Replace manual tag assembly in test helpers (makeOffer, makeAnswer, etc.)
with the existing build() methods from quartz event classes. This keeps
test tag structures in sync with production code automatically.
https://claude.ai/code/session_01AfRYTRCvtKqqDxeKQujUrx
Profile pictures are displayed at small fixed sizes (18-100dp) but Coil's
disk cache stores full-size originals, causing expensive re-decode on every
memory cache miss. With 60 avatars on screen, this creates significant I/O.
Adds a ProfilePictureFetcher with a dedicated ThumbnailDiskCache that stores
pre-resized 256px JPEG thumbnails (~5-10KB each). On cache hit, reads a tiny
file instead of re-decoding the full original. The zoomable full-screen dialog
bypasses this cache and loads the original via the normal Coil pipeline.
https://claude.ai/code/session_01PBJS3HDrurLP3i5n4tMsCv
Add 81 tests across two test suites covering the full NIP-AC spec:
- quartz/NipACStateMachineTest (31 tests): Protocol compliance test vectors
for event structure, tags, P2P/group flows, ICE serialization, staleness,
renegotiation glare rules, and multi-device support
- commons/CallManagerTest (50 tests): State machine integration tests using
real NostrSignerInternal with actual crypto, covering:
* Full call lifecycle (Idle → Offering/IncomingCall → Connecting → Connected → Ended → Idle)
* Call rejection, busy auto-reject, hangup from any state
* Self-event filtering (ICE, hangup, answer-elsewhere)
* Mid-call renegotiation (voice ↔ video)
* Group calls (mesh discovery, partial disconnect, invite peer)
* Interface-level tests with real signing + gift wrapping pipeline
* Full end-to-end P2P flow with two CallManager instances
Also adds test vector tables to NIP-AC.md spec for other implementers.
https://claude.ai/code/session_01AfRYTRCvtKqqDxeKQujUrx
Use EphemeralGiftWrapEvent (kind 21059) instead of GiftWrapEvent (kind
1059) for WebRTC call signaling. The ephemeral kind signals to relays
that these events are transient and should not be persisted, eliminating
the need for expiration tags on both inner signaling events and outer
wraps.
Changes:
- Remove expiration tags from all 6 call event types (25050-25055)
- Switch WebRtcCallFactory to produce EphemeralGiftWrapEvent wraps
- Update CallManager and CallController publishEvent types
- Update Account.publishCallSignaling signature
- Use CallManager.MAX_EVENT_AGE_SECONDS for staleness checks
- Update NIP-AC spec to document EphemeralGiftWrap usage
- Remove expiration-related tests
https://claude.ai/code/session_014kyBgZx7cNyeUXYWV25M4j
When a peer rejects a group call, CallManager updated its state but never
notified CallController to dispose the peer's PeerSession. This caused two
issues:
1. The rejected peer's PeerConnection lingered in HAVE_LOCAL_OFFER state
until the entire call ended (resource leak visible as "disposing 2 peer
sessions" when only 1 should exist).
2. In onPeerDisconnected, the allDisconnected check required ALL sessions
to be CLOSED. The stale session blocked this check, preventing automatic
hangup when ICE failed — meaning the remote peer never received a hangup
event and kept ringing.
Fix: invoke onPeerLeft when a peer rejects in Offering/Connecting/Connected
states, and make onPeerDisconnected treat sessions without a remote
description as inactive.
https://claude.ai/code/session_01UXaFKmHVPDyTzfHL8fkSw1
Four bugs fixed in group call signaling:
1. Self-event filtering: Own answer/ICE/hangup events echoed from relays
were processed as if from another peer, causing callees to attempt
callee-to-callee connections with themselves.
2. Lost peer discovery: When a callee received another callee's answer
while still ringing (IncomingCall state), the peer was silently
discarded. Now stored in discoveredCalleePeers and used for mesh
setup after accepting.
3. Duplicate onNewPeerInGroupCall: Both CallManager and CallController
triggered mesh setup for the same peer, causing duplicate log entries
and wasted work. Removed CallManager's redundant calls since
CallController handles this internally via onCallAnswerReceived.
