The web/PWA Amber path set compressionType=none; the native intents set neither
compressionType nor returnType, and the JS gzip decoder was never called. NIP-55
lets Amber gzip a large result (Signer1... base64), so a long nip44_decrypt (a
long DM) could arrive compressed and be parsed as garbage. Native intents now
send compressionType=none; the JS signer also gunzips any Signer1 result. Timeout
message no longer wrongly says Amber may not be installed.
Root cause: the stock nostr-signer-capacitor-plugin (0.0.5) tries Amber's
background ContentProvider first, which GrapheneOS blocks and never returns,
with no clean fallback — so signing hung 90s and Amber never appeared.
Fix (matches Amethyst, the reference NIP-55 client):
- New native Capacitor plugin AmberIntent (android-templates/AmberIntentPlugin.java)
that goes STRAIGHT to a foreground nostrsigner: intent via a properly-registered
ActivityResultLauncher, tracks each request by id, and times out gracefully.
- MainActivity now registers the launcher in onCreate (AndroidX contract) and
hands it to the plugin, while staying a safe stock BridgeActivity (no
WebViewClient override → no black screen).
- inject-main-activity.mjs now installs both MainActivity.java and AmberIntentPlugin.java.
- capacitorAmberSigner.ts rewired to registerPlugin('AmberIntent'); every op is a
user-visible Amber Approve/Reject popup; rejections surface as a friendly message.
- Removed the dead npm nostr-signer-capacitor-plugin dependency (and its phantom
coracle fork spec) plus its vite external entry — signing no longer uses it.
- Delete flow copy updated to the new foreground UX (Approve in Amber / Cancelled
on reject) instead of the diagnostic SIGN/PUBLISH narration.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
Verified against the plugin's Java source: signEvent/nip*Encrypt/Decrypt
IMMEDIATELY reject with "Signer package name not set" when no package is
configured — which is exactly what my earlier "omit packageName" change caused,
and that rejection surfaced as the endless spinner on delete/accept/zap.
With the package set, the plugin tries the content resolver, and when that
returns null (permission not remembered) it falls back to
startActivityForResult(nostrsigner:) — the SAME native call login uses that
reliably opens Amber. Restore native plugin signEvent/nip04/nip44 with
setPackageName + AMBER_PACKAGE, reconstruct signature-only responses, and remove
the unused JS-intent bridge. Bump to 1.2.9.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
The capacitor plugin's signEvent/encrypt query Amber's ContentProvider first,
which stalls on GrapheneOS and never launches Amber's approval UI — causing
delete, accept-invite, zaps, and every signing action to hang forever.
Add amberIntentBridge.ts which drives Amber through the reliable
intent:...scheme=nostrsigner flow (the same path login uses that actually opens
Amber), returning a promise resolved in-process via the org.skatespots.app://
deep-link callback (DeepLinkHandler → resolveAmberIntent). CapacitorAmberSigner
now routes signEvent/nip04/nip44 through this bridge instead of the stalling
ContentResolver path. Fast 60s fail-safe with a clear message instead of an
indefinite spinner.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
The likely real cause of "Amber never pops up, delete hangs" on
GrapheneOS: login only requested the get_public_key permission, so later
sign_event / encrypt / decrypt calls hit a per-action permission check
that routes through Amber's ContentProvider and stalls WITHOUT showing a
prompt. Now login pre-grants get_public_key, sign_event, nip04/nip44
encrypt+decrypt up front, so subsequent signing is a fast background
round-trip with no surprise permission stall.
IMPORTANT: existing sessions only granted get_public_key — users must log
out and back in once for Amber to grant the new permission set.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
ROOT CAUSE of "stuck on Deleting…" with no Amber prompt:
signEvent (and nip04/nip44 encrypt/decrypt) passed packageName to the
native plugin. With a packageName set, the plugin first tries Amber's
ContentProvider — a background, no-UI query that stalls on GrapheneOS and
never launches Amber's approval screen. So the user saw NO prompt and the
request timed out ("takedown taking longer than expected").
Login already worked around this by omitting packageName so the plugin
goes straight to startActivityForResult(nostrsigner:) which shows Amber's
UI. This mirrors that fix for every signing/crypto op: use the raw native
signer with NO packageName. Removed the now-unused getPlugin /
ensurePackageName / setPackageName machinery.
This is why deletes (and any signed action / encryption) never surfaced
Amber and hung.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
Compute event id locally in CapacitorAmberSigner.signEvent instead of
relying on Amber to echo it back. Amber (NIP-55) commonly returns only a
bare `signature` for kind:5 deletions and kind:1985 moderation labels,
leaving id empty — which failed the malformed-event check or emitted a
bad event that stalled the relay forever (the "stuck on Deleting…"
symptom). The id is a deterministic NIP-01 hash of the signed fields, so
we derive it ourselves, making signature-only responses work for every
kind including admin takedowns.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
Root cause found in the plugin's native getPublicKey (NostrSignerPlugin.java):
when a packageName is set, it FIRST tries Amber's ContentProvider (background,
no-UI). On GrapheneOS that cross-app ContentResolver query is blocked and never
returns, so it never falls through to startActivityForResult(nostrsigner:) —
the branch that actually opens Amber's approval UI. Result: eternal
"Waiting for Amber".
