Files
obelisk/scripts/indexer.mjs
T
tommy0103 b524339d85 feat(app): add Electron desktop UI and evolve memory/retrieval layer
Introduce an Electron app with session browser, memory list, and usage
  views (vanilla JS + Vue scaffolding). On the data layer: add content_type
  and is_meta to messages for transcript control-plane filtering, introduce
  FTS5-backed memory recall with safe tokenization, support memory archival
  via forget() through the renamed --attune runtime, and expose anchors on
  memory records.
2026-06-12 22:20:29 +08:00

290 lines
12 KiB
JavaScript

import { CLAUDE_DIR, openDb, rebuildMemoryFts, trunc, truncJson, extractText, extractContentType, extractMessageIsMeta, filePath, isDir, readLines, fs, path } from './db.mjs';
const PROJECTS_DIR = path.join(CLAUDE_DIR, 'projects');
const HISTORY_PATH = path.join(CLAUDE_DIR, 'history.jsonl');
function legacyProjectPathFromSlug(project) {
if (!project) return null;
return '/' + project.replace(/-/g, '/').replace(/^\//, '');
}
function normalizeObservedCwd(cwd) {
if (typeof cwd !== 'string' || !cwd.trim() || !path.isAbsolute(cwd)) return null;
return path.normalize(cwd);
}
function inferProjectPath(project, observedCwds = []) {
const byPath = new Map();
for (const cwd of observedCwds) {
const normalized = normalizeObservedCwd(cwd);
if (!normalized) continue;
const current = byPath.get(normalized) || { path: normalized, count: 0, first: byPath.size };
current.count++;
byPath.set(normalized, current);
}
const best = [...byPath.values()].sort((a, b) => b.count - a.count || a.first - b.first)[0];
return best?.path || legacyProjectPathFromSlug(project);
}
function discoverJsonlFiles() {
const files = [];
if (!fs.existsSync(PROJECTS_DIR)) return files;
let projects;
try { projects = fs.readdirSync(PROJECTS_DIR); } catch (e) { process.stderr.write(`Warning: cannot read projects dir: ${e.message}\n`); return files; }
for (const proj of projects) {
const projPath = path.join(PROJECTS_DIR, proj);
if (!isDir(projPath)) continue;
let entries;
try { entries = fs.readdirSync(projPath); } catch { continue; }
for (const f of entries) {
if (f.endsWith('.jsonl'))
files.push({ path: path.join(projPath, f), sessionId: f.slice(0, -6), project: proj, isSubagent: false });
}
for (const sd of entries) {
const saDir = path.join(projPath, sd, 'subagents');
if (!isDir(saDir)) continue;
let saEntries;
try { saEntries = fs.readdirSync(saDir); } catch { continue; }
for (const sf of saEntries) {
if (sf.endsWith('.jsonl'))
files.push({ path: path.join(saDir, sf), sessionId: sd, project: proj, isSubagent: true, agentId: sf.slice(0, -6) });
}
const wfRoot = path.join(saDir, 'workflows');
if (!isDir(wfRoot)) continue;
let wfDirs;
try { wfDirs = fs.readdirSync(wfRoot); } catch { continue; }
for (const wfDir of wfDirs) {
const wfPath = path.join(wfRoot, wfDir);
if (!isDir(wfPath)) continue;
let wfEntries;
try { wfEntries = fs.readdirSync(wfPath); } catch { continue; }
for (const wf of wfEntries) {
if (wf.endsWith('.jsonl'))
files.push({ path: path.join(wfPath, wf), sessionId: sd, project: proj, isSubagent: true, agentId: wf.slice(0, -6), workflowRunId: wfDir });
}
}
}
}
return files;
}
function needsReindex(db, fp) {
const mt = fs.statSync(fp).mtimeMs;
const row = db.prepare('SELECT mtime, lines_processed FROM index_state WHERE jsonl_path = ?').get(fp);
if (!row) return { needed: true, skip: 0 };
return mt > row.mtime ? { needed: true, skip: row.lines_processed } : { needed: false, skip: 0 };
}
function indexJsonl(db, fi) {
const { needed, skip } = needsReindex(db, fi.path);
if (!needed) return;
const mt = fs.statSync(fi.path).mtimeMs;
const ins = {
ses: db.prepare('INSERT OR REPLACE INTO sessions (id,title,project,project_path,started_at,ended_at,git_branch,version,message_count,jsonl_path) VALUES (?,?,?,?,?,?,?,?,?,?)'),
msg: db.prepare('INSERT OR REPLACE INTO messages (uuid,session_id,type,parent_uuid,timestamp,role,text,content_type,is_meta,model,is_sidechain,agent_id,input_tokens,output_tokens,cwd,skill) VALUES (?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)'),
