When a proxy lets neither WebSocket nor SSE through, the app opens a
poll connection: the server buffers its events (an SSEClient without a
stream), each GET waits up to 25 s, and room joins work like for SSE.
Tasks and lists now update live there too, and note sync takes its
notifications from it instead of a second long poll. Idle connections
close after two minutes. The profile shows such a connection as limited.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The old list (FLAW-01 to 27) was done and is removed; references to it
in other documents are gone or replaced by the content. New findings
from a review after v0.5, among them a high one: javascript: links in
task notes run on click (Markdown viewer), which with the session in
localStorage allows taking over accounts through a shared task list.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
WebSocket and EventSource put the 30-day session token into the URL,
where proxies log it. The app now gets a one-time ticket right before
connecting (POST /auth/stream-ticket, 60 s, bound to the token version);
?token= is no longer accepted for streams, non-browser clients may use
the Authorization header.
With BASE_URL set, CORS allows only that origin and WebSocket upgrades
only it or the requested host; credentials are off (no cookies exist).
Logout, ending all sessions, a new password and deactivation close all
live streams of the account; clients with a valid session reconnect
with a fresh ticket.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Turn the collaboration plan into module docs (Doku/modules/collaboration.md)
describing the implemented state; remove Doku/plans
- Header version comes from the build: git tag via APP_VERSION (CI passes
it for tag builds), otherwise "<package version>-dev"; package 0.2.0
- Sample data (setup option and demo mode): six inbox thoughts, two task
lists with Kanban states, subtasks and due dates, shopping and packing
lists, three notes in folders incl. a long nested outline; demo mode
shares the note and two lists with user/user
- e2e/screenshots.sh renews the README screenshots from demo data
- README: live collaboration, editor sharing, image tags, version badge
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Phase 5 of the collaboration plan:
- Remove PATCH /documents/{id}/blocks, PatchDocumentBlocks, the
DOC_BLOCKS_PATCHED event and its types, FlushDocumentState and
LoadDocumentState. PUT /documents/{id} rejects blocks (400) and only
changes title, icon, folder and pin; editors may now change title/icon.
- The to-do overview ticks via a single operation on /docs/{id}/sync
(new skipLog flag). Task links set the block tick with a targeted
operation and notify open editors; creating a linked task needs edit
rights.
- Keep the three share tables: DocRole already decides centrally, and
rewriting the visibility queries would only add risk (documented in plan).
- Tests adapted to the new paths; share e2e test now also ticks a to-do in
the overview.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Phase 4 of the collaboration plan:
- The sync engine listens for DOC_SEQ in the note's realtime room and syncs
right away. Without WebSocket/SSE it long-polls /docs/{id}/changes (25 s);
after three failures it falls back to polling every 3 s and retries long
polling when the browser comes back online.
- With push or long polling the engine only syncs every 7 s while holding
leases (renewal), otherwise every 30 s as a safety net. Focusing a block
announces its lease immediately (otherwise it could take up to 30 s).
- Realtime connection: SSE counts as connected only after its first event;
a stream that stays silent for 5 s (buffering proxy) switches SSE off for
the tab. App-level PING/PONG detects WebSockets that are accepted but
forward nothing and switches to SSE after 35 s.
- The editor shows the active channel for shared notes.
- Tests: engine channel tests (push, heartbeat, lease on focus, long
polling, interval fallback, switching), realtime fallback tests with fake
WebSocket/EventSource, Go tests for PONG and DOC_SEQ delivery, and
e2e/channels.e2e.mjs measuring latency behind six simulated proxy setups
(0.2-0.3 s on all live channels, ~3 s with plain polling).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Phase 1 of the collaboration plan (server side):
- internal/collab: block tree with parent IDs and fractional order keys,
operations with conflict rules (version check, block leases, conflict
copies instead of lost text, restore on late edits, cycle checks),
in-memory leases/sessions and long-poll notifier.
- Storage: per-note lock + transaction, append-only effect log (doc_ops),
idempotent results per opId (doc_op_results), snapshot, log paging,
central DocRole. Old flat block data is converted to the tree.
- The existing whole-document save (PUT/PATCH) runs as an adapter through
the same engine, so the current editor keeps working and writes to the log.
- API: GET /docs/{id}/snapshot, POST /docs/{id}/sync, GET /docs/{id}/changes
(long poll), POST /docs/{id}/leave; DOC_SEQ poke for WebSocket/SSE rooms.
- Backend accepts editor (write) shares for notes and folders; the forced
read-only migration is gone. The share dialog still offers read only.
- Tests: conflict matrix, order keys, leases, storage adapter round trip,
idempotency, API protocol incl. long poll, and a randomized simulation
(4 clients, lost/duplicated requests and responses, offline phases,
lease expiry) checking convergence, no lost text and exactly-once results.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Phase 0 of the collaboration plan:
- Remove BroadcastToAll; task, task list and workspace folder events now
only reach users who can see the resource. Task list events carry only
the list ID, clients reload with their own permissions.
- Reject unauthenticated WebSocket/SSE connections (401). Every room join
(URL room, JOIN_ROOM, SSE rooms) is checked via a room authorizer;
/events/rooms requires login and only accepts the caller's own sseId.
- Reorder rejects tasks the caller cannot access or edit (IDOR).
- Lift the server WriteTimeout for SSE streams so they are not cut after 30s.
- Fix folder share removal sending DOC_DELETED to the share ID.
- Tests for all of the above; docs and flaw list updated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Remove outdated claims (CRDT, DocRooms, presence, .md files on disk) and
document the actual architecture, API, permissions and known flaws.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
sandrodachandClaude Opus 5.5