- `mirror:` was wrong; correct displayplacer param is `mirror_of_display:<uuid>` - Same fix applied to the strip regex (extend path) and mirror detection check - Binary lookup now tries resources/bin → /opt/homebrew/bin → /usr/local/bin so dev/venue Macs with `brew install displayplacer` work without bundling - Updated TODO_AGENTS.md: marks auto-detection complete, documents one-time brew install step, bundling path, and optional per-device configStr override Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
8.0 KiB
Native App Agent Task List
Use this file to track steps for complex features or bug fixes. Status: Stable - ongoing development.
Current Investigation
- This started as an API contract review for the native Electron bootstrap path and expanded into a packaging/runtime issue after the deploy step stopped producing bundles.
- We now know the API side was not the root cause. The bootstrap request shape in
src/main/api_client.tswas wrong and has been corrected. - The packaging blocker has been diagnosed and fixed (see below).
Launcher Terminology Cleanup
- Align the native docs/comments with the Svelte-side terminology: Launch Profile = the Svelte config object keyed by extension; Native Template = the AppleScript or shell string Electron actually executes after the file lands in temp.
- Update
src/main/file_handlers.tscomments and any bridge-facing wording so they describe the resolvednative_templatestring accurately.launch_profilesis only the Svelte-side map; the IPC payload is the single executable string. Do not reintroducelaunch_scriptsas the public term. - Keep the source of truth in Svelte. Electron should remain a thin executor/copy layer.
- After source/docs updates, rebuild/regenerate
dist/main/file_handlers.jsfrom source; do not hand-edit the generated file. - If any parameter names remain awkward in source, preserve runtime behavior first and rename only when the signature ripple is understood.
What Was Fixed
- Updated the native bootstrap flow to use the direct
site_domain/searchrequest body expected by API V3. - Standardized the account-bypass header to
x-no-account-id: bypasswhere that narrow bypass is intended. - Removed a redundant
x-no-account-idheader from file download calls. - Rewrote the device lookup smoke test so it validates the real two-step bootstrap path end to end.
- Upgraded Electron from 34.x to 42.0.1.
- Replaced deprecated
electron-packagerwith@electron/packager20.0.0. - Added a
package:linuxsmoke test path so packaging failures can be isolated from macOS-specific behavior. - Fixed packaging hang on Node 26:
yauzl2.10.0 (used byextract-zipin@electron/packager) emits nodataevents on Node 26 streams, causing zip extraction to hang indefinitely. Fix: patched@electron/packager/dist/unzip.jsto usebsdtar(libarchive) instead ofextract-zip.bsdtarwas chosen over7zbecause7zrefuses macOS.appbundles with chained symlinks inside framework bundles. Patch is re-applied on everynpm installvia thepostinstallscript atscripts/patch-packager-unzip.js.
Verified So Far
npm run devworks once the Electron binary is present locally.- Manual Electron cache extraction restored a runnable checkout on this machine.
- API validation confirmed the backend responds correctly for:
event_device/{id}lookupsite_domain/search?limit=1with the directSearchQuerybody
- The returned
site_domain.account_idmatches the device account context in the verified bootstrap flow. - The SvelteKit frontend bootstrap path already follows the correct API contract and does not need the same fix.
npm run package:linuxnow producesbuilds/aether_launcher-linux-x64/with a complete bundle (confirmed 2026-05-11).npm run package:macnow producesbuilds/aether_launcher-darwin-x64/andbuilds/aether_launcher-darwin-arm64/withaether_launcher.appinside each (confirmed 2026-05-11). Initial fix used7zbut it refused chained symlinks inside macOS framework bundles; switched tobsdtar(libarchive) which handles both Linux and macOS zips correctly.deploy/deploy.shoutput directory names (aether_launcher-darwin-x64,aether_launcher-darwin-arm64) match packager output — no script changes needed.
Remaining Items
- Test that the packaged Linux binary runs end-to-end against the dev API.