4. No per-peer session cleanup: When a peer hung up but the call
continued, their WebRTC PeerSession was never disposed — it lingered
until ICE timeout. Added onPeerLeft callback and disposePeerSession
to cleanly close individual sessions and update video tracks.
Also added strategic debug logging for session lifecycle, peer
discovery, and disconnect handling.
https://claude.ai/code/session_01SWDsQJibYi1MMgvu55puSY
When a group member (e.g. Bob) rejects a call, the rejection event is
sent to all group members. The caller (Alice) correctly removes Bob,
but other callees (Charlie) were not removing Bob because the
IncomingCall handler in onCallRejected only handled self-rejection
for multi-device sync. Now it also removes the rejecting peer from
groupMembers, matching the behavior already present in onPeerHangup.
https://claude.ai/code/session_01PMwKCEU1fRDGNQ8WdX8CFW
Three issues addressed:
1. Make cleanup() exception-safe: wrap each WebRTC dispose call in
its own try-catch so that a failure in one (e.g. native crash
disposing a PeerConnection) does not skip releasing the camera,
audio mode, foreground service, or EGL context.
2. Upgrade the Idle safety net to call cleanup() instead of only
stopping ringing/notifications. If the Ended state is missed due
to StateFlow conflation, the Idle handler now performs full
resource cleanup (camera, WebRTC, audio mode, foreground service,
proximity wake lock). cleanup() is idempotent since all resources
are null-checked and nulled out.
3. Call cleanup() in AccountViewModel.onCleared() so that WebRTC
resources, camera, audio mode, and foreground service are released
when the ViewModel is destroyed (e.g. logout, activity recreation).
4. Clear processedEventIds when transitioning to Idle to prevent
unbounded memory growth across calls.
https://claude.ai/code/session_01GWRdrVAa29BsDkv8R7Z3Y9
Replaces the single-PeerConnection architecture with full mesh topology
where each participant maintains one PeerConnection per peer, as
specified by NIP-AC.
WebRtcCallSession: Refactored to a pure PeerConnection wrapper that
accepts a shared PeerConnectionFactory. No longer manages media sources,
tracks, or camera — those are now shared across all peer sessions.
CallController: Manages per-peer sessions via ConcurrentHashMap. Shared
resources (PeerConnectionFactory, EglBase, audio/video sources, camera)
are initialized once and reused. Each peer gets its own WebRtcCallSession
with per-peer ICE candidate routing. Supports callee-to-callee mesh
connections with pubkey-based tie-breaking to avoid ofer glare.
CallManager: New methods for per-peer offer/answer publishing
(publishOfferToPeer, publishAnswerToPeer, beginOffering). Forwards ALL
answers to CallController (not just the first). New callbacks
onNewPeerInGroupCall and onMidCallOfferReceived for mesh setup. Handles
mid-call offers (same call-id) when already in Connecting/Connected state.
AccountViewModel: Updated callback wiring to pass peer pubkey with
answer events and register new group call callbacks.
https://claude.ai/code/session_01J5fJx9YbSBx1BsBMctiAm8
Two bugs caused WebRTC group calls to get stuck in "Connecting":
1. CallController.onLocalIceCandidate() used currentPeerPubKey() which
returns only the first peer via firstOrNull(). ICE candidates were
gift-wrapped to only one peer; the others never received them.
Fixed by iterating over all currentPeerPubKeys().
2. CallManager.acceptCall() set Connecting.peerPubKeys to groupMembers
which includes the local user's own pubkey. This caused
currentPeerPubKey() to potentially return self, sending ICE
candidates to oneself instead of the caller.
Fixed by filtering out signer.pubKey from the peer set.
https://claude.ai/code/session_01J5fJx9YbSBx1BsBMctiAm8
In group calls (e.g. Alice calls Bob and Charlie), when Bob accepts the
call, a CallAnswerEvent is gift-wrapped to all group members including
Charlie. Charlie's CallManager, still in IncomingCall state, was treating
ANY CallAnswerEvent as "answered elsewhere" (intended for multi-device
scenarios) and ending the call prematurely.
