Commit Graph
540 Commits
Author SHA1 Message Date
jaredandClaude Opus 5 9363629ea2 feat(security): recovery key leaves the clipboard after 60 s, visibly (#156)
useSensitiveCopy: the recovery key's Copy button becomes 'Copied · clears in
60 s' and counts down; at zero the clipboard is cleared only if it still holds
the key (readText() where permitted — if the browser refuses to read, nothing
is wiped rather than risk eating something else). Any other copy made in the
app cancels the timer. No setting. Verified headless with a fake clock:
countdown ticks, clipboard emptied at 0; copying something else mid-countdown
cancels and leaves that content untouched.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 23:06:02 -04:00
jaredandClaude Opus 5 74d8e3119b feat(security): confirm before opening a link whose text names a different site (#122)
utils/linkSafety.ts compares the registrable domain the visible text claims
(when it looks like a URL/host) with the href's; a mismatch, or a punycode
(IDN) destination, renders the anchor as SuspiciousLink, whose click opens a
confirm — "It shows matrix.lotusguild.org but goes to evil.example." with the
full URL, Cancel / Open anyway (opens in a new tab with noopener). Honest links
are untouched: same registrable domain (youtube.com text over www.youtube.com,
bbc.co.uk over news.bbc.co.uk), plain-word text, mailto:, matrix.to and Lotus
permalinks, anchors with non-text children. Comparator unit-tested (incl. a
Cyrillic-а paypal homograph); verified headless that the phish and IDN
messages are flagged, the honest ones are not, the click shows the confirm and
does not navigate, Cancel keeps you put, Open anyway opens the real target.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 22:57:07 -04:00
jaredandClaude Opus 5 73b1d1e3a7 feat(calls): 'You're muted' nudge when talking into a muted mic (#117)
Host half of the talking-while-muted feature. The fork (0.25.0-lotus.9)
reports speakingWhileMuted for the local participant in io.lotus.call_state —
a level tap on the muted track, so cinny captures no audio. useMutedTalkWarning
polls that flag while muted; after 1.5 s of it being true it shows one sticky
toast, "You're muted — click to unmute", whose click unmutes; it fires at most
once per mute session (re-arms after unmute → mute), is retired the moment the
mic comes back on by any means, and never runs in Push-to-Talk mode. Setting
'Warn me when I talk while muted' (default on) sits next to AFK Auto-Mute.

Verified headless with a tone mic: no toast at +1 s, toast at +4 s, click
unmutes and clears it, second mute re-arms, unmute via the button clears it,
PTT mode → nothing.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 22:50:32 -04:00
jaredandClaude Opus 5 bbe91a24d8 fix(timeline): remember 'Read more' expansion per message for the session (#152)
CollapsibleBody kept its collapsed flag in component state, which the
timeline's windowed rendering throws away when a row unmounts. A module-level
Set<eventId> (never persisted, cleared on reload) now seeds the state, so
scrolling away and back, or a re-render, keeps an expanded message open.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 17:32:15 -04:00
jaredandClaude Opus 5 d2f56817b3 feat(calls): say when answering turns the camera on, offer audio-only (#135)
CI / Build & Quality Checks (push) Canceled after 0s
CI / Trigger Desktop Build (push) Canceled after 0s
CI / Secret scan (gitleaks) (push) Canceled after 0s
CI / Docker image build & smoke test (push) Canceled after 0s
CI / Playwright smoke (e2e) (push) Canceled after 0s
The only join path that skips the prescreen and can start with the camera on
is answering a video call with 'Join with Camera On' enabled. The incoming
dialog and the in-call banner now say 'Answering turns your camera on', label
the primary button 'Answer with camera' and add 'Answer without camera'.
Nothing changes for voice calls or when the setting is off. The other
no-prescreen paths already can't start on camera: the sidebar join uses a
session video pref that is only ever true after the user turned it on
themselves, and it is guarded by the same setting.

Verified headless: video call + setting on → notice + both buttons, 'Answer
without camera' joins with videoEnabled=false; voice call → plain Answer.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 17:26:31 -04:00
jaredandClaude Opus 5 53a80adb57 feat(mobile): long-press a message opens a bottom sheet of actions (#166)
On coarse-pointer devices the hover action bar is unreachable and iOS Safari
has no contextmenu on long-press. useLongPress (450 ms, < 10 px movement,
cancelled by scroll/lift) now opens an ActionSheet — slides up from the
bottom, drag handle, swipe-down or backdrop tap to dismiss — with the quick
reactions row + 'Add reaction' on top and the same folds Menu the desktop
popout uses (extracted to a shared menuJSX) beneath. Android's contextmenu
echo of the same press is dropped; the press-induced text/image selection is
cleared; the finger-lift's synthetic click is swallowed so it can't hit a menu
item; the sheet ignores 'outside' clicks for its first 600 ms. Desktop is
unchanged (right-click → anchored popout).