tc: db.prepare('INSERT OR REPLACE INTO tool_calls (id,message_uuid,session_id,name,input_json,file_path) VALUES (?,?,?,?,?,?)'),
tr: db.prepare('INSERT OR REPLACE INTO tool_results (tool_use_id,message_uuid,session_id,content,file_path,is_error) VALUES (?,?,?,?,?,?)'),
sum: db.prepare('INSERT OR REPLACE INTO summaries (id,session_id,timestamp,source,content) VALUES (?,?,?,?,?)'),
idx: db.prepare('INSERT OR REPLACE INTO index_state (jsonl_path,mtime,lines_processed) VALUES (?,?,?)'),
};
const existing = !fi.isSubagent ? db.prepare('SELECT * FROM sessions WHERE id = ?').get(fi.sessionId) : null;
const sm = {
started_at: existing?.started_at || null,
ended_at: existing?.ended_at || null,
git_branch: existing?.git_branch || null,
version: existing?.version || null,
title: existing?.title || null,
n: existing?.message_count || 0,
cwds: [],
};
let lineNum = 0;
readLines(fi.path, (line) => {
lineNum++;
if (lineNum <= skip) return;
let obj;
try { obj = JSON.parse(line); } catch { return; }
const sid = fi.sessionId;
const ts = obj.timestamp || null;
if (obj.type === 'ai-title' && obj.aiTitle) { sm.title = obj.aiTitle; return; }
if (obj.type === 'system' && obj.subtype === 'away_summary' && obj.content) {
ins.sum.run(obj.uuid || `${sid}-away-${ts}`, sid, ts, 'away_summary', obj.content);
return;
}
if (obj.type === 'system' && obj.subtype === 'turn_duration' && obj.parentUuid && obj.durationMs) {
db.prepare('UPDATE messages SET turn_duration_ms=? WHERE uuid=?').run(obj.durationMs, obj.parentUuid);
return;
}
if (obj.type !== 'user' && obj.type !== 'assistant') return;
if (ts && (!sm.started_at || ts < sm.started_at)) sm.started_at = ts;
if (ts && (!sm.ended_at || ts > sm.ended_at)) sm.ended_at = ts;
if (obj.gitBranch) sm.git_branch = obj.gitBranch;
if (obj.version) sm.version = obj.version;
sm.n++;
if (!fi.isSubagent && obj.cwd) sm.cwds.push(obj.cwd);
const msg = obj.message || {};
const text = extractText(msg.content);
const contentType = extractContentType(msg.content);
const isMeta = extractMessageIsMeta(obj, text);
const usage = msg.usage || {};
const aid = fi.isSubagent ? fi.agentId : (obj.agentId || null);
if (obj.uuid) {
ins.msg.run(obj.uuid, sid, obj.type, obj.parentUuid || null, ts,
msg.role || obj.type, text, contentType, isMeta, msg.model || null,
obj.isSidechain ? 1 : 0, aid, usage.input_tokens || null, usage.output_tokens || null,
obj.cwd || null, obj.attributionSkill || null);
}
if (obj.type === 'assistant' && Array.isArray(msg.content)) {
for (const b of msg.content) {
if (b.type === 'tool_use' && b.id)
ins.tc.run(b.id, obj.uuid, sid, b.name, truncJson(b.input || {}), filePath(b.name, b.input));
}
}
if (obj.type === 'user' && Array.isArray(msg.content)) {
for (const b of msg.content) {
if (b.type !== 'tool_result' || !b.tool_use_id) continue;
const rt = typeof b.content === 'string' ? b.content
: Array.isArray(b.content) ? b.content.map(c => c.text || '').join('\n') : '';
ins.tr.run(b.tool_use_id, obj.uuid, sid, trunc(rt), obj.toolUseResult?.filePath || null, b.is_error ? 1 : 0);
}
}
});
if (!fi.isSubagent) {
const pp = inferProjectPath(fi.project, sm.cwds);
ins.ses.run(fi.sessionId, sm.title, fi.project, pp, sm.started_at, sm.ended_at, sm.git_branch, sm.version, sm.n, fi.path);
}
ins.idx.run(fi.path, mt, lineNum);
}
function refreshSessionProjectPaths(db) {
const sessions = db.prepare('SELECT id, project FROM sessions').all();
const cwdStmt = db.prepare(`
SELECT cwd
FROM messages
WHERE session_id = ? AND cwd IS NOT NULL AND cwd != ''
ORDER BY timestamp IS NULL, timestamp
`);
const update = db.prepare('UPDATE sessions SET project_path = ? WHERE id = ?');
for (const session of sessions) {
const cwds = cwdStmt.all(session.id).map(row => row.cwd);
const projectPath = inferProjectPath(session.project, cwds);
if (projectPath) update.run(projectPath, session.id);
}
}
function indexSubagentMeta(db, fi) {
if (!fi.isSubagent) return;
const mp = fi.path.replace('.jsonl', '.meta.json');
if (!fs.existsSync(mp)) return;
try {
const meta = JSON.parse(fs.readFileSync(mp, 'utf8'));