The fix checks whether the answering/rejecting peer is actually the
current user (signer.pubKey) before treating it as an "answered/rejected
elsewhere" event. If it's a different group member, we simply ignore it
and keep ringing.
https://claude.ai/code/session_01EYzWB93PZRqw15QuQadCf4
Adds debug logging across CallManager, CallController, and WebRtcCallSession
to trace the full call lifecycle: SDP offer/answer creation, remote description
set success/failure, ICE candidate exchange, ICE connection state transitions,
and signaling state changes. The noOpSdpObserver is replaced with a logging
observer so setRemoteDescription success/failure is visible in logcat.
Filter logcat with: CallManager|CallController|WebRtcCallSession
https://claude.ai/code/session_01M3yyyRyu8KWJZ5PjNt4V2H
The group factory methods work correctly with any number of members,
including single-peer calls. Remove the if/else branches that
duplicated P2P vs group logic in acceptCall, rejectCall, hangup,
sendRenegotiation, and sendRenegotiationAnswer.
https://claude.ai/code/session_013h2E7spwHDgSjunqumsYgp
Adds a security setting to hide posts with more than X hashtags (t tags),
defaulting to 5. Applied at the feed filter level via FilterByListParams.match()
and Account.isAcceptable() to cover all feeds. Configurable in Security Filters
settings, with 0 to disable.
https://claude.ai/code/session_01NMVypgGSU9EjQA7hZ9uvjD
SDP payloads (offer, answer, renegotiate) are specific to individual
PeerConnections and cannot be shared across peers. This commit:
- Adds group-context overloads to createCallOffer, createCallAnswer,
and createRenegotiate that include all member p-tags but wrap to
a single target peer
- Removes createGroupRenegotiate (renegotiation is always per-peer)
- Updates sendRenegotiation/sendRenegotiationAnswer to take explicit
peerPubKey and use per-peer wrapping with group p-tags
- Updates invitePeer to include all existing group members + invitee
in p-tags so the new peer sees the full group composition
- Documents SDP per-peer vs sign-once distinction in NIP-AC spec
https://claude.ai/code/session_013h2E7spwHDgSjunqumsYgp
All signaling event types in group calls (answer, hangup, reject,
renegotiate) now include p-tags for every group member, matching the
pattern already used by CallOfferEvent. This allows signing the inner
event once and gift-wrapping it separately for each recipient.
- Add Set<HexKey> build() overloads to CallAnswerEvent, CallHangupEvent,
CallRejectEvent, and CallRenegotiateEvent
- Add createGroupCallAnswer, createGroupRenegotiate factory methods
- Fix createGroupHangup to sign once instead of per-peer
- Update createGroupReject to accept full member set
- Update CallManager to use group methods when in group call
- Document group call p-tag convention in NIP-AC spec
- Add tests for all group build overloads
https://claude.ai/code/session_013h2E7spwHDgSjunqumsYgp
- Include pending peers in hangup notification so ringing users get
immediate feedback instead of waiting for the 60s timeout
- Sign individual hangup events per peer with correct p-tag instead
of reusing a single event with the first peer's tag
- Stop treating ICE DISCONNECTED as terminal; only hang up on FAILED
since DISCONNECTED is often transient (network switch, packet loss)
- Send "busy" reject when receiving a call while already in one
https://claude.ai/code/session_01HpowdWMK77pA35ABn9XWpD
Add ability to invite new users into an ongoing call:
- CallManager.invitePeer() sends a call offer to a new peer using
the current callId, adding them to pendingPeerPubKeys
- CallController.invitePeer() exposes this to the UI layer
- Add PersonAdd button to connected call controls
- Add AddParticipantDialog with user search (reuses UserSuggestionState)
- New strings: call_add_participant, call_search_users
Rewrite CallScreenPreviews to showcase all call states:
1. Offering P2P call (Calling...)
2. Offering group call (3 avatars)
3. Connecting
4. Incoming voice call
5. Incoming video call
6. Incoming group call (4 avatars)
7. Connected P2P voice call with controls
8. Connected muted + video + speaker
9. Connected group call with pending peers
10. Connected with Bluetooth audio
11. Call ended
12. Large group (5+ members, +N badge)
13. PiP calling
14. PiP connected
https://claude.ai/code/session_01LR8NmFGdMoDTVcfKjL7HWR
The call now starts immediately when the first peer answers instead
of waiting for all participants. Remaining peers are tracked as
pending and shown in the UI with a "Waiting for others to join..."
indicator. When a pending peer answers, they move to the connected
set. If a pending peer rejects or hangs up, they're silently removed.