Verified with Playwright touch emulation (Pixel 7, CDP touch events):
long-press → sheet with 👍 / Add Reaction / Reply / Forward / Bookmark /
Remind Me / Reply in Thread / Copy Link / Pin / Delete; swipe-down dismisses;
a scroll gesture does not open it; tapping Reply closes it and arms the
composer. Still wants a real iPhone + Android pass before it's called done.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 13:59:03 -04:00
jaredandClaude Opus 5 0e2671891f feat(e2ee): undecryptable placeholder says why and offers the fix (#159)
CI / Build & Quality Checks (push) Canceled after 0s
CI / Trigger Desktop Build (push) Canceled after 0s
CI / Secret scan (gitleaks) (push) Canceled after 0s
CI / Docker image build & smoke test (push) Canceled after 0s
CI / Playwright smoke (e2e) (push) Canceled after 0s
'Unable to decrypt message' now carries one sentence per matrix-js-sdk
DecryptionFailureCode (describeDecryptionFailure, unit-tested against every
code so no raw code can leak into the copy) and, where something fixes it,
one button: no key backup → 'Set up key backup'; backup exists but this
session can't open it / key withheld for an unverified session → 'Unlock key
backup' / 'Verify this session' (both open Settings → Devices via a new
settingsRequestAtom that SettingsTab consumes); backup working or unknown
session (rust-crypto re-requests keys itself) → 'Retry', which re-runs
decryptEventIfNeeded. Sender-side problems are plain text. The raw code sits
in the placeholder's tooltip for support.

Verified headless on a fresh session in the encrypted seed room: each event
shows 'Sent before you signed in here, and no key backup exists…' with
tooltip HISTORICAL_MESSAGE_NO_KEY_BACKUP; the button opens Settings → Devices.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 13:23:25 -04:00
jaredandClaude Opus 5 f111b3c9af feat(media): timeline images open the gallery lightbox at that event (#219)
CI / Build & Quality Checks (push) Canceled after 0s
CI / Trigger Desktop Build (push) Canceled after 0s
CI / Secret scan (gitleaks) (push) Canceled after 0s
CI / Docker image build & smoke test (push) Canceled after 0s
CI / Playwright smoke (e2e) (push) Canceled after 0s
Clicking an image in the room or thread timeline now opens the same viewer
the media gallery uses — dark backdrop, sender + date, 1/N counter, ←/→ across
the room's media, +/-/0 and wheel/double-click zoom, download, jump to
message — positioned at the clicked event. RoomMediaLightbox feeds it from the
detached media timeline (#163); when the event isn't in the loaded window it
pages back (bounded, 6 pages) and shows the clicked image alone meanwhile, so
the viewer opens instantly. ImageContent gains onOpenViewer (RenderMessageContent
passes onOpenImageViewer); its built-in viewer remains for stickers, search
results, pins, notifications and avatars.

Verified headless: click → 'Media viewer' dialog focused, counter 1/2, '+' →
120 %, Jump to message present, Esc closes; an older image 10 pages up →
found at 10/45 after paging.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 13:13:28 -04:00
jaredandClaude Opus 5 8658ec05c3 perf: lazy-load seasonal overlays as their own chunks (#167)
CI / Build & Quality Checks (push) Canceled after 0s
CI / Trigger Desktop Build (push) Canceled after 0s
CI / Secret scan (gitleaks) (push) Canceled after 0s
CI / Docker image build & smoke test (push) Canceled after 0s
CI / Playwright smoke (e2e) (push) Canceled after 0s
All 11 seasonal overlays (particles + vanilla-extract keyframes) were in the
main JS and CSS for everyone year-round. Each is now a React.lazy chunk fetched
only when that season is active (auto date or override) or previewed in
Settings → Appearance; the schedule/override decision stays in the main bundle.

