mirror of
https://github.com/kuhyx/todo-app.git
synced 2026-11-11 20:34:00 +01:00
The desktop app is a browser now, so it cannot write files. This adds the other half: a loopback HTTP server that serves the web build and owns the two files the browser cannot touch -- `~/todo/BACKLOG.md`, which the user's tooling and the todo MCP server read, and a disk copy of the note log so a wiped Chrome profile is not a total-loss event for unsynced notes. Notable details: - Binds to 127.0.0.1 only. The endpoints overwrite files in the user's home with no authentication, so a routable bind would let anything on the network rewrite the backlog. - Serves explicit content types. CanvasKit refuses a `.wasm` labelled anything else and the app then renders nothing at all. - Mirroring to disk is debounced. The store persists on every keystroke, so mirroring eagerly meant an HTTP request plus a whole-log write per character; IndexedDB still gets every keystroke, only the durability copy waits. Separately, this fixes a pre-existing destructive bug found while testing: `settings_screen_test` did real file I/O and overwrote the user's canonical `~/todo/BACKLOG.md` with a single fake note on *every* `flutter test` run. It went unnoticed because the app rewrites the file whenever it launches. The export destination is now injectable and the test redirects it to a temp dir; verified that a full suite run leaves the real file byte-identical. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
95 lines
3.3 KiB
Dart
95 lines
3.3 KiB
Dart
import 'dart:async';
|
|
|
|
import 'package:crdt_sync/crdt_sync.dart';
|
|
import 'package:idb_shim/idb.dart';
|
|
import 'package:todo/data/desktop_backup_client.dart';
|
|
|
|
/// [LogPersistence] backed by IndexedDB, with a best-effort copy pushed to the
|
|
/// desktop wrapper so the notes also exist on disk.
|
|
///
|
|
/// IndexedDB rather than `localStorage` because the latter caps at roughly
|
|
/// 5-10MB per origin and is evicted more eagerly; the note log is the primary
|
|
/// copy of the backlog and has to outlive storage pressure.
|
|
///
|
|
/// The disk copy exists because moving the desktop app into a browser would
|
|
/// otherwise leave a wiped Chrome profile as a total-loss event for notes that
|
|
/// have not synced to GitHub yet. It is deliberately fire-and-forget: the
|
|
/// wrapper is an optimisation, and the app must stay fully usable when it is
|
|
/// absent (for instance when the same build is opened in a plain browser tab).
|
|
class WebLogPersistence implements LogPersistence {
|
|
/// Creates a persistence over [database], mirroring writes to [backup].
|
|
WebLogPersistence({
|
|
required Database database,
|
|
this.backup,
|
|
this.mirrorDebounce = const Duration(seconds: 2),
|
|
}) : _db = database;
|
|
|
|
final Database _db;
|
|
|
|
/// Wrapper client that mirrors writes to disk, or null when absent.
|
|
final DesktopBackupClient? backup;
|
|
|
|
/// How long to coalesce disk mirroring after the last write.
|
|
///
|
|
/// The store persists on *every* keystroke, so mirroring eagerly would mean
|
|
/// an HTTP request and a whole-log disk write per character. IndexedDB is
|
|
/// still written synchronously with each keystroke; only the durability copy
|
|
/// is debounced.
|
|
final Duration mirrorDebounce;
|
|
|
|
Timer? _mirrorTimer;
|
|
|
|
/// Object store holding a single record: the serialised log.
|
|
static const storeName = 'log';
|
|
|
|
/// Key of that single record.
|
|
static const recordKey = 'notes';
|
|
|
|
/// IndexedDB database name.
|
|
static const databaseName = 'todo';
|
|
|
|
/// Opens (creating if needed) the IndexedDB database backing the log.
|
|
static Future<Database> openDatabase(IdbFactory factory) => factory.open(
|
|
databaseName,
|
|
version: 1,
|
|
onUpgradeNeeded: (event) {
|
|
final db = event.database;
|
|
if (!db.objectStoreNames.contains(storeName)) {
|
|
db.createObjectStore(storeName);
|
|
}
|
|
},
|
|
);
|
|
|
|
@override
|
|
Future<String?> read() async {
|
|
final txn = _db.transaction(storeName, idbModeReadOnly);
|
|
final value = await txn.objectStore(storeName).getObject(recordKey);
|
|
await txn.completed;
|
|
if (value is String && value.isNotEmpty) return value;
|
|
// Nothing local yet (fresh profile, or the browser evicted the database).
|
|
// Fall back to the wrapper's on-disk copy so a cleared profile recovers
|
|
// instead of silently starting from an empty backlog.
|
|
return backup?.readLog();
|
|
}
|
|
|
|
@override
|
|
Future<void> write(String text) async {
|
|
final txn = _db.transaction(storeName, idbModeReadWrite);
|
|
await txn.objectStore(storeName).put(text, recordKey);
|
|
await txn.completed;
|
|
_scheduleMirror(text);
|
|
}
|
|
|
|
void _scheduleMirror(String text) {
|
|
final client = backup;
|
|
if (client == null) return;
|
|
_mirrorTimer?.cancel();
|
|
_mirrorTimer = Timer(mirrorDebounce, () {
|
|
unawaited(client.writeLog(text));
|
|
});
|
|
}
|
|
|
|
/// Cancels any pending disk mirror. Call when tearing the store down.
|
|
void dispose() => _mirrorTimer?.cancel();
|
|
}
|