Compare commits

...
Author SHA1 Message Date
poli0981andClaude Opus 5.5 0a36f54fb1 docs(handoff): the beta.18 traps name the Vulkan launch e2e race that failed #288's CI
On an elevated runner the harness starts through CreateProcessWithTokenW, its "[SKIP]" line never reaches the host's
output, and a guard scan that hits the exiting harness (ModuleScanFailed) fails the test instead of skipping it.
Re-run; the deterministic fix is a task.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-07 20:37:54 +07:00
poli0981andClaude Opus 5.5 201a89d23b release: 0.1.0-beta.18 - the CHANGELOG section with how to update from beta.17, the accuracy block's release line, README's release lines, LIMITATIONS' version, HANDOFF's beta.18 section and owner-only item 14 (beta.18 PR-4)
VERSION stays 0.1.0: the tag's numeric core. The owner (2026-10-07) asked for beta.18 - the two chips beta.17 left and
the page text that blends into the dark window; the train follows beta.17's: merge on green CI, then the signed tag
v0.1.0-beta.18 on the merge commit. wiki.yml publishes the wiki after the Release run succeeds.

- CHANGELOG: [Unreleased] becomes [0.1.0-beta.18] - 2026-10-07, with its summary and "Updating from 0.1.0-beta.17";
  tools/release-notes.ps1 -Version 0.1.0-beta.18 extracts it (3948 characters).
- The accuracy block's release line (legal/ACCURACY.md, embedded in README and the Disclaimer; accuracy-check green),
  README's release lines and count, LIMITATIONS' version.
- HANDOFF: beta.17's START HERE struck; the beta.18 section - no new decisions; the traps (a page in the NavigationView
  inherits nothing; the accent is not in the theme dictionary; the gate's own refusals look like findings on the wire;
  Safety_Consent_* strings are versioned with the disclosure; a release's natives were -dirty by its own hand; the
  CaptureHost e2e flake met again); owner-only item 14.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-07 20:17:17 +07:00
poli0981andClaude Opus 5.5 a6fad3c05c fix(release): the natives carry the clean checkout's build id - read before the date substitution, handed to CMake, read back out of the binaries (beta.18 PR-3)
The chip from beta.17, open since beta.4 (HANDOFF §0.1.0-beta.4, "Found and NOT fixed"): the owner's Dashboard read
"Overlay build v0.1.0-beta.17-dirty", and beta.16's the same. The release run's own log says why:
"RELEASE_DATE = 2026-10-07 in 3 file(s)" and, 22 s later, "FL_BUILD_ID = v0.1.0-beta.17-dirty". release.yml writes the
release date into the tracked legal/*.md before the gate, and CMake's configure reads `git describe --dirty` after it.
CI's own checkout is clean (main's last run: "FL_BUILD_ID = 15b4c6b0ecc1").

- release.yml: a step before the substitution reads describe from the clean checkout (12 digits, as CMake does),
  refuses one that is already dirty or longer than 31 characters, and exports FL_BUILD_ID. After the publish,
  tools/buildid-check.ps1 reads it back out of the four natives that compile it.
- src/native/FrameLedger.Shm/CMakeLists.txt: the environment's FL_BUILD_ID is taken as given, after a check for the
  characters a C string literal carries without escaping; otherwise describe as before. Its comment said a released
  build was exact - "CI configures and builds once from a clean checkout" - corrected in place.
- tools/buildid-check.ps1 (new): looks for the id as the C string it is compiled as (characters + NUL) in each
  binary's bytes; nothing is loaded - loading the Overlay would run its DllMain. -SelfTest, eight cases, both
  directions, runs in the gate (build.ps1 'buildid-check').

Measured here: ./build.ps1 native with FL_BUILD_ID=b18-envtest.1+x_y printed "FL_BUILD_ID = b18-envtest.1+x_y", and
buildid-check over the four built DLLs: "ok - 4 native binaries carry 'b18-envtest.1+x_y'"; with FL_BUILD_ID='bad "id"'
the configure stopped ("holds a character no build id has"). On this branch's own describe (dirty, "+"), the tool
refused the id itself and found it in all four binaries.

Docs: 12_BUILD (§Release-time token substitution; gate 14c), 13_CI_CD (§release.yml), HANDOFF (§0.1.0-beta.4's bullet
struck, dated), CLAUDE.md (tools), wiki Releasing, CHANGELOG [Unreleased].

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-07 20:17:17 +07:00
14 changed files with 279 additions and 38 deletions

No files matched your search

+19
View File
@@ -116,6 +116,22 @@ jobs:
"dry_run=$($dry.ToString().ToLowerInvariant())" >> $env:GITHUB_OUTPUT
Write-Host "$(if ($dry) { 'DRY RUN of' } else { 'releasing' }) $full (pre-release: $pre)"