Changes:
- Add pendingPeerPubKeys to Connecting and Connected states
- Add allPeerPubKeys helper on Connected for UI rendering
- Split peers into connected/pending on first answer in onCallAnswered
- Handle subsequent answers in Connecting and Connected states
- Carry pending peers through Connecting -> Connected transition
- Show all peers (connected + pending) in call UI avatars
- Display "Waiting for others to join..." when pending peers exist
https://claude.ai/code/session_01LR8NmFGdMoDTVcfKjL7HWR
When a peer hangs up or rejects in a group call, remove them from the
participant set instead of ending the entire call. The call only ends
when no peers remain (all others left). Similarly, if a peer rejects
during the Offering phase of a group call, remove them but continue
ringing remaining peers.
For incoming group calls, if the original caller hangs up the call
ends immediately. If another group member leaves, the call continues
as long as at least 2 members (including self) remain.
https://claude.ai/code/session_01LR8NmFGdMoDTVcfKjL7HWR
Extends the NIP-AC P2P call signaling to support group calls by
sending individual gift-wrapped offers to each member of the group,
all sharing the same call-id.
- CallOfferEvent: add multi-pubkey build overload, groupMembers(),
isGroupCall(), and recipientPubKeys() helpers
- WebRtcCallFactory: add GroupResult type, createGroupCallOffer(),
createGroupHangup(), and createGroupReject() methods
- CallState: change peerPubKey to peerPubKeys (Set<HexKey>) across
Offering, Connecting, Connected, and Ended states; add groupMembers
field to IncomingCall
- CallManager: add initiateGroupCall() that creates a single signed
offer with p tags for every callee and gift-wraps it individually;
hangup sends to all peers in a group call
- CallController: add initiateGroupCall() entry point
- Tests: add group call offer tests for p tags, call-id, call-type,
and expiration
https://claude.ai/code/session_01WafqMofFQfWCQA5TcLvHRb
When a user is logged in on multiple devices, all devices ring on an
incoming call. Previously, accepting or rejecting on one device left the
others still ringing.
Now, acceptCall() and rejectCall() also publish a self-addressed
gift-wrapped event (to the user's own pubkey) so other devices receive
it and transition out of the IncomingCall state. Added ANSWERED_ELSEWHERE
end reason and handling in onCallAnswered/onCallRejected for IncomingCall
state.
https://claude.ai/code/session_01X9juzyPYWqxRMCgDWEw8R1
When toggling video mid-call, addVideoTrack triggers
onRenegotiationNeeded but the callback was empty — the remote peer
never received the updated SDP, so video never appeared on their side.
This wires up the full renegotiation flow using the existing
CallRenegotiateEvent (kind 25055):
- WebRtcCallSession: forward onRenegotiationNeeded to a callback
- CallController: create and send renegotiation offers, handle
incoming renegotiation offers by creating answers
- CallManager: route CallRenegotiateEvent, allow CallAnswerEvent
during Connected state for renegotiation answers
https://claude.ai/code/session_01WdmqktFBjXKB6PYRZ5b4FD
- Add case-insensitive role matching in ParticipantTag (accepts both
"host" and "Host" from other clients per NIP-53 spec)
- Add PARTICIPANT role to ROLE enum per NIP-53 spec
- Add MeetingSpaceEvent (30312), MeetingRoomEvent (30313), and
MeetingRoomPresenceEvent (10312) consumption in LocalCache
- Create rendering composables for MeetingSpaceEvent and MeetingRoomEvent
with status flags (OPEN/PRIVATE/CLOSED for spaces, LIVE/PLANNED/ENDED
for rooms) and participant display
- Add meeting kinds to all discovery feed relay filters (global, authors,
hashtag, geohash, community)
- Add meeting kinds to ChannelCardCompose and LiveActivityCard thumb
rendering
- Add meeting kinds to KindRegistry search aliases
https://claude.ai/code/session_01CJ6xfY9gRA1c5r4T1SrUuv
1. Move UI toggle state out of CallState.Connected:
- Removed isAudioMuted/isVideoEnabled/isSpeakerOn from domain
CallState. These are UI concerns, not call state.