const tok = db.prepare('SELECT COALESCE(SUM(input_tokens),0)+COALESCE(SUM(output_tokens),0) as t FROM messages WHERE agent_id=?').get(fi.agentId);
const ts = db.prepare('SELECT MIN(timestamp) as t0, MAX(timestamp) as t1 FROM messages WHERE agent_id=?').get(fi.agentId);
const dur = ts?.t0 && ts?.t1 ? new Date(ts.t1).getTime() - new Date(ts.t0).getTime() : null;
if (fi.workflowRunId) {
db.prepare('INSERT OR REPLACE INTO workflow_agents (agent_id,run_id,session_id,agent_type,description) VALUES(?,?,?,?,?)').run(fi.agentId, fi.workflowRunId, fi.sessionId, meta.agentType||null, meta.description||null);
} else {
db.prepare('INSERT OR REPLACE INTO subagents VALUES(?,?,?,?,?,?,?)').run(fi.agentId, fi.sessionId, meta.toolUseId||null, meta.agentType||null, meta.description||null, dur, tok?.t||0);
}
} catch (e) { process.stderr.write(`Warning: failed to read subagent meta ${mp}: ${e.message}\n`); }
}
function indexWorkflows(db) {
if (!fs.existsSync(PROJECTS_DIR)) return;
let projects;
try { projects = fs.readdirSync(PROJECTS_DIR); } catch { return; }
for (const proj of projects) {
const pp = path.join(PROJECTS_DIR, proj);
if (!isDir(pp)) continue;
let entries;
try { entries = fs.readdirSync(pp); } catch { continue; }
for (const sd of entries) {
const wd = path.join(pp, sd, 'workflows');
if (!isDir(wd)) continue;
let wfFiles;
try { wfFiles = fs.readdirSync(wd); } catch { continue; }
for (const f of wfFiles) {
if (!f.endsWith('.json')) continue;
try {
const wf = JSON.parse(fs.readFileSync(path.join(wd, f), 'utf8'));
if (!wf.runId) continue;
const ac = db.prepare('SELECT COUNT(*) as c FROM workflow_agents WHERE run_id=?').get(wf.runId);
db.prepare('INSERT OR REPLACE INTO workflows (run_id,session_id,task_id,script,result_json,timestamp,agent_count,duration_ms,total_tokens,status,workflow_name) VALUES(?,?,?,?,?,?,?,?,?,?,?)').run(
wf.runId, sd, wf.taskId||null, wf.script||null,
wf.result ? JSON.stringify(wf.result) : null, wf.timestamp||null, ac?.c||0,
wf.durationMs||null, wf.totalTokens||null, wf.status||null, wf.workflowName||null);
const progress = wf.workflowProgress || [];
for (const item of progress) {
if (item.type !== 'workflow_agent' || !item.agentId) continue;
db.prepare('UPDATE workflow_agents SET phase=?, label=?, model=?, state=?, duration_ms=?, tokens=?, tool_calls=? WHERE agent_id=?').run(
item.phaseTitle||null, item.label||null, item.model||null, item.state||null,
item.durationMs||null, item.tokens||null, item.toolCalls||null, 'agent-' + item.agentId);
}
} catch (e) { process.stderr.write(`Warning: failed to index workflow ${f}: ${e.message}\n`); }
}
}
}
}
function indexHistory(db) {
if (!fs.existsSync(HISTORY_PATH)) return;
readLines(HISTORY_PATH, (line) => {
try {
const o = JSON.parse(line);
if (o.sessionId && o.title) db.prepare('UPDATE sessions SET title=? WHERE id=? AND title IS NULL').run(o.title, o.sessionId);
} catch (e) { process.stderr.write(`Warning: malformed history line: ${e.message}\n`); }
});
}
const BUILD_DEBOUNCE_MS = 30000;
function buildIndex({ force = false } = {}) {
const db = openDb();
if (!force) {
const last = db.prepare("SELECT mtime FROM index_state WHERE jsonl_path='__last_build__'").get();
if (last && Date.now() - last.mtime < BUILD_DEBOUNCE_MS) { db.close(); return; }
}
if (force) {
db.prepare("DELETE FROM index_state WHERE jsonl_path != '__last_build__'").run();
}
const files = discoverJsonlFiles();
for (const f of files) {
db.exec('BEGIN');
try {
indexJsonl(db, f);
indexSubagentMeta(db, f);
db.exec('COMMIT');
} catch (e) {
db.exec('ROLLBACK');
process.stderr.write(`Warning: failed to index ${f.path}: ${e.message}\n`);
}
}
db.exec('BEGIN');
try {
indexWorkflows(db);
refreshSessionProjectPaths(db);
indexHistory(db);
db.exec("INSERT INTO messages_fts(messages_fts) VALUES('rebuild')");
rebuildMemoryFts(db);
db.prepare("INSERT OR REPLACE INTO index_state (jsonl_path, mtime, lines_processed) VALUES ('__last_build__', ?, 0)").run(Date.now());
db.exec('COMMIT');
} catch (e) {
db.exec('ROLLBACK');
process.stderr.write(`Warning: failed to finalize index: ${e.message}\n`);
}
db.close();
}
export { buildIndex, inferProjectPath, refreshSessionProjectPaths };