Fix: getAmberPublicKeyViaPlugin no longer calls setPackageName and no longer
passes packageName; it passes only the get_public_key permission. With no
package name the plugin skips the content-resolver branch and launches the
intent directly, opening Amber's UI. Amber returns both pubkey and its package,
so later signing still works. Paired with the auto-injected <queries> manifest
block so the OS can resolve the intent in the first place.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
Two concrete root causes for the eternal "Waiting for Amber…":
1. setPackageName could hang forever on GrapheneOS. getPlugin() awaited it with
no timeout, so getPublicKey() (and its 90s timeout) was never even reached.
Now setPackageName races a 3s timeout and continues regardless — the Amber
package is passed explicitly on every call anyway, so it's non-essential.
2. AndroidManifest.xml had no <queries> block. On Android 11+/GrapheneOS the OS
silently blocks the nostrsigner intent without it, so Amber never opens. Added
scripts/inject-amber-queries.mjs (idempotent) and wired it into android:init /
android:sync / android:build so the block is guaranteed present in every APK,
surviving cap sync regeneration.
Bump version to 1.1.9.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
Root cause: LoginDialog and useCurrentUser detected the native app via
window.Capacitor, which is unreliable inside the WebView (false at mount / not
attached in some builds). So the ZapStore APK fell into the web intent:// redirect
flow and never opened Amber ("Waiting for Amber…" hang).
Fix: add isNativeAmberAvailable() in capacitorAmberSigner.ts using the imported
@capacitor/core Capacitor + native plugin presence, and use it (evaluated at
click time) in LoginDialog.handleAmberLogin and in useCurrentUser's signer
selection. The APK now calls getAmberPublicKeyViaPlugin() → native intent → Amber
opens directly, no redirects.
Bump version to 1.1.7.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
Symptom: crew logo upload spins forever, and crew creation appears to do
nothing. Both go through Amber signEvent. Login (getPublicKey) works because
it uses Amber's silent content-resolver path, but signEvent must launch
Amber's approval Activity and wait for the result — and if that result never
comes back (buggy plugin build, user backs out, missing permission) the
bridge promise never settles, so the UI hangs with no error.
Changes:
- withTimeout() wraps every Amber bridge call (sign/getPublicKey/nip04/nip44)
with a 90s cap. A hung round-trip now rejects with a clear, actionable
message instead of spinning forever, which also unblocks the CrewForm
(isUploading/isPublishing finally resolve).
- setPackageName is now called ONCE per session (cached promise) instead of
before every op — repeated calls are needless bridge chatter and can wedge
on some builds.
NOTE (root cause, fixed on the build machine): package.json requests the
GitHub `main` build of nostr-signer-capacitor-plugin, but package-lock.json
still pins the old npm 0.0.5 (which the APK bundles). 0.0.5 is the version
prone to the signEvent activity-result hang. Regenerate the lockfile with
`npm install` so the `main` build is used — see chat for steps.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
Root cause (same class as the camera bug): capacitorAmberSigner.ts used
`await import('nostr-signer-capacitor-plugin')`, but that package is marked
external in vite.config.ts and the native build uses inlineDynamicImports, so
the bare specifier survives in the APK bundle and the Android WebView can't
resolve it at runtime. The import threw, getPlugin() returned null, and every
Amber signing op — including the Blossom upload auth event that a Crew logo
upload needs — failed with "Capacitor Amber plugin not available".
Fix: access the native signer via Capacitor's registerPlugin('NostrSignerPlugin')
from the bundled @capacitor/core, and reimplement the plugin's thin JS wrapper
logic directly against the native bridge (options-object API per the plugin's
main source). This removes the broken dynamic import entirely and is version-
independent as long as the native plugin is registered. Covers both native
Amber login (getAmberPublicKeyViaPlugin) and all signing/encryption paths.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
- JSObject from Capacitor needs JSON.parse(JSON.stringify()) to be
accessible as plain JS object with property access
- Restore AMBER_PACKAGE as argument (null was causing broken URI)
- Remove debug toast from LoginDialog
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
The Content Resolver silent path returns a cached/wrong key.
Passing null skips it and always opens Amber UI so the user
explicitly selects their key and the correct pubkey is returned.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
NostrSigner.npubToHex() is called before putting result in "npub" field,
so the value is already a 64-char hex pubkey. Skip nip19.decode() and
use it directly. This was causing the "data must be at least 8 chars" error.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
The actual plugin source uses positional args, not object params:
- getPublicKey(packageName)
- signEvent(packageName, eventJson, id, npub)
- nip04Encrypt/Decrypt(packageName, text, id, pubKey, npub)
- nip44Encrypt/Decrypt(packageName, text, id, pubKey, npub)
- isExternalSignerInstalled(packageName)
Previous implementation was based on wrong README docs.
Now based on actual dist/esm/index.js source.
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
- Use named export { NostrSignerPlugin } not mod.default
- setPackageName takes a plain string not { packageName: string }
- isExternalSignerInstalled takes packageName as argument
- These were the exact bugs causing "plugin not available" error
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
- Add CapacitorAmberSigner: a full NostrSigner implementation backed by
the native nostr-signer-capacitor-plugin (login, signEvent, nip04,
nip44 all go through Amber via native Android intents)
- LoginDialog: on Capacitor, call getAmberPublicKeyViaPlugin() directly
instead of the intent:// URL redirect — returns a Promise, no
callback URL or polling needed. Shows clear error if Amber not installed.
- useCurrentUser: on Capacitor amber logins use CapacitorAmberSigner
so every publish/sign goes through the plugin, not window.location.href
- Web/PWA path unchanged — still uses URL redirect + AmberSigner
Co-authored-by: shakespeare.diy <assistant@shakespeare.diy>
