (touch a codex file → reparsed idempotently, unchanged files skipped). lint + typecheck clean, 119/119.
2026-07-08 20:43:54 +08:00
parent 1c346039f1
commit 0598c29aad
8 changed files with 474 additions and 469 deletions
@@ -0,0 +1,220 @@
+// Codex provider adapter (see docs/adr/0001).
+//
+// Pure: discovers Codex rollout files and parses one into a record stream. It
+// never touches the Obelisk database. Unlike claude, codex is a FULL-REPARSE
+// adapter: it buffers every line and re-emits every record on each run, because
+// the event_msg ↔ response_item dedup needs whole-file (bidirectional) knowledge
+// (the matching pair sits ±1 line apart but in either order). Hence the session
+// record uses countMode 'total' (persist replaces the count, never accumulates).
+// The per-line logic mirrors the original indexCodexJsonl.
+
+import { createRequire } from 'node:module';
+const require = createRequire(import.meta.url);
+const fs = require('node:fs');
+
+import { trunc, truncJson, readLines } from '../db.mjs';
+import {
+  discoverCodexJsonlFiles, normalizeObservedCwd, projectSlugFromPath,
+  codexRawId, codexDbId, codexCallId, codexLineUuid, codexParentThreadId,
+  codexIsGuardianThread, codexAgentNickname, codexAgentRole, codexUsage,
+  codexEventText, codexMessagePayloadText, codexVisibleMessageKey,
+  codexToolInput, codexToolOutput,
+} from '../indexer.mjs';
+
+import type { Cursor, DiscoverContext, IndexRecord, IndexUnit, MessageRecord, Provider } from './types.ts';
+
+export const name = 'codex';
+
+export function discover(_ctx: DiscoverContext): IndexUnit[] {
+  return discoverCodexJsonlFiles().map((f: any) => ({ key: f.path, sessionId: '', meta: { source: 'codex' } }));
+}
+
+export function* parse(unit: IndexUnit, _cursor: Cursor): Generator<IndexRecord, Cursor> {
+  const mtime = fs.statSync(unit.key).mtimeMs;
+  const records: { lineNum: number; obj: any }[] = [];
+  let lineNum = 0;
+  readLines(unit.key, (line: string) => {
+    lineNum++;
+    try { records.push({ lineNum, obj: JSON.parse(line) }); } catch { /* skip malformed */ }
+  });
+  const outCursor = `${mtime}:${lineNum}`;
+
+  const metaRecord = records.find(r => r.obj?.type === 'session_meta' && r.obj.payload?.id);
+  if (!metaRecord) return outCursor;
+
+  const meta = metaRecord.obj.payload;
+  const threadRawId = codexRawId(meta.id) as string;
+  if (codexIsGuardianThread(meta, records)) {
+    yield { kind: 'delete-session', sessionId: codexDbId(threadRawId) as string };
+    return outCursor;
+  }
+
+  const parentRawId = codexParentThreadId(meta);
+  const sessionId = codexDbId(parentRawId || threadRawId) as string;
+  const agentId = (parentRawId ? codexDbId(threadRawId) : null) as string | null;
+  const isSidechain: 0 | 1 = agentId ? 1 : 0;
+  const project = projectSlugFromPath(normalizeObservedCwd(meta.cwd));
+  const lineUuid = (n: number): string => codexLineUuid(threadRawId, n) as string;
+
+  const out: IndexRecord[] = [];
+  const msgByUuid = new Map<string, MessageRecord>();
+  const sm = {
+    started_at: (meta.timestamp || metaRecord.obj.timestamp || null) as string | null,
+    ended_at: (meta.timestamp || metaRecord.obj.timestamp || null) as string | null,
+    git_branch: (meta.git?.branch || null) as string | null,
+    version: (meta.cli_version || null) as string | null,
+    title: null as string | null,
+    n: 0,
+    lastMessageUuid: null as string | null,
+    lastTextAssistantUuid: null as string | null,
+    totalInputTokens: 0,
+    totalOutputTokens: 0,
+  };
+
+  let currentCwd = normalizeObservedCwd(meta.cwd);
+  let currentModel: string | null = null;
+  const eventMessageKeys = new Set<string>();
+  const callMessageUuids = new Map<string, string>();
+
+  const updateBounds = (ts: string | null) => {
+    if (!ts) return;
