| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
A suite run let ~90 throwaway test wallets subscribe and publish on public
relays from one home address, and nos.lol and relay.damus.io banned it.
tools/puppeteer.js wraps puppeteer-core: every socket to a non-local host
gets an in-page stand-in relay (EVENT -> OK, REQ -> EOSE). Tests that read
real content launch with { relays: 'read' } (real reads, publishes answered
in place); feed-live, chat-layout and cordn-ui launch with { relays: 'live' }.
LIVE_RELAYS=read|1|host,... overrides per run.
profile-key-qr: measure the QR again if a repaint swapped it out.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
|
| |
|
|
|
|
|
| |
wizard's import line.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Tmfg8k5f3TbNJS34qiqHAp
|
|
|
New messages rang in Vector but sat unseen in v3 until a reload. The
resume hook only rebuilds subscriptions after an absence it was told
about; a socket that errors while the page stays visible (a wifi hop, a
laptop lid, a phone freezing the tab without the away timer noticing,
another tab clearing that timer first) is dropped from the nostr-tools
pool with every subscription it carried, and nothing dials it again.
A watchdog now runs whenever the app is on screen: on focus, on becoming
visible, and every ten seconds between. It asks the pool whether each
relay the inbox and the rooms listen on still holds a live socket (a dial
hung past 20s counts as dead, and a relay that closed a subscription on us
counts too) and rebuilds the lot when one is missing, dropping a stuck
relay first so the rebuild dials afresh. Rebuilds retry on a growing
backoff while a relay keeps refusing. tools/chat-watchdog-test.js errors
every socket under an open room and checks the recovery.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JuwQqNWtDJcxPWtEm9134z
|