- Added StateFlows for toggles in CallController (isAudioMuted,
isVideoEnabled, isSpeakerOn) with proper toggle methods.
- CallScreen reads toggle state from CallController flows.
- Removed toggleAudioMute/toggleVideo/toggleSpeaker from
CallManager (no longer manages UI state).
2. Move ICE candidate parsing to quartz layer:
- Added candidateSdp(), sdpMid(), sdpMLineIndex() parsers and
serializeCandidate() to CallIceCandidateEvent in quartz.
- Removed companion object with parseIceCandidate/
serializeIceCandidate from CallController.
- CallController now uses quartz-layer parsing directly.
- Updated IceCandidateSerializationTest accordingly.
3. Add helper methods to CallManager:
- currentCallId() and currentPeerPubKey() extract from any
active state, reducing repeated when expressions.
4. Fix empty callId bug in ChatroomScreen:
- Navigation to ActiveCall now uses callManager.currentCallId()
instead of hardcoded empty string.
https://claude.ai/code/session_017hZm7yu7CzmcQgZGSaqSXS
Temporary diagnostic logging to identify why calls hang at
Connecting. Logs added to:
- CallManager.onSignalingEvent: event kind, age, state
- CallController: answer received, ICE candidates (buffered vs
direct), flush count, local ICE generation
https://claude.ai/code/session_017hZm7yu7CzmcQgZGSaqSXS
- Change EXPIRATION_SECONDS from 300 to 20 for all call signaling
events (offer, answer, ICE, hangup, reject, renegotiate)
- Change MAX_EVENT_AGE_SECONDS from 30 to 20 to match
- Update NIP-AC doc accordingly
- Fix SecurityException crash on SDK 36: phoneCall foreground service
type requires MANAGE_OWN_CALLS permission. Switch to microphone
type which only needs RECORD_AUDIO (already granted).
https://claude.ai/code/session_017hZm7yu7CzmcQgZGSaqSXS
- Deduplicate signaling events via processedEventIds set in
CallManager to prevent duplicate processing from multiple relays
- BackHandler on CallScreen calls hangup() before navigating back
- KeepScreenOn composable adds FLAG_KEEP_SCREEN_ON during calls
and clears it on dispose
https://claude.ai/code/session_017hZm7yu7CzmcQgZGSaqSXS
Two fixes for call connectivity:
1. ICE candidate buffering: Candidates arriving before the WebRTC
session exists (callee ringing) or before remote description is
set (caller waiting for answer) are now queued in
pendingIceCandidates and flushed once setRemoteDescription is
called. This was the root cause of calls getting stuck at
"Connecting" — ICE candidates were silently dropped.
2. Stale event filter: All signaling events older than 30 seconds
are discarded in CallManager.onSignalingEvent() to prevent
old cached events from triggering phantom calls.
Also: removed cleanup() from WebRTC onDisconnected callback to
avoid double-cleanup race with the CallManager state observer.
https://claude.ai/code/session_017hZm7yu7CzmcQgZGSaqSXS
CallManager now auto-resets from Ended to Idle after 2 seconds via
transitionToEnded(). Previously reset() was called from a
LaunchedEffect in CallScreen which could be cancelled when the
composable was disposed via popBack(), leaving the state stuck at
Ended and silently dropping subsequent incoming calls.