check-bundle-size: largest chunk 332.4 kB → 320.1 kB gzip; main CSS 68.4 kB →
56.5 kB; total gzip 1551.6 kB → 1561.3 kB (per-chunk overhead, only paid when
a season is on). Verified headless: no theme chunk requested at startup with
'auto' out of season; the Settings grid fetches the 11 previews; picking
Halloween renders the full-screen overlay (20 particle nodes).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 12:59:28 -04:00
jaredandClaude Opus 5 2e7915d086 feat(a11y): screen-reader announcements for call events (#168)
CI / Build & Quality Checks (push) Canceled after 0s
CI / Trigger Desktop Build (push) Canceled after 0s
CI / Secret scan (gitleaks) (push) Canceled after 0s
CI / Docker image build & smoke test (push) Canceled after 0s
CI / Playwright smoke (e2e) (push) Canceled after 0s
A visually hidden aria-live=polite region (rendered by CallEmbedProvider so it
outlives the embed) announces joins/leaves — batched over 1.5 s: 'alice
joined', 'alice and bob joined', '3 people joined' — your own mute/unmute,
deafen/undeafen and screenshare start/stop, and 'Call ended'. Nothing visible,
nothing audible for anyone else, no setting. Verified headless by observing
the region: bob joined → You are muted → … → bob left → Call ended.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 12:55:01 -04:00
jaredandClaude Opus 5 6df160a7bf fix(upload): plain-language upload failure text instead of the raw MatrixError (#213)
The upload card printed the SDK's toString — 'MatrixError: [413] nope
(http://<hs>/_matrix/media/v3/upload?filename=…)'. describeUploadError() maps
the common cases to one sentence: 413/M_TOO_LARGE → 'This file is larger than
the server allows (limit N)' using m.upload.size when known, 429 → 'Slow down —
try again in a moment.', 401/403 → 'The server refused this upload: <server
text>', 5xx/transport after the retry loop → 'Couldn't reach the server. Check
your connection and retry.', other 4xx → the server's own sentence, URL
stripped. Both card renderers use it; the raw error is still console.warn-ed
by uploadContent for debugging. Unit-tested; verified headless with routed
413/403/503 responses.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 12:31:26 -04:00
jaredandClaude Opus 5 34574178a9 feat(calls): toast when a room admin turns off screen sharing / camera mid-call (#223)
Re-measured the revoke on the current stack (guard fix + fork lotus.4): the
SFU drops the sharer's tracks and EC's own button, the captured tracks and our
control bar all follow within ~2-4 s — the 30 s lag is gone. What remained was
that the button simply vanished with no explanation, so a call-lifetime hook
now toasts 'Screen sharing was turned off by a room admin.' (or the camera
variant) when a track ends while the room policy forbids it.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 01:12:25 -04:00
jaredandClaude Opus 5 070a1ea012 fix(calls): hang up explicitly when answering another call while in one (#195)
Answering from the in-call banner is the only call-to-call switch path; it
started the new embed and let the atom dispose the old one, which just removed
the iframe — our m.call.member in the old room lingered ~17 s (until the
delayed leave expired), so everyone there still saw us in the call.