# The Overlay's build id is `git describe --dirty` at CMake's configure
# (src/native/FrameLedger.Shm/CMakeLists.txt), and the next step edits tracked
# files: every release until beta.18 shipped "<tag>-dirty". So the id is read
# here, from the clean checkout, and handed to CMake through the environment;
# buildid-check reads it back out of the published binaries.
- name: Build id (the clean checkout)
shell: pwsh
run: |
$ErrorActionPreference = 'Stop'
$id = "$(git describe --always --dirty --abbrev=12)".Trim()
if ($LASTEXITCODE -ne 0 -or -not $id) { throw 'git describe failed' }
if ($id.EndsWith('-dirty')) { throw "the checkout is dirty before the release has touched it: $id" }
if ($id.Length -gt 31) { throw "build id '$id' is $($id.Length) characters; FlShmHandshake::buildId holds 31" }
"FL_BUILD_ID=$id" >> $env:GITHUB_ENV
Write-Host "FL_BUILD_ID = $id"
# 12_BUILD §Release-time token substitution: the effective date of a legal
# document is the date it ships. The only placeholder that survives into the
# repository, substituted here and nowhere else; the App embeds legal/*.md at
@@ -200,6 +216,9 @@ jobs:
# beta.14 (D50): the managed half too - the eight managed binaries, carrying the tag's full version.
./tools/versioninfo-check.ps1 -BuildDir out/app -ManagedDir out/app -ExpectedProductVersion '${{ steps.version.outputs.version }}'
if ($LASTEXITCODE -ne 0) { throw 'versioninfo-check over the published tree' }
# beta.18: the native binaries carry the build id read from the clean checkout, exactly - never "<tag>-dirty".
./tools/buildid-check.ps1 -BuildDir out/app -Expected $env:FL_BUILD_ID
if ($LASTEXITCODE -ne 0) { throw 'buildid-check over the published tree' }
$size = (Get-ChildItem out/app -Recurse -File | Measure-Object Length -Sum).Sum / 1MB
Write-Host ("published tree: {0:N0} MB" -f $size)
+14
View File
@@ -28,6 +28,15 @@ numeric core and does not move; the heading is the full version — corrected 20
## [Unreleased]
## [0.1.0-beta.18] - 2026-10-07
**The eighteenth pre-release: every page readable in the dark theme, notices that say what happened in words, and a
release whose build no longer calls itself dirty.** The same unsigned installer: verify its hash against
`SHA256SUMS.txt`, then *More info → Run anyway*.
**Updating from `0.1.0-beta.17`:** Help ▸ Check for updates, then *Restart to update* with no game running. Nothing in
the database changes and nothing is shown again for you to accept; sessions already stored keep their status.
### Fixed
- **Page titles and section headings read in the dark theme again.** Every page's title and its section headings — Dashboard,
@@ -54,6 +63,11 @@ numeric core and does not move; the heading is the full version — corrected 20
- **A viewer over a copy leaves every agent setting alone.** The user-mode exception option, the three Recording numbers
and the update channel stayed live in `--data-dir`, where no agent reads them (`ViewerModeTests`).
- `docs/08_UI.md` listed a "PawnIO missing" banner that was never built; struck, with the banners that do exist.
- **A release's overlay build no longer calls itself dirty.** The Dashboard read *Overlay build v0.1.0-beta.17-dirty* (and
beta.16's, and every release's before): the release workflow writes the release date into the legal documents before
the build, and the build id was read from the tree after that. It is read from the clean checkout first and handed to
the build, and `tools/buildid-check.ps1` reads it back out of the published binaries (`docs/12_BUILD.md` §Release-time
token substitution). Nothing changes for a game: both sides of the ring's handshake always carried the same id.
## [0.1.0-beta.17] - 2026-10-07
+2 -1
View File
@@ -139,7 +139,8 @@ tools/ # accuracy-check, changelog-check, chokepoint-che
# (+ license-overrides/), notice-check, package-closure-check,
# release-notes, rules-validate, test-artifacts-check, vendor-exports,
# versioninfo-check, vklayer-blastradius, fps-impact-runbook,
# wiki-build (beta.17: builds and checks the wiki; the wiki-check gate)
# wiki-build (beta.17: builds and checks the wiki; the wiki-check gate),
# buildid-check (beta.18: the release's build id in the shipped natives)
# (six names added 2026-10-04)
# (PowerShell). This line used to name three, one of which
# — resx-audit — ~~does not exist~~ exists since 2026-09-13 with
+1 -1
View File
@@ -5,7 +5,7 @@ hooking on for — and reading the graphics calls the game makes. That is why it
resolution, the upscaler, frame generation and ray tracing apart. It is also where its limits come from. This
page lists them in plain words; each item links to the document that has the details.
*Beta software: this page describes version 0.1.0-beta.17.*
*Beta software: this page describes version 0.1.0-beta.18.*
## Games it will not measure
+3 -3
View File
@@ -12,7 +12,7 @@ pages.
