lfg2025/archy
1
0
Fork 0
Code Issues 25 Pull Requests 4 Actions Packages Projects Releases 74 Wiki Activity

feat(content): owned-content persistence + Fedimint paid downloads, fmcd caps fix, FIPS warm-path perf

Browse Source 
Buyer-side paid downloads now persist: purchases are cached on disk
(content_owned.rs) keyed by (seller onion, content_id), the gallery shows
an "Owned" badge unblurred, and items view/play in-app from the local
cache with no re-payment or reliance on a browser download (which
silently failed on the mobile companion). New RPCs content.owned-list /
content.owned-get. Validated e2e .116<-.198 (paid 100 sats via Fedimint,
166KB jpeg returns, survives restart).

fedimint-clientd manifest: restore the standard container capability set
(CHOWN/DAC_OVERRIDE/FOWNER/SETUID/SETGID) so fmcd's startup chown of an
existing-federation /data succeeds instead of dying EPERM (#7). Confirmed
the orchestrator applies these to the running container.

FIPS perf: tighten the supervisor warm-path keepalive 45s -> 25s so peer
paths stay inside the ~30-60s NAT cold window. Dials now reliably land on
FIPS instead of re-punching and falling back to Tor. Measured to the same
peer: cloud browse 18-22s -> 0.4s; full Fedimint paid download 29s -> 11s
(residual is the seller-side guardian reissue round-trip).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
archipelago
2026-06-20 18:58:52 -04:00
co-authored by Claude Opus 4.8
parent b0c9bd2a0c
commit db7d424bff
12 changed files with 544 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files
neode-ui/src/main.ts
+31 -2
View File
@@ -110,10 +110,39 @@ function recordError(source: string, err: unknown, info?: string) {
app.config.errorHandler = (err, _instance, info) => recordError('Vue Error', err, info)
// After a frontend deploy the browser's cached index.html still points at the
// OLD hashed chunks (e.g. AppSession-Cq93o4ao.js), which 404 — Vite then throws
// "Failed to fetch dynamically imported module". The fix is to reload once so
// the browser pulls the fresh index + chunk map. Guard with sessionStorage so a
// genuinely-broken chunk can't trap us in a reload loop.
function isStaleChunkError(err: unknown): boolean {
const msg = (err as { message?: string })?.message ?? String(err ?? '')
return /Failed to fetch dynamically imported module|error loading dynamically imported module|Importing a module script failed|ChunkLoadError|dynamically imported module/i.test(msg)
}
function reloadOnceForStaleChunk(err: unknown): boolean {
if (!isStaleChunkError(err)) return false
try {
const KEY = 'archy-chunk-reload-at'
const last = Number(sessionStorage.getItem(KEY) || '0')
// Only reload if we haven't already tried in the last 10s (loop guard).
if (Date.now() - last < 10_000) return false
sessionStorage.setItem(KEY, String(Date.now()))
} catch { /* sessionStorage unavailable — reload anyway, once is better than stuck */ }
// Cache-bust the navigation so the stale index isn't served again.
location.reload()
return true
}
// Vue's errorHandler only catches errors raised synchronously inside Vue's
// lifecycle/reactivity. Async rejections and plain runtime errors (e.g. a JS
// API missing in an older WebView) slip past it, so catch those too.
window.addEventListener('error', (ev) => recordError('window.error', ev.error ?? ev.message))
window.addEventListener('unhandledrejection', (ev) => recordError('unhandledrejection', ev.reason))
window.addEventListener('error', (ev) => {
if (reloadOnceForStaleChunk(ev.error ?? ev.message)) return
recordError('window.error', ev.error ?? ev.message)
})
window.addEventListener('unhandledrejection', (ev) => {
if (reloadOnceForStaleChunk(ev.reason)) return
recordError('unhandledrejection', ev.reason)
})
app.mount('#app')