hangupAndWait moves out of LogoutDialog into plugins/call/hangup.ts and
handleAnswer now hangs up, waits for our membership to clear (bounded 4 s),
disposes the old embed itself (its HangupCall echo would otherwise land after
startCall and clear the NEW embed from the atom — seen in testing), then joins
the new call. Measured headless: old membership gone in 1 s, new call live
with both participants.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 01:04:26 -04:00
jaredandClaude Opus 5 bbcbdad55a refactor(calls): drop the deafen iframe-DOM fallback now that the fork owns it (#209)
CallControl.setSound()/applyScreenshareAudioMuted() no longer poke <audio>.muted
inside the EC frame, and the per-membership re-apply (useCallMemberSoundSync)
is gone: with the pin at 0.25.0-lotus.4 the fork applies deafen and the
screenshare-audio mute through its own renderer, including for late joiners.
Verified headless: deafen set before a second participant joins mutes their
track on arrival; deafen/undeafen leaves the screenshare-audio mute in place.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 00:53:24 -04:00
jaredandClaude Opus 5 a9e0b893be fix(calls): opaque backdrop for the PiP call window (#224)
The fork renders EC transparent so the room wallpaper shows through the
full-size embed; in PiP the element behind the frame is whatever room you
navigated to, so the timeline bled through the participant tiles. The PiP
container now gets the surface colour (the full embed keeps the wallpaper).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-19 00:45:38 -04:00
jaredandClaude Opus 5 50b4e2c16c fix(calls): incoming ring stops on every device once answered or declined elsewhere, or when the caller hangs up (#161)
CI / Build & Quality Checks (push) Successful in 1m30s
CI / Docker image build & smoke test (push) Skipped
CI / Secret scan (gitleaks) (push) Successful in 6s
CI / Trigger Desktop Build (push) Successful in 6s
CI / Playwright smoke (e2e) (push) Successful in 1m39s
The incoming-call dialog only went away on Ignore/Answer/Reject on THAT
device or when the notification lifetime expired, so a DM call answered
on the desktop kept the phone ringing for up to two minutes, and a caller
who gave up left everyone ringing. While a ring is showing we now watch
the room's MatrixRTC session and timeline: our own membership from any
device (answered elsewhere), our own RTCDecline for this ring (declined
elsewhere), or an empty session after it has settled (caller hung up)
all dismiss it. Verified with two alice devices + bob on the local
LiveKit stack: answer elsewhere → dismissed; decline elsewhere →
dismissed; caller End → both dialogs gone in 0.5 s.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-18 22:27:13 -04:00
jaredandClaude Opus 5 230d147ec1 fix(calls): logging out mid-call hangs up first so no ghost MatrixRTC membership is left behind (#29)
Logout stopped the client with the call still joined; the m.call.member
state (expires 4 h) stayed and everyone saw the user 'in call'. The
logout dialog now sends HangupCall and waits (≤4 s) until our own
membership is gone from the room's RTC session before stopping the
client. Verified on the local LiveKit stack: membership count 1 → 0,
logout completes in ~2 s.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-18 22:13:45 -04:00
jaredandClaude Opus 5 53823f5466 fix(media viewer): lightbox focuses itself on open (keys worked only after a click); timeline viewer gets +/-/0 keys, double-click zoom and dialog semantics (#164)
CI / Build & Quality Checks (push) Successful in 1m31s
CI / Docker image build & smoke test (push) Skipped
CI / Secret scan (gitleaks) (push) Successful in 5s
CI / Trigger Desktop Build (push) Successful in 7s
CI / Playwright smoke (e2e) (push) Successful in 2m19s
Gallery lightbox: FocusTrap had initialFocus:false, so focus stayed on the
tile behind the overlay and ←/→/Esc/+/- were dead until the user clicked
inside. Timeline ImageViewer: no keyboard zoom, no role/aria-modal/label,
focus landed on the <img>; now mirrors the lightbox (+ = / - / 0,
double-click toggles 1×↔2×, role=dialog aria-modal labelled by the file
name, focuses on open).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-18 21:29:24 -04:00
jaredandClaude Opus 5 5fc90feef9 feat(forward): Discord-style provenance — 'Forwarded from <sender> in <room> · <time>' with jump to the original
CI / Build & Quality Checks (push) Successful in 1m31s
CI / Docker image build & smoke test (push) Skipped
CI / Secret scan (gitleaks) (push) Successful in 7s
CI / Trigger Desktop Build (push) Successful in 5s
CI / Playwright smoke (e2e) (push) Successful in 2m7s
Forwarded events carried no origin at all; they read as the forwarder's own
words. buildForwardContent now stamps io.lotus.forwarded (sender, ts,
room_id, event_id; re-forwards keep the original stamp) and the main and
thread timelines render a reply-style header above the message that jumps
to the original when the viewer is in the source room (sender + time only
otherwise — the source room's name is not leaked). Unit-tested; verified
end to end with Playwright (header text, event content, jump, re-forward).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-18 21:07:33 -04:00
jaredandClaude Opus 5 2e8244dc67 a11y: SettingTile labels its switch/select with the tile title (#185)
CI / Build & Quality Checks (push) Canceled after 0s
CI / Trigger Desktop Build (push) Canceled after 0s
CI / Secret scan (gitleaks) (push) Canceled after 0s
CI / Docker image build & smoke test (push) Canceled after 0s
CI / Playwright smoke (e2e) (push) Canceled after 0s
Every settings toggle was an icon-only role=switch with no accessible
name (36 axe 'button-name' criticals in User Settings, 3 in Room
Settings). SettingTile now gives the title an id and points any unlabelled
switch/input/select in its 'after' slot at it via aria-labelledby.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-18 19:19:54 -04:00
jaredandClaude Opus 5 22371f8156 fix(composer): no LaTeX conversion inside a typed markdown fence or backtick span (#184 O3)
In markdown mode each paragraph line is serialised before parseBlockMD
joins them, so $x$ inside a ``` fence became data-mx-maths markup
inside the resulting <pre><code> (rendered as math in a code block).
Track fence state across lines and skip math for fenced lines and for
backtick code spans. Unit tests added; verified in the browser.

Also: scripts/dev-homeserver.sh enables MSC4140 delayed events so
scheduled messages work locally.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-18 18:26:15 -04:00
jaredandClaude Opus 5 58a716c734 fix(editor): opening a thread on a pristine composer crashed the app — give each Slate editor its own initial value
CI / Build & Quality Checks (push) Successful in 1m36s
CI / Docker image build & smoke test (push) Skipped
CI / Secret scan (gitleaks) (push) Successful in 7s
CI / Trigger Desktop Build (push) Successful in 6s
CI / Playwright smoke (e2e) (push) Successful in 2m24s
CustomEditor passed one module-level initialValue array to every <Slate>.
slate-react keys its node→path weak maps by node identity, so mounting
the thread composer re-mapped the shared nodes to the new editor and the
main composer threw "Unable to find the path for Slate node" on its next
render, taking the whole client to the error boundary. Anything that had
already edited the main editor (typing, a restored draft) replaced its
nodes and masked the bug, which is why it was intermittent.

