Files
platform/src/servers/sidecar/email/imap-sync.ts
T
pastilhasandClaude Opus 5 0c90216c7a email: keep an account's sync position inside its own emails.db
The messages were in emails.db and the position — last_sync_at, and per-folder uidvalidity/lastuid —
was a jsonb column on email_accounts in Postgres. Two stores for one fact, with an edge that only shows
up when you try to move a mailbox to another machine.

The expensive part of an email account is the first sync: hours of IMAP for a large mailbox, which is
exactly why "copy emails.db to the new server" is the obvious way to bring one across. With the position
in Postgres that silently does not work — the new server's column is empty, !last_sync_at says first
sync, and the whole mailbox downloads again on top of the one just restored.

The other direction is quieter and worse. Restore an OLDER emails.db while Postgres holds a NEWER
position and the sidecar skips every message between the two, permanently, because nothing looks below
lastuid again. Re-syncing is slow; skipping mail is data loss nobody notices.

Not a new idea — the Gmail path already read SQLite and fell back to Postgres, backfilling so the
fallback was taken once. Only the IMAP path had not followed. This extracts that pattern so both use one
copy, and unifies the isFirstSync fork in accounts.ts, which is how the two drifted apart to begin with.

The file wins over Postgres, always, and only migrates when it holds nothing at all. Topping up a
partial position from Postgres would reintroduce precisely the divergence this removes.

email_accounts.sync_meta is kept and marked legacy rather than dropped: it is the one-time backfill
source for every account created before this, and dropping it would strand any that has not synced
since. Nothing writes to it now.

11 tests on the migration, aimed at both expensive failures — migrating when we should not, and failing
to migrate an account that predates the change.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-08-10 13:39:13 +00:00