<!-- accuracy-block:begin -->
> ⚠ **What FrameLedger actually measures today — 2026-10-06.** The software is a beta; its latest
> **pre-release is `0.1.0-beta.17`** (2026-10-07), an unsigned installer built from that tag with its
> **pre-release is `0.1.0-beta.18`** (2026-10-07), an unsigned installer built from that tag with its
> checksums published beside it. The source holds the desktop app
> (library, store import, charts, settings) and the background Agent, which records a session when a
> game in the library runs (unless you switched its recording off), injects only into games you enabled
@@ -163,7 +163,7 @@ Elevation is **optional — for everything.** Hooked capture is the normal path
## Install
> **The latest pre-release is `v0.1.0-beta.17` (2026-10-07); before it `v0.1.0-beta.16` (2026-10-06), `v0.1.0-beta.15` (2026-10-06), `v0.1.0-beta.14` (2026-10-06), `v0.1.0-beta.13` (2026-10-04), `v0.1.0-beta.12` (2026-10-04), `v0.1.0-beta.11` (2026-10-03), `v0.1.0-beta.10` (2026-09-28), `v0.1.0-beta.9` (2026-09-26), `v0.1.0-beta.8` (2026-09-25), `v0.1.0-beta.7` (2026-09-23), `v0.1.0-beta.6` (2026-09-23), `v0.1.0-beta.5` (2026-09-22), `v0.1.0-beta.4` (2026-09-21), `v0.1.0-beta.3` (2026-09-21), `v0.1.0-beta.2` (2026-09-17) and `v0.1.0-beta.1` (2026-09-16).** They are pre-releases: read `CHANGELOG.md`'s section for the one you install — what it measures, what it does not yet, and how to update — before installing. A source build is `docs/12_BUILD.md`.
> **The latest pre-release is `v0.1.0-beta.18` (2026-10-07); before it `v0.1.0-beta.17` (2026-10-07), `v0.1.0-beta.16` (2026-10-06), `v0.1.0-beta.15` (2026-10-06), `v0.1.0-beta.14` (2026-10-06), `v0.1.0-beta.13` (2026-10-04), `v0.1.0-beta.12` (2026-10-04), `v0.1.0-beta.11` (2026-10-03), `v0.1.0-beta.10` (2026-09-28), `v0.1.0-beta.9` (2026-09-26), `v0.1.0-beta.8` (2026-09-25), `v0.1.0-beta.7` (2026-09-23), `v0.1.0-beta.6` (2026-09-23), `v0.1.0-beta.5` (2026-09-22), `v0.1.0-beta.4` (2026-09-21), `v0.1.0-beta.3` (2026-09-21), `v0.1.0-beta.2` (2026-09-17) and `v0.1.0-beta.1` (2026-09-16).** They are pre-releases: read `CHANGELOG.md`'s section for the one you install — what it measures, what it does not yet, and how to update — before installing. A source build is `docs/12_BUILD.md`.
1. Download the latest `FrameLedger.App-win-Setup.exe` from [Releases](https://github.com/poli0981/frameledger/releases). It installs into `%LOCALAPPDATA%\FrameLedger.App`; your data stays in `%LOCALAPPDATA%\FrameLedger`, and uninstalling asks before touching it.
2. SmartScreen may warn — releases are not code-signed (free, open-source project). Verify the SHA-256 checksum published with each release, then **More info → Run anyway**.
@@ -212,4 +212,4 @@ GPU telemetry is layered so the project never depends on a proprietary vendor li
---
**Status:** pre-release (beta). Seventeen tagged pre-releases so far; see `CHANGELOG.md` for what each one changed and the accuracy block at the top of this file for what is and is not measured. It is not feature-complete, it certainly still has bugs, and `legal/DISCLAIMER.md` §0 says what that means for you.
**Status:** pre-release (beta). Eighteen tagged pre-releases so far; see `CHANGELOG.md` for what each one changed and the accuracy block at the top of this file for what is and is not measured. It is not feature-complete, it certainly still has bugs, and `legal/DISCLAIMER.md` §0 says what that means for you.
+6
View File
@@ -488,6 +488,12 @@ function Invoke-ProjectGates {
Write-Step 'release-notes'
Invoke-Checked 'release-notes (self-test)' { & (Join-Path $repo 'tools/release-notes.ps1') -SelfTest }
# beta.18: the build id a release hands CMake (read from the clean checkout before the date substitution edits
# legal/*.md) is read back out of the published native binaries by release.yml. Self-test only here — the live
# pass needs that id, which only a release has (12_BUILD §Release-time token substitution).
Write-Step 'buildid-check'
Invoke-Checked 'buildid-check (self-test)' { & (Join-Path $repo 'tools/buildid-check.ps1') -SelfTest }
# 09_I18N's gate, built with the first .resx family (P3 PR-2, 2026-09-13): every key in en/vi/ja, ja
# Safety_* marked for human review, the generated accessor current. Self-test first (at least four
# cases must go red), then the live pass; a tree with no family is red, not skipped.
+11
View File
@@ -272,6 +272,16 @@ substitutes it with ~~the tag date~~ the run's UTC date (corrected 2026-10-04) w
address — is resolved in the source files, because an unsubstituted token in a
document the app displays for acceptance (FR-11) is a defect, not a template.
**The build id is read before it (beta.18).** The substitution edits tracked files, and the native build id is
`git describe --dirty` at CMake's configure (`src/native/FrameLedger.Shm/CMakeLists.txt`): every release until beta.18
shipped `<tag>-dirty` — the owner's Dashboard read *Overlay build v0.1.0-beta.17-dirty*, and that run's log printed
`FL_BUILD_ID = v0.1.0-beta.17-dirty` two lines after `RELEASE_DATE = 2026-10-07 in 3 file(s)`. `release.yml` now reads
describe from the clean checkout first, refuses one that is already dirty or longer than 31 characters, and passes it
as the environment's `FL_BUILD_ID`, which CMake takes as given (checked for the characters a C string literal carries);
after the publish, `tools/buildid-check.ps1` reads it back out of the four native binaries that compile it, by their
bytes — nothing is loaded. A developer's build still runs describe itself (`+` marks a dirty tree when the long form
would not fit).