Reproduced with Playwright (fresh login → room → click "N replies"):
crashed every time; typing one character first avoided it. Fixed by
creating the initial value per instance (useState).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-18 18:14:05 -04:00
jaredandClaude Opus 5 9d84f9153a fix(invite): the header Invite dialog's QR was still fetched from api.qrserver.com — render it locally (#192)
a899d7d3 moved the Room Settings share QR to qrcode.react but the
InviteUserPrompt (header ⋮ → Invite, Room Intro, space menu) kept the
third-party <img>, which leaked the room link and — since the prod CSP
img-src no longer allows that host — rendered as a broken image. Same
QRCodeSVG + white quiet zone as RoomShareInvite. Verified with
Playwright: SVG present, no remote <img>, zero external requests.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-18 00:47:37 -04:00
jaredandClaude Opus 5 470b5217ae fix(composer): one-row toolbar with uniform 32px buttons on every viewport
CI / Build & Quality Checks (push) Successful in 1m53s
CI / Docker image build & smoke test (push) Skipped
CI / Secret scan (gitleaks) (push) Successful in 6s
CI / Trigger Desktop Build (push) Successful in 6s
CI / Playwright smoke (e2e) (push) Successful in 1m38s
The composer looked off in two ways, both confirmed by rendering CustomEditor
with RoomInput's exact props and measuring the buttons headlessly:

Desktop: the Lotus additions (location, poll, voice, schedule) used
`Icon size="100"` (18px) inside the same `IconButton size="300"` as the
upstream Aa/sticker/emoji/send buttons (24px icons), so one row mixed
32×32, 26×26 and a 28×19 "GIF" text stub. Every button is now 32×32: the
four small icons use the default icon size and the GIF label sits in a
1.5rem box, the same footprint as an icon. The mic's idle button in
VoiceMessageRecorder gets the same treatment since it lives in this row.

Phones: d6159997 let the before|editable|after row flex-wrap at <=750px, but
folds' Scroll (the editable's wrapper) is `width: 100%`, so the row ALWAYS
broke into three stacked lines — "+" alone on top, the input flush against
the left edge on its own line (the :first-child padding selectors no longer
matched), and emoji/draft/send left-aligned underneath. e1bb8301's "+"
overflow menu was meant to produce [ + | input | emoji | send ] but never
could while the row wrapped. The row no longer wraps (upstream behaviour);
instead the collapse into the "+" overflow is keyed on the viewport
(ScreenSize.Mobile) as well as the touch UA, so a phone-width window on a
desktop UA — iPad desktop mode, split-screen PWA, docked window — also
collapses instead of rendering ten controls inline and clipping Send behind
the editor's overflow:hidden. The "Draft saved" label moves into the overflow
row in compact mode so the inline row stays [ + | input | emoji | count |
send ]. The editable's vertical padding grows to 19px at phone width (only
when the row actually has buttons) so the text sits level with the 44px
touch targets instead of hugging the top of the row. Those touch targets now
also apply the shared MobileTouchTarget class, matching the recorder button.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-17 01:14:48 -04:00
jaredandClaude Opus 5 3dead4b3e1 fix(calls): incoming-call ringtone respects quiet hours, DND, Focus Assist and snooze
CI / Build & Quality Checks (push) Canceled after 11s
CI / Trigger Desktop Build (push) Canceled after 0s
CI / Secret scan (gitleaks) (push) Canceled after 0s
CI / Docker image build & smoke test (push) Canceled after 0s
CI / Playwright smoke (e2e) (push) Canceled after 0s
The "should we make noise" predicate used for message sounds is extracted
into useNotificationsQuiet() (unit-tested) and applied to the ringtone in
both the full-screen incoming-call overlay and the compact in-call banner.
The overlay/banner still show so the call can be answered; only the audio
is skipped. Join/media paths untouched.

Fixes #28

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-15 22:43:17 -04:00
jaredandClaude Opus 5 e9d07513d2 merge: upstream cinny v4.12.7 into lotus
Brings the 4.12.7 fixes: "%" sign wrapping below the composer input and
emoji autocompletion overwriting the preceding element. The release's
security half (sanitize-html, react-router-dom) was already applied in
91def3ad (we are on newer versions than upstream ships).