435 lines
16 KiB
TypeScript

import { createHash } from 'node:crypto';
import { openEmailDb, upsertFromRawEml } from './store';
import { readSyncMeta, writeSyncMeta } from './sync-meta';
import { refreshGoogleAccessToken } from '../../api/integrations/google-auth';
import {
getEmailAccount,
getUserIntegration,
upsertUserIntegration,
getServerIntegration,
updateEmailAccountStatus,
getDockPaths,
setDockPaths,
} from 'officerdb';
import type { SyncStep, SyncCtx } from './sync-runner';
// Thrown by the handlers for failures that a retry cannot fix (missing credentials, deleted account).
// The queue used it to skip retrying; here it is an ordinary error whose message reaches the sync state.
export class PermanentError extends Error {}
// ── Types for job meta (passed by the API server at enqueue time) ──
type EmailSyncMeta = {
emailAccountId: number;
userEmail: string;
account: {
id: number;
userId: number;
email: string;
imapHost: string;
imapPort: number;
imapSecure: boolean;
provider: string;
authType: string;
credentials: Record<string, unknown>;
};
imapAuth: { user: string; pass?: string; accessToken?: string };
saved?: number;
};
// ── Stable ID from Message-Id header ──
function messageIdToStableId(raw: string): string | null {
const match = raw.match(/^Message-Id:\s*<?([^>\s]+)>?/im);
if (!match?.[1]) return null;
return createHash('sha1').update(match[1]).digest('hex').slice(0, 16);
}
// ── IMAP folder → label mapping ──
const SPECIAL_USE_LABEL_MAP: Record<string, string> = {
'\\Inbox': 'inbox',
'\\Sent': 'sent',
'\\Drafts': 'draft',
'\\Flagged': 'starred',
'\\Trash': 'trash',
'\\Junk': 'spam',
'\\All': 'archive',
};
const SKIP_SPECIAL_USE = new Set(['\\Trash', '\\Junk', '\\All']);
type FolderInfo = {
specialUse?: string;
path: string;
flags: Set<string>;
status?: { uidNext?: number; uidValidity?: number };
};
function shouldSkipFolder(folder: FolderInfo): boolean {
if (folder.flags.has('\\Noselect') || folder.flags.has('\\NonExistent')) return true;
if (folder.specialUse && SKIP_SPECIAL_USE.has(folder.specialUse)) return true;
return false;
}
function folderToLabel(folder: FolderInfo): string {
if (folder.specialUse && SPECIAL_USE_LABEL_MAP[folder.specialUse]) {
return SPECIAL_USE_LABEL_MAP[folder.specialUse]!;
}
if (folder.path === 'INBOX') return 'inbox';
return folder.path.toLowerCase();
}
// ── OAuth token refresh ──
async function resolveImapAuth(meta: EmailSyncMeta): Promise<{ user: string; pass?: string; accessToken?: string }> {
if (meta.account.authType !== 'oauth') return meta.imapAuth;
// Refresh OAuth token if expired
const userGoogle = await getUserIntegration(meta.account.userId, 'google');
const config = userGoogle?.config as Record<string, unknown> | undefined;
if (!config?.refreshToken) {
throw new PermanentError('Google OAuth not configured — reconnect your Google account');
}
const expiresAt = config.expiresAt as number | undefined;
const tokenExpired = !expiresAt || expiresAt < Date.now() + 60_000;
if (!tokenExpired && config.accessToken) {
return { user: meta.account.email, accessToken: config.accessToken as string };
}
console.log('[email-sync] Refreshing OAuth token');
const refreshed = await refreshGoogleAccessToken(config.refreshToken as string);
const serverGoogle = await getServerIntegration('google');
await upsertUserIntegration({
userId: meta.account.userId,
provider: 'google',
serverIntegrationId: serverGoogle?.id,
config: { ...config, accessToken: refreshed.accessToken, expiresAt: refreshed.expiresAt },
});
return { user: meta.account.email, accessToken: refreshed.accessToken };
}
// ── Handler ──
const emailSyncHandler = {
type: 'email-sync',
retry: { delayMs: 15 * 60 * 1000, maxRetries: 5 },
steps: [
{
name: 'Sync emails',
run: async (ctx: SyncCtx) => {
const { ImapFlow } = await import('imapflow');
const meta = ctx.meta as unknown as EmailSyncMeta;
const { account, userEmail } = meta;
// Resolve auth (refreshes OAuth token if needed)
const imapAuth = await resolveImapAuth(meta);
// Always re-read the position rather than trusting job meta, which can be stale after a retry.
const freshAccount = await getEmailAccount(account.id);
if (!freshAccount) throw new PermanentError(`Email account ${account.id} not found`);