## Local quality gate (pre-push)
`./build.ps1 check`:
@@ -301,6 +311,7 @@ Seven self-test cases, both directions; `./tools/notice-check.ps1 -Fix` writes a
13d. **`tools/wiki-build.ps1 -SelfTest`, then `-Check`** (beta.17, owner decisions D53/D54, 2026-10-07) — assembles the GitHub Wiki from `wiki/`, `guide/` and `LIMITATIONS.md` into a temporary folder, as `wiki.yml` does at a release (`13_CI_CD` §wiki.yml), and fails on a link to a file, folder or section that does not exist, two sources for one page, a page that neither Home nor the sidebar leads to, a sidebar that lists the guide in another order than `guide/README.md`, a backticked repository path in a `wiki/` page that names nothing, and a version number written into a `wiki/` page (the footer says which release the wiki shows). Nine self-test cases, both directions
14. `tools/changelog-check.ps1 -SelfTest` — nine cases, five expected RED. The live half needs a pull request's changed-file list and is supplied by `ci.yml`
14b. `tools/release-notes.ps1 -SelfTest` — five cases over a fixture changelog (a dated and an undated section, a missing one, a prefix that must not match, an empty one). The live half needs a tag's version and runs in `release.yml`, where a missing or empty section is a red release (P4 PR-5)
14c. **`tools/buildid-check.ps1 -SelfTest`** (beta.18) — eight cases, both directions: a binary built from a dirty tree (`-dirty`, or the short form's `+`), one with no id, a missing one, and an expected id that is dirty or not an id are red. The live half needs the release's clean-checkout id and runs in `release.yml` over the published tree (§Release-time token substitution)
15. `tools/resx-audit.ps1` — ~~**skipped loudly; it does not exist, and no `.resx` file does either**~~ **built 2026-09-13 (P3 PR-2)**: `-SelfTest` first (11 fixture cases, both directions), then the live pass over every `Strings.resx` family under `src/` — key sets equal across en/vi/ja, `Safety_*` in `ja` marked for review or signed, `Strings.Designer.cs` current with `tools/resx-gen.ps1`. A tree with no family is red (`09_I18N` §Translation workflow)
16. **struct-mirror** — reads this run's `.trx` and fails when `ShmLayoutMirrorTests` did not execute, so deleting the mirror test is red as well as breaking it
16b. **`tools/test-artifacts-check.ps1`** — `-SelfTest` first (seven cases, two RED), then live: an `overlay-*.log` created since this gate started, still in the real `%LOCALAPPDATA%\FrameLedger\logs` after the tests, whose first line names a `hook-harness*` image, fails the gate. It is what makes a test binary without its log sweep red (`17_HOOK_ENGINE` §Native logging, 2026-09-15); a game's log never counts
+4
View File
@@ -127,6 +127,10 @@ FrameLedger uses the **`poli0981/.github` ops repo** where its templates fit, an
`VERSION` file (`12_BUILD` §Version); and `CHANGELOG.md` must carry a **non-empty** `## [x.y.z]` section for the
full version, extracted by `tools/release-notes.ps1` (self-tested on every `build.ps1 check`). A missing
section is a red release, not an empty note — the ledger's header said the opposite for five weeks.
- **The build id from the clean checkout** (beta.18): `git describe --always --dirty --abbrev=12` before anything edits a
tracked file — a dirty one or one over 31 characters stops the run — into `FL_BUILD_ID` for CMake. Until beta.18 the
substitution below ran first and every release's natives called themselves `<tag>-dirty` (`12_BUILD` §Release-time
token substitution).
- `{{RELEASE_DATE}}` substituted in `legal/*.md` with the run's UTC date **before the build**, because the App
embeds those documents (FR-11); any `{{` token left anywhere after that fails the job.
- **The identical gate** — `./build.ps1 check`, native first, every tool including `versioninfo-check` against
+49 -3
View File
@@ -1410,9 +1410,11 @@ stops there"~~ — the override itself is gone since 2026-09-22 (D22).
**Found and NOT fixed — each needs its own PR.**
- ~~**A refused session is not recorded, though the notice says it is.**~~ **Fixed 2026-09-22 (D23):** the loop holds a
refused session open until the target exits, so Tier 2 has a producer and the sentence is true for every refusal.
- **A released Overlay logs its build as `v0.1.0-beta.3-dirty`**: the CI checkout is dirty by the time CMake runs
- ~~**A released Overlay logs its build as `v0.1.0-beta.3-dirty`**: the CI checkout is dirty by the time CMake runs
`git describe`. Both sides of the handshake compile the same string, so nothing breaks; it is a release that calls
itself dirty.
itself dirty.~~ **Fixed 2026-10-07 (beta.18 PR-3):** the dirt was `release.yml`'s own date substitution in
`legal/*.md`; the workflow reads describe from the clean checkout before it and hands it to CMake as `FL_BUILD_ID`,
and `tools/buildid-check.ps1` reads it back out of the published binaries (`12_BUILD` §Release-time token substitution).