Also: CallController now observes CallManager state and auto-cleans
up the WebRTC session when a call ends (handles peer hangup case).
https://claude.ai/code/session_017hZm7yu7CzmcQgZGSaqSXS
Connects all call infrastructure so calls flow through the system:
- EventProcessor routes unwrapped call events (offer/answer/ICE/
hangup/reject) from gift wraps to CallManager
- CallController orchestrates WebRTC session lifecycle: creates
PeerConnection, generates SDP offers/answers, exchanges ICE
candidates via gift-wrapped events, and manages foreground service
- AccountViewModel initializes CallManager + CallController and
wires answer/ICE callbacks between them
- Account.publishCallSignaling() publishes gift-wrapped events
- DM chat top bar gets a call button (1-on-1 rooms only) that
initiates a voice call and navigates to ActiveCall screen
- ActiveCall route registered in AppNavigation with CallScreen
https://claude.ai/code/session_017hZm7yu7CzmcQgZGSaqSXS
Implements P2P calling over Nostr relays using WebRTC for media
transport and NIP-59 Gift Wraps for encrypted signaling. No custom
server required — only public STUN servers for NAT traversal.
Protocol layer (quartz/nip100WebRtcCalls):
- 6 new event kinds (25050-25055): offer, answer, ICE candidate,
hangup, reject, renegotiate
- WebRtcCallFactory for creating and gift-wrapping signaling events
- CallIdTag and CallTypeTag for event metadata
- Events registered in EventFactory
Call state machine (commons/call):
- CallState sealed interface with full lifecycle states
- CallManager orchestrating signaling and state transitions
- Follow-gate spam prevention: only followed users can ring,
non-follows are silently ignored
Android WebRTC integration (amethyst/service/call):
- WebRtcCallSession wrapping Google WebRTC PeerConnection
- CallForegroundService for keeping calls alive in background
- IceServerConfig with default public STUN servers
- User-configurable TURN server support
Android UI (amethyst/ui/call):
- CallScreen with offering, connecting, connected, and ended states
- IncomingCallUI with accept/reject buttons
- ConnectedCallUI with mute, video toggle, speaker, and timer
- Call button added to 1-on-1 DM chat header
- ActiveCall route added to navigation
https://claude.ai/code/session_017hZm7yu7CzmcQgZGSaqSXS
Rename the existing BookmarkListEvent (kind 30001) to OldBookmarkListEvent
and create a new BookmarkListEvent with kind 10003 as per the updated spec.
- OldBookmarkListEvent (kind 30001): Kept as-is for backward compatibility,
displayed as "Old Bookmarks" in the Bookmark Lists screen
- BookmarkListEvent (kind 10003): New replaceable event, all new bookmark
operations (save/check/remove) use this kind
- Both kinds are displayed as separate items in the Bookmark Lists screen
- Old Bookmarks screen includes a "Move All to New Bookmarks" button that
migrates public bookmarks to public and private to private
- Created PrivateReplaceableTagArrayEvent base class for replaceable events
with encrypted private tags (kind 10003 is in the 10000-19999 range)
- Updated all relay filters, search queries, and profile views to fetch
both kinds
- Updated Android and Desktop modules
https://claude.ai/code/session_01U9sjQHQMVVHxYiesoXjqop
When a user hits Quote, the new post screen opens with the nostr URI
pre-filled but the cursor was at the end (after the URI). This moves
the cursor to the beginning so the user can immediately start typing
their commentary.
https://claude.ai/code/session_01RPseKC9GJ8hs3GGBR85ezw
When events share the same createdAt timestamp, the tiebreaker sort by id
must be ascending to produce a stable, deterministic order. Fixed several
locations that either had no secondary sort, used descending id sort, or
used .reversed() which incorrectly flipped both sort directions.
https://claude.ai/code/session_01RvuyPf1x9wLf2DCgsSXLGz
`equalImmutableLists` is public and uses referential equality (`===`) per element rather than structural equality (`==`). Without documentation, callers may assume normal list equality and get false negatives for value-equal but distinct instances. This is a non-obvious contract and should be documented explicitly to prevent subtle logic bugs.
Affected files: ListUtils.kt
Signed-off-by: Nguyen Van Nam <nam.nv205106@gmail.com>
Claude