// Opened here rather than further down because the position now lives inside it. Postgres is
// passed only as the one-time backfill for accounts that predate the move — see sync-meta.ts.
const db = openEmailDb(userEmail, account.email);
const syncMeta: Record<string, unknown> = readSyncMeta(db, freshAccount.syncMeta as Record<string, unknown>);
const isIncremental = !!syncMeta.last_sync_at;
await updateEmailAccountStatus(account.id, 'syncing');
console.log(`[email-sync] ${isIncremental ? 'Incremental' : 'Initial'} sync for ${account.email}`);
const MAX_RECONNECTS = 10;
const RECONNECT_DELAY_MS = 5_000;
let reconnects = 0;
let saved = 0;
let skipped = 0;
let errors = 0;
let allDone = false;
// Load existing IDs for dedup (once, shared across reconnections)
const existingIds = new Set<string>();
const rows = db.query('SELECT id FROM emails').all() as Array<{ id: string }>;
for (const row of rows) existingIds.add(row.id);
try {
while (!allDone && reconnects <= MAX_RECONNECTS) {
if (reconnects > 0) {
console.log(
`[email-sync] Reconnecting (${reconnects}/${MAX_RECONNECTS}) after ${RECONNECT_DELAY_MS / 1000}s...`,
);
await ctx.updateProgress({
current: saved + skipped,
total: 0,
label: `Reconnecting (${reconnects}/${MAX_RECONNECTS})...`,
});
await new Promise((r) => setTimeout(r, RECONNECT_DELAY_MS));
// Re-read syncMeta from DB to get latest saved UIDs
const updated = await getEmailAccount(account.id);
if (updated?.syncMeta) {
Object.assign(syncMeta, updated.syncMeta as Record<string, unknown>);
}
} else {
await ctx.updateProgress({ current: 0, total: 0, label: 'Connecting...' });
}
const client = new ImapFlow({
host: account.imapHost,
port: account.imapPort,
secure: account.imapSecure,
auth: imapAuth,
logger: false,
socketTimeout: 30 * 60 * 1000,
});
const connState = { error: null as Error | null };
client.on('error', (err: Error) => {
console.log(`[email-sync] IMAP connection error: ${err.message}`);
connState.error = err;
});
try {
await client.connect();
console.log(`[email-sync] IMAP connected${reconnects > 0 ? ` (reconnect ${reconnects})` : ''}`);
const folders = (await client.list({
statusQuery: { uidNext: true, uidValidity: true },
})) as FolderInfo[];
// Filter to folders that still need syncing
const foldersToSync: Array<{ folder: FolderInfo; lastUid: number }> = [];
for (const folder of folders) {
if (shouldSkipFolder(folder)) continue;
const uidValidityKey = `imap_uidvalidity:${folder.path}`;
const lastUidKey = `imap_lastuid:${folder.path}`;
const storedUidValidity = syncMeta[uidValidityKey] as string | undefined;
const storedLastUid = syncMeta[lastUidKey] as string | undefined;
const uidValidity = folder.status?.uidValidity ? String(folder.status.uidValidity) : null;
const uidNext = folder.status?.uidNext ?? 0;
let lastUid =
storedUidValidity && uidValidity === storedUidValidity && storedLastUid
? parseInt(storedLastUid, 10)
: 0;
if (storedUidValidity && uidValidity && storedUidValidity !== uidValidity) {
console.log(`[email-sync] UIDVALIDITY changed for ${folder.path} — will re-scan`);
lastUid = 0;
}
if (lastUid > 0 && uidNext <= lastUid + 1) continue;
foldersToSync.push({ folder, lastUid });
}
if (foldersToSync.length === 0) {
console.log('[email-sync] All folders synced');
allDone = true;
await client.logout().catch(() => {});
break;
}
console.log(`[email-sync] ${foldersToSync.length} folder(s) to sync`);
let connectionLost = false;
for (let fi = 0; fi < foldersToSync.length; fi++) {
const { folder, lastUid } = foldersToSync[fi]!;
if (connState.error) {
connectionLost = true;
break;
}
console.log(`[email-sync] Syncing folder ${fi + 1}/${foldersToSync.length}: ${folder.path}`);
let lock;
try {
lock = await client.getMailboxLock(folder.path);
} catch (err) {
const errMsg = err instanceof Error ? err.message : String(err);
if (errMsg.includes('not available') || errMsg.includes('timeout') || errMsg.includes('closed')) {
console.log(`[email-sync] Connection lost at folder ${folder.path}`);
connectionLost = true;
break;
}
console.log(`[email-sync] Skipping folder ${folder.path}: ${errMsg}`);
continue;
}
try {
const mailbox = client.mailbox;
if (!mailbox) continue;