- ~~**An install over a running Agent half-replaces the folder** (three failed starts, `Microsoft.Data.Sqlite` not
found).~~ **A GUESS, WITHDRAWN the same evening, and it had already been published in beta.4's notes.** Nothing was
installed between the failures (20:59–21:01) and the start that worked (21:02); the PC had been restarted at 20:57
@@ -2039,7 +2041,7 @@ added.*
- **The desktop app forbids polling CI** (no `gh` loop, Monitor or ScheduleWakeup): it wakes the session on a failure, a
conflict or a review comment, never on green, so a green PR waits for the owner's word or the owner's auto-merge.
## **START HERE** — 2026-10-07, 0.1.0-beta.17: text that follows a theme switch, scroll bars hidden everywhere, a wiki
## ~~**START HERE**~~ — 2026-10-07, 0.1.0-beta.17: text that follows a theme switch, scroll bars hidden everywhere, a wiki *(struck 2026-10-07: the head is §beta.18 below)*
*Status is `CHANGELOG.md` `[0.1.0-beta.17]`; this is what no other file carries. The owner's report (2026-10-07, three
screenshots of beta.16): Settings' descriptions kept the old theme's colour after a switch; the Dashboard kept its
@@ -2082,6 +2084,44 @@ release.*
**Owner-only, added to the list below:** item 13.
## **START HERE** — 2026-10-07, 0.1.0-beta.18: page titles readable in the dark theme, notices in words, a clean build id
*Status is `CHANGELOG.md` `[0.1.0-beta.18]`; this is what no other file carries. The owner's report (2026-10-07, five
screenshots of beta.17 in the dark theme): every page's title and section headings near-black on the dark window — and
the two chips beta.17 left (§beta.17, "Found while writing the guide"). Three stacked PRs — #286 the page text, #287
the notices and the rest of that list, #288 the release build id — then the release.*
**No new decisions.** The train is the owner's request of 2026-10-07 ("beta18 with the two chip tasks, and the text that
blends into the background in dark mode").
**Traps met on the way:**
- **A page in the `NavigationView` inherits nothing.** `NavigationViewContentPresenter` is a `Frame`, and a Frame sets
`InheritanceBehavior.SkipToAppNow`; WPF UI 4.3.0 makes the TextBlock default black. A page on screen during a theme
switch takes the window's new colour through the switch and a page navigated to afterwards does not, so it came and
went. Every check before hosted pages straight in a window and measured a page the App never draws — host a page in the
presenter (`PagesLoadTests.Hosted`, `16_WPFUI_SYNTAX` §Gotchas).
- **The accent is not in the theme dictionary.** WPF UI writes it into `Application.Resources` at every
`ApplicationThemeManager.Apply(…, updateAccent: true)`; a test that swaps only the dictionary measures the other theme's
accent (`Visuals.UnderTheme` derives it as the App does; `systemAccentColor: true` reads the machine's own palette).
- **The gate's own refusals look like findings on the wire.** `HookedCaptureGate` puts a label ("kill switch", "previously
blocked") where a family goes; whatever reads a family as "an anti-cheat was found" asks the end reason first. A test
of what the user reads builds the event as the Agent does (`NoticesFromTheAgentTests`), never by hand — the
hand-written events were why every notice test was green.
- **`Safety_Consent_*` strings are versioned with the consent disclosure** (`SafetyDisclosure`): a message that is wrong
on one path gets a key of its own rather than a changed meaning.
- **A release's natives were `-dirty` by the release's own hand:** its date substitution edits `legal/*.md` before CMake
reads describe. The id is read first, handed in, and read back out of the binaries (`tools/buildid-check.ps1`).
- **Met again:** `CaptureHostEndToEndTests.WhenTheTargetExitsTheHostStopsAndSaysWhy` (`TargetNotRunning` on a slow
runner) failed #286's first CI run; the job was re-run. **And a new one** failed #288's:
`…LaunchModeOnAVulkanTargetAttachesToTheLayersRingAndInjectsNothing`. A hosted runner has the Vulkan loader and no
device, so the harness exits 77 with a `[SKIP]` line — but the runner is elevated, the launch goes through
`CreateProcessWithTokenW` (`UnelevatedProcess`), the child inherits no stdout, and the line never reaches the host's
output; when the guard's scan hits the exiting harness (`ModuleScanFailed`) the test fails instead of skipping. Re-run;
the deterministic fix is a task of its own.
**Owner-only, added to the list below:** item 14.
## Owner-only — no PR can close these
1. **§S23-2 — branch protection.** `Rules / validate` is not a required status check on
@@ -2177,6 +2217,12 @@ release.*
halves and a footer naming `v0.1.0-beta.17` — then tick *Restrict editing to collaborators only* (Settings ▸ General
▸ Features ▸ Wikis), or anyone can change a page until the next release overwrites it. (e) The `ja` strings marked
`review`.