Resolved: kept our package.json/lockfile (the fork is ahead on every
dependency — React 19, router 7, i18next 26…), version 4.12.7-lotus, our
README/CONTRIBUTING, our "Lotus Chat vN" branding via pkg.version, and
the deleted GitHub-only workflows stay deleted. Note upstream announces
it is replacing matrix-js-sdk with its own SDK and pausing PRs; future
syncs will get harder.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-15 13:07:32 -04:00
jaredandClaude Opus 5 2e14980de6 fix(settings): validate seasonalThemeOverride; commit page zoom on blur; GIF picker opt-in
- seasonalThemeOverride is coerced to 'auto' when the persisted value is
  outside the union, like every other enum setting (tested).
- PageZoomInput commits on blur as well as Enter.
- New gifPickerEnabled setting (default off) with a disclosure that GIF
  searches go to Giphy; the picker doesn't mount and the composer button
  is hidden while off.

Fixes #74
Fixes #75
Fixes #68

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 20:28:41 -04:00
jaredandClaude Opus 5 6b16c789a4 fix(seasonal): Earth Day globe clears the composer strip
Fixes #72

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 20:28:41 -04:00
jaredandClaude Opus 5 a25777a77a fix(privacy): location map loads on click; CollapsibleBody tracks reduced motion live
m.location mounted an openstreetmap.org iframe on render, acting as a
read beacon; it's now behind a "Load map" button. CollapsibleBody uses
useReducedMotion() instead of a one-time matchMedia snapshot.

Fixes #66
Fixes #85

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 20:28:41 -04:00
jaredandClaude Opus 5 3adba22ddf fix(security): bound KaTeX rendering of remote LaTeX
maxSize 10, maxExpand 100, trust false, strict ignore, and a 5000-char
cap (rendered as plain text beyond it), so a remote sender can't blow out
layout or CPU with \rule{99999em} or deep macro expansion.

Fixes #65

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 20:28:41 -04:00
jaredandClaude Opus 5 49c4641ca8 fix(calls): warn before unload during a joined call
Register beforeunload while joined so Ctrl+R / tab close prompts instead
of silently dropping the call. Also removes the setPipMode effect (#59).

Fixes #58

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 20:28:41 -04:00
jaredandClaude Opus 5 a29be7953b fix(soundboard): upload cap counts staged clips once and per batch
CI / Build & Quality Checks (push) Successful in 1m26s
CI / Trigger Desktop Build (push) Successful in 28s
The guard double-counted staged uploads and read a stale count for every
file in a batch, so a 60-file drop bypassed the cap while a nearly-full
pack refused early. Partition the batch with a running count.

Fixes #31

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 19:46:06 -04:00
jaredandClaude Opus 5 23ee156f2f fix(calls): PTT, deafen hotkey and AFK auto-mute live for the whole call
They were wired inside CallControls, which only renders while the call
room is selected, so navigating away (PiP) or opening the in-call chat on
mobile silently disabled all three — AFK auto-mute exactly when it
mattered. Move them into useCallHotkeys + useAfkAutoMute mounted from the
embed-lifetime CallUtils, gated on joined; CallControls keeps only the
PTT chip (pttActiveAtom).

Also: window blur/focus release the mic only while a PTT key is actually
held, so a deliberate hands-free unmute survives a click into the iframe
(#27); iframe-side listeners re-bind on the iframe load event so they
survive an EC reload (#60). The #23 modifier/interactive guards are
preserved and unit-tested.

Fixes #9
Fixes #27
Fixes #60

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 19:46:05 -04:00
jaredandClaude Opus 5 6bd2903de1 fix(profile): private notes no longer lose typed text
The store-sync effect re-applied the stored note on every notification,
including the echo of the note's own save, reverting text typed after
the debounce fired; and closing the panel inside the 800ms debounce
dropped the pending save. Add a dirty ref that suppresses the resync
while there are unsaved edits, and flush the pending save on unmount.

Fixes #18

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 14:48:35 -04:00
jaredandClaude Fable 5.1 c4aa1567d7 fix(call): answering an incoming call honours cameraOnJoin and mic/sound prefs
handleAnswer hard-coded { microphone: true, video, sound: true }, so
pressing Answer on a video call published the camera even with the
"Join with Camera On" setting off (the default) and ignored a persisted
muted/deafened preference. Build the answer prefs the same way every other
join path does; PTT's forceAudioOff still applies downstream.

Fixes #8

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 02:05:34 -04:00
jaredandClaude Fable 5.1 a6ddafb446 fix(seasonal): Halloween fog layers drop willChange under reduced motion
The three fog bands pinned compositor layers even when their animation was
'none' (always the case in the settings swatch). Match the sibling layers.