Root Cause Summary (Packaging Hang)
- Tool chain: Node 26.1.0 +
@electron/packager20.0.0 +extract-zip2.0.1 +yauzl2.10.0 - Symptom:
npm run package:linuxexits 0 but produces no output. Debug log shows it starts extraction but never finishes. - Root cause:
yauzlopens a read stream for the first zip entry, but on Node 26, nodataevents are ever emitted on that stream. Thepipeline(readStream, writeStream)call inextract-zipblocks forever. - Fix: Replace the one-liner
extractElectronZipfunction innode_modules/@electron/packager/dist/unzip.jswith achild_process.execSynccall tobsdtar -xf.bsdtarwas chosen over7zbecause7zrefuses macOS.appbundles with chained symlinks (e.g.Electron Framework.framework/Libraries → Versions/Current/Libraries). Apostinstallnpm script re-applies this patch after eachnpm install. - Build-time dependency:
libarchive(providesbsdtar) must be installed on the build host. On Arch:pacman -S libarchive; macOS: included in Xcode CLT orbrew install libarchive; Ubuntu/Debian:apt install libarchive-tools.
References
- Electron 42 release notes: https://www.electronjs.org/blog/electron-42-0
- Related Electron packaging discussion: https://github.com/aaddrick/claude-desktop-debian/pull/587
- Electron packaging/runtime change reference: https://github.com/electron/electron/pull/49328
- yauzl Node 26 stream issue:
yauzl2.10.0 uses legacy Node streams (streams1 style); Node 26 changed stream internal behavior soopenReadStreamreturns a stream that never emitsdatawithout a proper pipeline consumer.
Notes
- Was on Electron 34.
- The problem is not the backend API keys or the frontend site bootstrap flow.
- The packaging fix is a node_modules patch, not upstream. If
@electron/packagerorextract-zipreleases a Node 26-compatible version, thepostinstallscript should be removed.
set_display_layout — displayplacer Setup & Status (updated 2026-05-20)
Reference: jakehilborn/displayplacer
Current state (2026-05-20):
- ✅ Handler auto-detects displays via
displayplacer list— no per-deviceconfigStrrequired for normal use - ✅ Binary lookup order:
resources/bin/displayplacer(bundled) →/opt/homebrew/bin/displayplacer(Apple Silicon Homebrew) →/usr/local/bin/displayplacer(Intel Homebrew) - ✅ Correct
mirror_of_display:<uuid>syntax used (wasmirror:— wrong, now fixed) - ✅ Failures logged to Electron console (
[Launcher] set_display_layout:) instead of silently swallowed - ✅ Display Mode toggle added to Launcher config (Native OS section) — Extend/Mirror buttons always visible, no Technical Mode required
One-time setup on each venue Mac:
brew install displayplacer
The binary is not bundled in the Electron build yet. Homebrew installs it to /opt/homebrew/bin/ (Apple Silicon) or /usr/local/bin/ (Intel) — both are in the fallback lookup chain.
Bundling the binary (future):
To ship displayplacer inside the .app bundle without requiring Homebrew on venue Macs:
- Copy the
displayplacerbinary toresources/bin/displayplacer - Mark it executable:
chmod +x resources/bin/displayplacer - Add to
package.jsonextraResourcesso@electron/packagerincludes it
Optional per-device override (manual tuning):
For rooms where auto-detection produces the wrong result, store the raw configStr in event_device.data_json:
{
"displayplacer_config_extend": "<output of displayplacer list in extended layout>",
"displayplacer_config_mirror": "<output of displayplacer list in mirrored layout>"
}
The handler accepts configStr and uses it directly, bypassing auto-detection. Pass it from the Svelte call site if needed.
displayplacer quick reference:
displayplacer list— prints current display info and a ready-to-run config string at the bottomdisplayplacer "<display_string>" "<display_string>"— applies layout- Mirror syntax: add
mirror_of_display:<primary_uuid>to the secondary display string - Extend syntax: set
origin:(<x>,0)with non-overlapping x offsets per display