14. **beta.18 on real hardware.** (a) **The dark theme:** every page's title and section headings — Dashboard, Games, a
game's page, Compare, Settings, Logs — read light on the dark window, after starting in Dark, after Light → Dark, and
after moving between pages. (b) **The overlay build** on the Dashboard's *Capture agent* card reads `v0.1.0-beta.18`,
without `-dirty`. (c) **Disable all hooking:** turning it on says a game being measured stops within 30 seconds; with
it on, a hooked game's notice reads "All hooking is switched off in Settings, so nothing was measured." (d) The `ja`
strings marked `review`, `Safety_Unhooked_Unnamed_Format` among them.
**Answered 2026-08-05, do not re-ask:** remove `gameguard` and keep `guard` (approved
over a red `Rules` gate, with the reasoning recorded in the merge commit); vendor NVAPI
+1 -1
View File
@@ -12,7 +12,7 @@
HTML comments are not part of the block.
-->
> ⚠ **What FrameLedger actually measures today — 2026-10-06.** The software is a beta; its latest
> **pre-release is `0.1.0-beta.17`** (2026-10-07), an unsigned installer built from that tag with its
> **pre-release is `0.1.0-beta.18`** (2026-10-07), an unsigned installer built from that tag with its
> checksums published beside it. The source holds the desktop app
> (library, store import, charts, settings) and the background Agent, which records a session when a
> game in the library runs (unless you switched its recording off), injects only into games you enabled
+1 -1
View File
@@ -114,7 +114,7 @@ Frame timing is derived from high-resolution timestamps taken at the moment the
<!-- accuracy-block:begin -->
> ⚠ **What FrameLedger actually measures today — 2026-10-06.** The software is a beta; its latest
> **pre-release is `0.1.0-beta.17`** (2026-10-07), an unsigned installer built from that tag with its
> **pre-release is `0.1.0-beta.18`** (2026-10-07), an unsigned installer built from that tag with its
> checksums published beside it. The source holds the desktop app
> (library, store import, charts, settings) and the background Agent, which records a session when a
> game in the library runs (unless you switched its recording off), injects only into games you enabled
+47 -27
View File
@@ -22,40 +22,60 @@ target_compile_features(FrameLedger.Shm INTERFACE cxx_std_20)
#
# KNOWN LIMIT, stated rather than discovered: this runs at CONFIGURE time, so the
# `-dirty` suffix reflects the tree when CMake last ran, not when the DLL was
# last compiled. For a released build that is exact — CI configures and builds
# once from a clean checkout. For a developer editing between configures it is
# approximate, and the field is a version-skew signal rather than a provenance
# guarantee. Making it exact means re-running git on every build, which is a cost
# on every incremental compile for a value only the attach handshake reads.
execute_process(
COMMAND git describe --always --dirty --abbrev=12
WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}
OUTPUT_VARIABLE FL_BUILD_ID_RAW
OUTPUT_STRIP_TRAILING_WHITESPACE
ERROR_QUIET
)
if(NOT FL_BUILD_ID_RAW)
set(FL_BUILD_ID_RAW "unknown")
endif()
# buildId is char[32] and the guard side compares the whole array, so a value
# that would be truncated must be caught here rather than becoming a permanent
# mismatch nobody can explain.
string(LENGTH "${FL_BUILD_ID_RAW}" FL_BUILD_ID_LEN)
# AFTER THE FIRST TAG the long form stopped fitting on a developer's machine: `v0.1.0-beta.2-2-gd3e74988a7bc-dirty`
# is 35 characters, so every local configure with an uncommitted change failed here and the documented workaround
# was "commit before you build" (measured 2026-09-17, again 2026-09-21). The id is a version-skew signal, not a
# provenance guarantee (above), so the fallback spends fewer characters on the same facts: an 8-digit abbreviation
# and `+` for dirty. A released build is a clean checkout AT a tag and never reaches this branch. Still too long
# after that is still fatal, for the reason the check exists.
if(FL_BUILD_ID_LEN GREATER 31)
# last compiled. This comment said that for a released build it was exact, "CI
# configures and builds once from a clean checkout", and it was not (corrected
# 2026-10-07, beta.18): release.yml writes the release date into the tracked
# legal/*.md before the gate configures, and every release until beta.18 shipped
# "<tag>-dirty" — the owner's Dashboard read "Overlay build v0.1.0-beta.17-dirty",
# and the run's log "FL_BUILD_ID = v0.1.0-beta.17-dirty" two lines after
# "RELEASE_DATE = 2026-10-07 in 3 file(s)". A release now hands the id in (below). For a
# developer editing between configures it is approximate, and the field is a
# version-skew signal rather than a provenance guarantee. Making it exact means
# re-running git on every build, which is a cost on every incremental compile for a
# value only the attach handshake reads.
#
# A RELEASE HANDS THE ID IN: release.yml reads describe from the clean checkout, before
# anything edits a tracked file, and passes it as the environment's FL_BUILD_ID; it is
# taken as given here, and tools/buildid-check.ps1 then reads it back out of the
# published binaries. It goes into a C string literal, so it must be what describe can
# print and nothing a literal would have to escape.
if(DEFINED ENV{FL_BUILD_ID} AND NOT "$ENV{FL_BUILD_ID}" STREQUAL "")
set(FL_BUILD_ID_RAW "$ENV{FL_BUILD_ID}")
if(NOT FL_BUILD_ID_RAW MATCHES "^[0-9A-Za-z._+-]+$")
message(FATAL_ERROR "FL_BUILD_ID '${FL_BUILD_ID_RAW}' from the environment holds a character no build id has (0-9 A-Z a-z . _ + -).")