Fixes #73

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 02:04:55 -04:00
jaredandClaude Fable 5.1 4bea48959e fix(seasonal): keep Arcade + Deep Space overlays out of the chat text
CI / Build & Quality Checks (push) Successful in 1m39s
CI / Trigger Desktop Build (push) Successful in 7s
Arcade:
- perspective grid is masked through the central timeline column so it
  frames the chat instead of striping messages; lower opacity, thinner
  non-scaling strokes (also fixes the grid vanishing in the 76px swatch)
- neon horizon line is broken through the middle so it never underlines
  a message; softer sun glow
- "1UP" score + "INSERT COIN" moved from top-left/bottom-centre (space
  bar / composer) to a single bottom-right HUD stack; font-size clamps
  to 0 via cqw inside the settings swatch so no more clipped glyphs
- scanlines: 4px pitch at lower opacity (light theme was fully striped);
  ambient wash slightly lighter

Deep Space:
- starfield 28 → 62 stars across the two parallax layers
- base violet wash slightly lighter for light themes

SeasonalPreview wrapper is now a size container (container-type) so
overlays can scale/hide fixed-px details with cqw. LOTUS_FEATURES.md
theme table now describes what both overlays actually render.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
2026-09-12 01:29:33 -04:00
wjaaaaaaat c434e0dda3 fix: emoji autocompletion overwriting preceding element (#3064)
* fix getPrevWorldRange to exclude empty text children

this prevents the range from encroaching on the node of elements like emojis and pings to prevent them from being overwritten

* fix typo of "word" as "world"

* trigger pr checks
2026-08-11 14:23:15 +10:00
jaredandClaude Opus 4.8 477df4ae32 fix(composer): stop the intermittent autocomplete-insert crash
Picking an autocomplete item (mention/emoji/command — all inline voids)
occasionally tripped the composer error boundary, forcing a page refresh, even
though the element had already inserted. Root cause (traced through slate-react):
moveCursor deferred its cursor work to setTimeout(0), leaving the caret on the
just-inserted void's zero-width edge whose DOM (a U+FEFF node) isn't populated on
that tick. slate-react's commit-phase selection sync then calls
setBaseAndExtent(voidEdge, 1) and throws IndexSizeError mid-render → boundary.

Prevention: do the cursor work SYNCHRONOUSLY, in the same commit as the insert —
Transforms.move (escapes the void into the real trailing text node) then
insertText(' '). The caret is then always a resolvable text point when the
selection sync runs. (moveCursor's focus stays deferred+guarded, unchanged.)

Recovery (belt-and-suspenders): the composer error boundary is now recoverable —
a "Reload composer" button (resetErrorBoundary) + onReset Transforms.deselect
clears a transient bad selection so it remounts with the draft intact, no page
refresh. + role="alert" for screen readers.

Three review agents: two root-caused the exact slate-react throw and proved the
try/catch-only version merely recovered; a third reproduced the transforms
headlessly and caught that a first "sync insertText WITHOUT move" attempt hit
Slate's void guard (space dropped, caret trapped) — the move is required to
escape the void. Not unit-testable (needs the live DOM + the timing race).
Gate-green (tsc, eslint, prettier, 925 tests, build).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-26 16:12:52 -04:00
jaredandClaude Opus 4.8 8a461610f4 fix(low-tail): MLocation permalink uses validated floats; PolicyListViewer doc
- MLocation "Open in OpenStreetMap" permalink built its URL from the raw
  parseGeoUri strings (location.latitude/longitude) while the embedded map iframe
  used the parseFloat + isFinite-validated lat/lon. Use lat/lon in the permalink
  too, so a malformed geo: substring can't reach the URL (they're already proven
  finite a few lines above and used identically in mapSrc).
- LOTUS_FEATURES claimed the Policy List Viewer has "Subscribe (join) /
  unsubscribe (leave) controls for each list" and lists subscribed lists.
  Verified against PolicyListViewer.tsx: it's a room-ID/alias input viewer that
  displays a joined policy room's rules read-only — no subscribe controls, no
  subscribed-lists listing. Corrected the doc to match.

Two low-tail bug-hunt findings from LOTUS_TODO. Gate-green (tsc, eslint,
prettier, build).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-24 19:57:34 -04:00
jaredandClaude Opus 4.8 08e191008b fix(call): release auto-enabled pip spotlight when returning to the call room
In picture-in-picture with an active screenshare, spotlight is auto-enabled so
the share fills the pip window (tracked via pipAutoSpotlightRef). The release
branch sat behind `if (!pipMode) return`, so navigating BACK to the call room
(pipMode → false) early-returned and never released it — the spotlight stayed
stuck on with the ref latched true.

The effect now guards only on `!callEmbed`, computes wantSpotlight = pipMode &&
pipScreenshare, and releases whenever that's false (screenshare ends OR pip
ends). The ref still gates release so we only ever undo a spotlight we enabled,
never the user's. Two reviewer-prescribed hardenings folded in: reset the ref
when callEmbed is torn down (kills a stale cross-call latch), and a comment that
control.spotlight is deliberately not a dep (re-adding it would fight the user).