+    if (!sm.started_at || ts < sm.started_at) sm.started_at = ts;
+    if (!sm.ended_at || ts > sm.ended_at) sm.ended_at = ts;
+  };
+
+  const insertMessage = ({ uuid, type, role, text = null, contentType = 'text', timestamp, isMeta = 0 }: {
+    uuid: string; type: string; role: string; text?: string | null; contentType?: string; timestamp: string | null; isMeta?: 0 | 1;
+  }) => {
+    const rec: MessageRecord = {
+      kind: 'message', uuid, session_id: sessionId, type, parent_uuid: sm.lastMessageUuid,
+      timestamp: timestamp || null, role, text: trunc(text), content_type: contentType,
+      is_meta: isMeta, model: currentModel, is_sidechain: isSidechain, agent_id: agentId,
+      input_tokens: null, output_tokens: null, cwd: currentCwd, skill: null, source: 'codex',
+    };
+    out.push(rec);
+    msgByUuid.set(uuid, rec);
+    sm.lastMessageUuid = uuid;
+    if (!agentId) sm.n++;
+    if (type === 'assistant' && contentType === 'text') sm.lastTextAssistantUuid = uuid;
+    updateBounds(timestamp);
+    return uuid;
+  };
+
+  // First pass: collect visible event_msg keys so duplicate response_items drop.
+  for (const { obj } of records) {
+    if (obj?.type !== 'event_msg') continue;
+    const payload = obj.payload || {};
+    if (payload.type !== 'user_message' && payload.type !== 'agent_message') continue;
+    const text = codexEventText(payload);
+    if (text === null) continue;
+    eventMessageKeys.add(codexVisibleMessageKey(payload.type === 'user_message' ? 'user' : 'assistant', text));
+  }
+
+  for (const { lineNum: currentLine, obj } of records) {
+    const ts = obj.timestamp || null;
+    if (obj.type === 'session_meta') {
+      if (obj.payload?.cwd) currentCwd = normalizeObservedCwd(obj.payload.cwd) || currentCwd;
+      if (obj.payload?.git?.branch) sm.git_branch = obj.payload.git.branch;
+      if (obj.payload?.cli_version) sm.version = obj.payload.cli_version;
+      updateBounds(obj.payload?.timestamp || ts);
+      continue;
+    }
+    if (obj.type === 'turn_context') {
+      currentCwd = normalizeObservedCwd(obj.payload?.cwd) || currentCwd;
+      currentModel = obj.payload?.model || currentModel;
+      updateBounds(ts);
+      continue;
+    }
+    if (obj.type === 'event_msg') {
+      const payload = obj.payload || {};
+      if (payload.type === 'user_message' || payload.type === 'agent_message' || payload.type === 'agent_reasoning') {
+        const text = codexEventText(payload);
+        if (text === null) continue;
+        const isReasoning = payload.type === 'agent_reasoning';
+        insertMessage({
+          uuid: lineUuid(currentLine),
+          type: payload.type === 'user_message' ? 'user' : 'assistant',
+          role: payload.type === 'user_message' ? 'user' : 'assistant',
+          text, contentType: isReasoning ? 'thinking' : 'text', timestamp: ts,
+        });
+        continue;
+      }
+      if (payload.type === 'collab_agent_spawn_end' && payload.call_id && payload.new_thread_id) {
+        const uuid = insertMessage({ uuid: lineUuid(currentLine), type: 'assistant', role: 'assistant', text: null, contentType: 'tool_use', timestamp: ts });
+        const toolId = codexCallId(payload.call_id) as string;
+        const description = payload.new_agent_nickname || payload.new_agent_role || 'Agent';
+        const input = {
+          description, subagent_type: payload.new_agent_role || 'Agent', prompt: payload.prompt || '',
+          new_thread_id: payload.new_thread_id, model: payload.model || null, reasoning_effort: payload.reasoning_effort || null,
+        };
+        out.push({ kind: 'tool_call', id: toolId, message_uuid: uuid, session_id: sessionId, name: 'Agent', input_json: truncJson(input) as string, file_path: null });
+        callMessageUuids.set(toolId, uuid);