endif()
string(LENGTH "${FL_BUILD_ID_RAW}" FL_BUILD_ID_LEN)
else()
execute_process(
COMMAND git describe --always --dirty=+ --abbrev=8
COMMAND git describe --always --dirty --abbrev=12
WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}
OUTPUT_VARIABLE FL_BUILD_ID_RAW
OUTPUT_STRIP_TRAILING_WHITESPACE
ERROR_QUIET
)
if(NOT FL_BUILD_ID_RAW)
set(FL_BUILD_ID_RAW "unknown")
endif()
# buildId is char[32] and the guard side compares the whole array, so a value
# that would be truncated must be caught here rather than becoming a permanent
# mismatch nobody can explain.
string(LENGTH "${FL_BUILD_ID_RAW}" FL_BUILD_ID_LEN)
# AFTER THE FIRST TAG the long form stopped fitting on a developer's machine: `v0.1.0-beta.2-2-gd3e74988a7bc-dirty`
# is 35 characters, so every local configure with an uncommitted change failed here and the documented workaround
# was "commit before you build" (measured 2026-09-17, again 2026-09-21). The id is a version-skew signal, not a
# provenance guarantee (above), so the fallback spends fewer characters on the same facts: an 8-digit abbreviation
# and `+` for dirty. A released build hands its id in (above) and never reaches this branch. Still too long
# after that is still fatal, for the reason the check exists.
if(FL_BUILD_ID_LEN GREATER 31)
execute_process(
COMMAND git describe --always --dirty=+ --abbrev=8
WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}
OUTPUT_VARIABLE FL_BUILD_ID_RAW
OUTPUT_STRIP_TRAILING_WHITESPACE
ERROR_QUIET
)
string(LENGTH "${FL_BUILD_ID_RAW}" FL_BUILD_ID_LEN)
endif()
endif()
if(FL_BUILD_ID_LEN GREATER 31)
message(FATAL_ERROR
+118
View File
@@ -0,0 +1,118 @@
# SPDX-License-Identifier: GPL-3.0-only
# Copyright (C) 2026 poli0981 - additional terms under GPLv3 section 7: see NOTICE
#Requires -Version 7.0
<#
.SYNOPSIS
Fails when a shipped native binary does not carry the release's build id, exactly.
.DESCRIPTION
FL_BUILD_ID is `git describe` at CMake's configure (src/native/FrameLedger.Shm/CMakeLists.txt), compiled into the
Overlay, the guard, the NVAPI bridge and the process-stats reader; the ring's handshake compares the Overlay's with the
guard's, and the App shows the Overlay's as "Overlay build". release.yml wrote the release date into the tracked
legal/*.md before the gate configured, so describe read the tree as dirty and every release until beta.18 shipped
"<tag>-dirty" (the owner's Dashboard, 0.1.0-beta.16 and beta.17). The workflow now reads the id from the clean
checkout, hands it to CMake as the environment's FL_BUILD_ID, and runs this over the published tree.
The id is looked for as the C string it is compiled as — its characters and the NUL after them — in each binary's
bytes. Nothing is loaded: loading the Overlay would run its DllMain in this process.
.PARAMETER BuildDir
The directory holding the native binaries (release.yml: out/app).
.PARAMETER Expected
The id they must carry: describe of the clean checkout. One that marks a dirty tree is refused too.
.PARAMETER SelfTest
Runs the fixture cases below, in a scratch directory, and exits 0 only if every one behaves.
#>
[CmdletBinding()]
param(
[string]$BuildDir,
[string]$Expected,
[switch]$SelfTest
)
Set-StrictMode -Version Latest
$ErrorActionPreference = 'Stop'
# The four that compile FL_BUILD_ID in (FrameLedger.Shm's INTERFACE define): FlGetBuildId, FlGuardBuildId, FlNvBuildId, FlPsBuildId.
$binaries = 'FrameLedger.Overlay.dll', 'FrameLedger.Guard.dll', 'FrameLedger.NvapiBridge.dll', 'FrameLedger.ProcessStats.dll'
function Get-BuildIdProblem {
param([string]$Directory, [string]$Id)
$problems = [System.Collections.Generic.List[string]]::new()
# What describe can print, and what a C string literal carries without escaping (CMakeLists.txt checks the same).
if ([string]::IsNullOrEmpty($Id) -or $Id -notmatch '^[0-9A-Za-z._+-]{1,31}$') {
$problems.Add("the expected id '$Id' is not a build id (1 to 31 of [0-9A-Za-z._+-])")
return , $problems.ToArray()
}
if ($Id.EndsWith('-dirty', [StringComparison]::Ordinal) -or $Id.EndsWith('+', [StringComparison]::Ordinal)) {
$problems.Add("the expected id '$Id' marks a dirty tree; a release is built from a clean checkout")
