mirror of
https://github.com/kuhyx/diet-guard.git
synced 2026-11-11 19:33:55 +01:00
Krzysztof kuhy Rudnicki
c12faa3920
Completes the move: `linux/` is gone, replaced by the web build served by the
wrapper in a Chrome `--app` window. The icons and the .desktop installer move
out of `linux/` first -- they are not Flutter scaffolding and are still
needed; the entry now points at the packaged launcher and matches the
`--class=diet_guard_app` the wrapper passes, so the taskbar shows the app's
icon rather than a browser's.
install_arch.sh carries over the three traps each of which produced an
install that looked fine until launched: `dart build cli` rather than
`dart compile exe` (which refuses any package whose dependency graph has
native build hooks), `options=('!strip')` (makepkg would otherwise discard
the AOT snapshot, leaving a binary that runs as the bare Dart VM printing a
usage message), and no browser in `depends` (this machine's policy uninstalls
`chromium` seconds later, failing dependency resolution).
Also adds the frame-cadence instrument the performance claim is measured
with, armed by `--dart-define=DIET_GUARD_FRAME_STATS=1` and read with
`String.fromEnvironment` -- `bool.fromEnvironment` only accepts the literal
`true`, so `=1` would evaluate to false and a whole run would silently
produce no data. It reports frame-to-frame cadence rather than build/raster
durations, since work that fits the budget says nothing about the frame rate
actually achieved.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
76 lines
2.5 KiB
Dart
76 lines
2.5 KiB
Dart
/// Optional frame-cadence instrument, armed with
|
|
/// `--dart-define=DIET_GUARD_FRAME_STATS=1`.
|
|
///
|
|
/// Reports **frame-to-frame cadence**, not build/raster durations: work that
|
|
/// fits the budget says nothing about the frame rate actually achieved, which
|
|
/// is the number the desktop migration exists to move.
|
|
///
|
|
/// Read with `String.fromEnvironment`, never `bool.fromEnvironment` -- the
|
|
/// latter only accepts the literal `true`, so `=1` would silently evaluate to
|
|
/// false and a whole measurement run would produce no data while looking like
|
|
/// a broken callback.
|
|
library;
|
|
|
|
import 'dart:async';
|
|
import 'dart:convert';
|
|
import 'dart:ui' show FramePhase;
|
|
|
|
import 'package:diet_guard_app/services/desktop_wrapper.dart';
|
|
import 'package:flutter/scheduler.dart';
|
|
import 'package:http/http.dart' as http;
|
|
|
|
/// Whether the instrument is armed for this build.
|
|
const frameStatsEnabled =
|
|
String.fromEnvironment('DIET_GUARD_FRAME_STATS') == '1';
|
|
|
|
/// Starts collecting frame-to-frame intervals and periodically posts a
|
|
/// summary to the desktop wrapper, which writes it to disk.
|
|
///
|
|
/// The wrapper is used as the sink because a browser cannot write a file and
|
|
/// reading its console from outside needs a debugger attached; a POST lands
|
|
/// somewhere a measurement run can simply read.
|
|
// coverage:ignore-start
|
|
void startFrameStats({Duration reportEvery = const Duration(seconds: 2)}) {
|
|
final intervalsMs = <double>[];
|
|
int? previousMicros;
|
|
|
|
SchedulerBinding.instance.addTimingsCallback((timings) {
|
|
for (final timing in timings) {
|
|
final micros = timing.timestampInMicroseconds(FramePhase.rasterFinish);
|
|
if (previousMicros != null) {
|
|
intervalsMs.add((micros - previousMicros!) / 1000);
|
|
}
|
|
previousMicros = micros;
|
|
}
|
|
});
|
|
|
|
Timer.periodic(reportEvery, (_) {
|
|
if (intervalsMs.isEmpty) return;
|
|
final sample = [...intervalsMs]..sort();
|
|
intervalsMs.clear();
|
|
double percentile(double fraction) =>
|
|
sample[(sample.length * fraction).floor().clamp(
|
|
0,
|
|
sample.length - 1,
|
|
)];
|
|
unawaited(
|
|
http.post(
|
|
Uri.parse(
|
|
'$desktopWrapperOrigin${WrapperPaths.documents}'
|
|
'frame_stats.json',
|
|
),
|
|
body: jsonEncode({
|
|
'frames': sample.length,
|
|
'p50_ms': percentile(0.50),
|
|
'p95_ms': percentile(0.95),
|
|
'max_ms': sample.last,
|
|
'fps_at_p50': 1000 / percentile(0.50),
|
|
'fps_at_p95': 1000 / percentile(0.95),
|
|
}),
|
|
),
|
|
);
|
|
});
|
|
}
|
|
|
|
// coverage:ignore-end
|