+        out.push({ kind: 'subagent', agent_id: codexDbId(payload.new_thread_id) as string, session_id: sessionId, parent_tool_use_id: toolId, agent_type: payload.new_agent_role || null, description });
+        continue;
+      }
+      if (payload.type === 'task_complete') {
+        if (sm.lastTextAssistantUuid && payload.duration_ms !== undefined) {
+          out.push({ kind: 'message-turn-duration', uuid: sm.lastTextAssistantUuid, turn_duration_ms: payload.duration_ms || null });
+        }
+        updateBounds(ts);
+        continue;
+      }
+      if (payload.type === 'token_count') {
+        const usage = codexUsage(payload);
+        if (usage.inputTokens !== null) sm.totalInputTokens = usage.inputTokens;
+        if (usage.outputTokens !== null) sm.totalOutputTokens = usage.outputTokens;
+        if (sm.lastTextAssistantUuid && (usage.inputTokens !== null || usage.outputTokens !== null)) {
+          const rec = msgByUuid.get(sm.lastTextAssistantUuid);
+          if (rec) { rec.input_tokens = usage.inputTokens; rec.output_tokens = usage.outputTokens; }
+        }
+        continue;
+      }
+      if (payload.type === 'thread_name_updated' && payload.thread_name) sm.title = payload.thread_name;
+      continue;
+    }
+    if (obj.type !== 'response_item') continue;
+    const payload = obj.payload || {};
+    if (payload.type === 'message' && payload.role !== 'developer') {
+      const text = codexMessagePayloadText(payload);
+      const role = payload.role || 'assistant';
+      if (text !== null && !eventMessageKeys.has(codexVisibleMessageKey(role, text))) {
+        insertMessage({ uuid: lineUuid(currentLine), type: role === 'user' ? 'user' : 'assistant', role, text, contentType: 'text', timestamp: ts });
+      }
+      continue;
+    }
+    if (['function_call', 'custom_tool_call', 'tool_search_call', 'web_search_call'].includes(payload.type) && payload.call_id) {
+      const uuid = insertMessage({ uuid: lineUuid(currentLine), type: 'assistant', role: 'assistant', text: null, contentType: 'tool_use', timestamp: ts });
+      const name = payload.name || payload.tool || payload.type.replace(/_call$/, '');
+      const toolId = codexCallId(payload.call_id) as string;
+      out.push({ kind: 'tool_call', id: toolId, message_uuid: uuid, session_id: sessionId, name, input_json: truncJson(codexToolInput(payload)) as string, file_path: null });
+      callMessageUuids.set(toolId, uuid);
+      continue;
+    }
+    if (['function_call_output', 'custom_tool_call_output', 'tool_search_output'].includes(payload.type) && payload.call_id) {
+      const toolId = codexCallId(payload.call_id) as string;
+      out.push({ kind: 'tool_result', tool_use_id: toolId, message_uuid: callMessageUuids.get(toolId) || '', session_id: sessionId, content: trunc(codexToolOutput(payload) || ''), file_path: null, is_error: payload.is_error ? 1 : 0 });
+    }
+  }
+
+  if (agentId) {
+    const started = sm.started_at ? new Date(sm.started_at).getTime() : null;
+    const ended = sm.ended_at ? new Date(sm.ended_at).getTime() : null;
+    const tokenTotal = (sm.totalInputTokens || 0) + (sm.totalOutputTokens || 0);
+    out.push({
+      kind: 'subagent', agent_id: agentId, session_id: sessionId,
+      agent_type: codexAgentRole(meta), description: codexAgentNickname(meta),
+      duration_ms: started && ended ? ended - started : null, total_tokens: tokenTotal || null,
+    });
+  } else {
+    out.push({
+      kind: 'session', id: sessionId, title: sm.title, project,
+      started_at: sm.started_at, ended_at: sm.ended_at, git_branch: sm.git_branch, version: sm.version,
+      message_count: sm.n, countMode: 'total', jsonl_path: unit.key, source: 'codex',
+    });
+  }
+
+  yield* out;
+  return outCursor;
+}
+
+export const codexProvider: Provider = { name, discover, parse };