}
# Latin-1 maps every byte to one char, so a byte search is an ordinal string search.
$needle = $Id + [char]0
foreach ($name in $binaries) {
$path = Join-Path $Directory $name
if (-not (Test-Path -LiteralPath $path -PathType Leaf)) {
$problems.Add("$name is missing")
continue
}
$text = [Text.Encoding]::Latin1.GetString([IO.File]::ReadAllBytes($path))
if ($text.IndexOf($needle, [StringComparison]::Ordinal) -lt 0) {
$problems.Add("$name does not carry '$Id'")
}
}
return , $problems.ToArray()
}
if ($SelfTest) {
$root = Join-Path ([IO.Path]::GetTempPath()) ('fl-buildid-selftest-' + [guid]::NewGuid().ToString('N'))
$id = 'v0.1.0-beta.18'
$cases = @(
@{ Name = 'every binary carries the id passes'; Expected = $id; Ids = @($id, $id, $id, $id); Bad = 0 },
@{ Name = 'a commit id with no tag passes'; Expected = '15b4c6b1a2c3'; Ids = @('15b4c6b1a2c3', '15b4c6b1a2c3', '15b4c6b1a2c3', '15b4c6b1a2c3'); Bad = 0 },
@{ Name = "the Overlay built from a dirty tree FAILS"; Expected = $id; Ids = @("$id-dirty", $id, $id, $id); Bad = 1 },
@{ Name = "the short fallback's dirty mark FAILS"; Expected = $id; Ids = @($id, "$id+", $id, $id); Bad = 1 },
@{ Name = 'a binary with no id at all FAILS'; Expected = $id; Ids = @($id, $id, 'unknown', $id); Bad = 1 },
@{ Name = 'a missing binary FAILS'; Expected = $id; Ids = @($id, $id, $id, $null); Bad = 1 },
@{ Name = 'an expected id that marks a dirty tree FAILS'; Expected = "$id-dirty"; Ids = @("$id-dirty", "$id-dirty", "$id-dirty", "$id-dirty"); Bad = 1 },
@{ Name = 'an expected id no build has FAILS'; Expected = 'v0.1.0 "beta"'; Ids = @($id, $id, $id, $id); Bad = 1 }
)
$failed = 0
try {
$i = 0
foreach ($c in $cases) {
$i++
$dir = Join-Path $root "case$i"
New-Item -ItemType Directory -Path $dir -Force | Out-Null
for ($b = 0; $b -lt $binaries.Count; $b++) {
if ($null -eq $c.Ids[$b]) { continue }
# A string among other bytes, as the linker leaves it in .rdata.
$body = [Text.Encoding]::Latin1.GetBytes("MZ`0`0padding`0" + $c.Ids[$b] + "`0more`0")
[IO.File]::WriteAllBytes((Join-Path $dir $binaries[$b]), $body)
}
$got = Get-BuildIdProblem -Directory $dir -Id $c.Expected
$ok = ($got.Count -gt 0) -eq ($c.Bad -gt 0)
Write-Host (" {0,-4} {1}" -f ($(if ($ok) { 'ok' } else { 'FAIL' })), $c.Name) -ForegroundColor ($(if ($ok) { 'DarkGray' } else { 'Red' }))
if (-not $ok) { $failed++ }
}
}
finally {
if (Test-Path -LiteralPath $root) { Remove-Item -LiteralPath $root -Recurse -Force }
}
if ($failed -gt 0) { Write-Host "buildid-check self-test FAILED ($failed)" -ForegroundColor Red; exit 1 }
Write-Host "buildid-check self-test: $($cases.Count) cases, both directions" -ForegroundColor Green
exit 0
}
if (-not $BuildDir) { throw 'buildid-check: -BuildDir is required (or -SelfTest)' }
$problems = Get-BuildIdProblem -Directory $BuildDir -Id $Expected
if ($problems.Count -gt 0) {
$problems | ForEach-Object { Write-Host " $_" -ForegroundColor Red }
Write-Host "buildid-check: the native binaries in $BuildDir do not all carry '$Expected' (src/native/FrameLedger.Shm/CMakeLists.txt, 12_BUILD)." -ForegroundColor Red
exit 1
}
Write-Host "buildid-check: ok - $($binaries.Count) native binaries carry '$Expected'" -ForegroundColor Green
exit 0
+3 -1
View File
@@ -19,9 +19,11 @@ The full documents are [`docs/13_CI_CD.md`](../docs/13_CI_CD.md) and [`docs/11_U
statement (`legal/ACCURACY.md` and its copies) and `LIMITATIONS.md`. `VERSION` changes only with the numeric core.
2. After it merges, a signed tag `vx.y.z` (or `vx.y.z-beta.N`) on the merge commit starts `release.yml`:
- it checks the tag against `VERSION` and that the changelog has the section;
- reads the build id from the clean checkout, before anything is edited;
- writes the release date into the legal documents;
- runs the whole gate;
- publishes the App and the agent (self-contained, ReadyToRun) and checks the published tree;
- publishes the App and the agent (self-contained, ReadyToRun) and checks the published tree, the build id in the
native binaries included;
- packs them with Velopack, with a delta from the previous release;
- writes `SHA256SUMS.txt`;
- creates the GitHub release from the changelog section, marked pre-release for a beta.