const uidValidity = String(mailbox.uidValidity);
const effectiveLastUid = syncMeta[`imap_uidvalidity:${folder.path}`] === uidValidity ? lastUid : 0;
let maxUid = effectiveLastUid;
const label = folderToLabel(folder);
const range = effectiveLastUid > 0 ? `${effectiveLastUid + 1}:*` : '1:*';
try {
const fetchOpts: Record<string, unknown> = { source: true, uid: true };
if (account.provider === 'gmail') fetchOpts.labels = true;
for await (const msg of client.fetch(range, fetchOpts, { uid: true })) {
if (msg.uid <= effectiveLastUid) continue;
maxUid = Math.max(maxUid, msg.uid);
if (!msg.source) {
errors++;
continue;
}
const raw = msg.source.toString('utf-8');
const id = messageIdToStableId(raw);
if (!id) {
errors++;
continue;
}
if (existingIds.has(id)) {
skipped++;
continue;
}
const labels = [label];
try {
upsertFromRawEml({
db,
id,
raw,
integration: account.provider,
emailAccount: account.email,
labels,
});
existingIds.add(id);
saved++;
} catch {
errors++;
}
if ((saved + skipped) % 50 === 0) {
const progressLabel = `${folder.path}${saved.toLocaleString()} saved, ${skipped.toLocaleString()} skipped`;
console.log(`[email-sync] ${progressLabel}`);
await ctx.updateProgress({ current: saved + skipped, total: 0, label: progressLabel });
if (maxUid > effectiveLastUid) {
syncMeta[`imap_lastuid:${folder.path}`] = String(maxUid);
writeSyncMeta(db, syncMeta);
}
}
}
} catch (fetchErr) {
const errMsg = fetchErr instanceof Error ? fetchErr.message : String(fetchErr);
if (errMsg.includes('Nothing to fetch')) {
// No messages in range — normal
} else if (
errMsg.includes('not available') ||
errMsg.includes('timeout') ||
errMsg.includes('closed')
) {
console.log(`[email-sync] Connection lost during fetch in ${folder.path}: ${errMsg}`);
connectionLost = true;
} else {
console.log(`[email-sync] Fetch error in ${folder.path}: ${errMsg}`);
errors++;
}
}
syncMeta[`imap_uidvalidity:${folder.path}`] = uidValidity;
if (maxUid > effectiveLastUid) {
syncMeta[`imap_lastuid:${folder.path}`] = String(maxUid);
}
console.log(`[email-sync] ${folder.path} done: ${saved} saved, ${skipped} skipped, ${errors} errors`);
await ctx.updateProgress({
current: saved + skipped,
total: 0,
label: `${folder.path}: ${saved} saved`,
});
writeSyncMeta(db, syncMeta);
} finally {
lock.release();
}
if (connectionLost) break;
}
await client.logout().catch(() => {});
if (connectionLost) {
console.log(`[email-sync] Connection lost after saving ${saved} emails — will reconnect`);
reconnects++;
continue;
}
allDone = true;
} catch (err) {
await client.logout().catch(() => {});
if (reconnects < MAX_RECONNECTS) {
console.log(`[email-sync] Error: ${err instanceof Error ? err.message : String(err)} — will reconnect`);
reconnects++;
continue;
}
throw err;
}
}
if (!allDone) {
throw new Error(`IMAP sync incomplete after ${MAX_RECONNECTS} reconnection attempts`);
}
// Mark sync complete. Last, and only after every message above is stored: if this never runs,
// the position stays behind and those messages are fetched again. Re-fetching is wasteful;
// advancing past mail that was never stored would lose it silently.
syncMeta.last_sync_at = new Date().toISOString();
writeSyncMeta(db, syncMeta);
} finally {
db.close();
}
console.log(
`[email-sync] Done: saved ${saved}, skipped ${skipped}, errors ${errors}, reconnects ${reconnects}`,
);
ctx.meta.saved = saved;
await ctx.updateProgress({ current: 1, total: 1, label: `Done — ${saved} new emails` });
},
},
{
name: 'Finalize',
run: async (ctx: SyncCtx) => {
const meta = ctx.meta as unknown as EmailSyncMeta;
const saved = meta.saved ?? 0;
await updateEmailAccountStatus(meta.account.id, 'synced');
if (saved > 0) {
try {
const paths = await getDockPaths(meta.account.userId);
if (!paths) {
const defaults = ['/', '/files', '/automation', '/dashboards', '/chat'];
await setDockPaths(meta.account.userId, [...defaults, '/email']);
} else if (!paths.includes('/email')) {
await setDockPaths(meta.account.userId, [...paths, '/email']);
}
} catch {
// Non-fatal
}
}
console.log(`[email-sync] ${meta.account.email} status set to synced`);
},
},
],
};
/** The handler's steps, for the sidecar's own runner. */
export const imapSyncSteps: SyncStep[] = emailSyncHandler.steps as SyncStep[];