Bug-hunt finding from LOTUS_TODO. Two review agents verified against
CallControl.ts (ref-gating, deps, idempotency, cross-embed self-heal); [live] —
the code fix is unambiguous but confirming screenshare→pip→back wants a real
call. Gate-green (tsc, eslint, prettier, 925 tests, build).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-24 19:54:29 -04:00
jaredandClaude Opus 4.8 d416c62b4c fix(seasonal): auto theme re-evaluates over time; auto clears chat background
- The "auto" seasonal theme was computed once at mount, so a long-lived session
  never crossed a season/holiday-window boundary. SeasonalEffect now re-evaluates
  on an hourly ticker (auto mode only) AND refreshes on entering auto — the
  interval only runs while auto, so a stale mount-time timestamp would otherwise
  resurface on a pinned/off → auto switch (the exact frozen-at-mount bug, caught
  in review). The decision is extracted to a pure resolveSeasonTheme(override,
  now) in seasonSchedule.ts (removing an unsafe cast) and unit-tested.

- Selecting seasonal "auto" while a chat background was set was a silent no-op:
  the seasonal picker only cleared the background for a *specific* theme, and the
  overlay is suppressed while a background is set. Now any active seasonal mode
  ("auto" included) clears the background; only "off" leaves it — symmetric with
  the background picker (which sets seasonal "off"). The overlay guard stays as a
  backstop for legacy persisted state.

Bug-hunt findings from LOTUS_TODO. Three review passes (the 2nd caught the
switch-into-auto staleness); +2 unit tests. Gate-green (tsc, eslint, prettier,
922 tests, build).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-24 17:34:51 -04:00
jaredandClaude Opus 4.8 29ff16546a fix: avatar-decoration live-update + CDN override + profile 404; DND badge color
Avatar decorations (useAvatarDecoration.ts / ProfileDecoration.tsx):
- invalidateDecorationCache now notifies a per-user listener set (and clears the
  give-up counter), so changing your own decoration updates mounted avatars
  (timeline, member list) live instead of only after a remount. Concurrent
  re-fetches de-dupe via the existing `pending` map.
- Picker grid thumbnails use decorationUrl() instead of the raw DECORATION_CDN
  literal, so a VITE_DECORATION_CDN override no longer breaks the grid while
  real avatars work.
- Settings reads the full /profile/{userId} instead of the /{field}
  sub-resource, which 404s (console error) for anyone without a decoration set
  — matching the pattern already used by useAvatarDecoration.

Presence (Presence.tsx): PresenceBadge renders DND (unavailable + status 'dnd')
as red "Do Not Disturb" to match PresenceRingAvatar and the settings picker;
it was the lone outlier showing a yellow "Idle".

Bug-hunt findings from LOTUS_TODO. Two review agents (correctness +
upstream-behavior); gate-green (tsc, eslint, prettier, 914 tests, build). Both
flagged only pre-existing edge notes (in-flight piggyback staleness, 'dnd'
free-text collision shared with the ring avatar) — neither introduced here.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-24 15:37:16 -04:00
Claude e1bb8301f0 fix(composer): collapse mobile action buttons behind a "+" overflow menu
On phones the composer's 7-8 secondary action buttons wrapped into a tall
multi-row stack ("massive height"). Mobile now shows a single compact row —
[ + | input | emoji | send ] — where "+" toggles a collapsible row (above the
formatting toolbar) holding attach, GIF, location, poll, voice, formatting and
schedule. Desktop is unchanged (isMobile === false; the mobile branches are
never entered and composerOverflow stays null).

The after-builder stashes the collapsed buttons in a render-local `let` that
the bottom slot reads; safe because JSX props evaluate in source order within
one render (verified by review). Emoji/Send stay inline; the emoji and GIF
PopOut anchors still resolve wherever their button renders.

Review fixes folded in: the "+" toggle uses aria-expanded + aria-controls
(dropped the redundant aria-pressed) pointing at the labelled role="group"
overflow row; the voice recorder's idle mic button gets the @media-gated
MobileTouchTarget 44px target so the overflow row is uniformly tappable.

Two review agents (correctness + UX/a11y); gate-green (tsc, eslint, prettier,
914 tests, build). Visual confirmation still wants a real device per
LOTUS_TESTING.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-24 14:44:13 -04:00
Claude 4154cae55a fix(embeds): animate GIF previews; add Mixcloud/Deezer; misc embed fixes
CI / Build & Quality Checks (push) Successful in 11m21s
CI / Trigger Desktop Build (push) Successful in 34s
GIF previews rendered but never played: Synapse's /thumbnail endpoint
flattens animated GIFs to a still first frame. GifCard and the generic OG
card now request the original via /download (no width/height) for GIFs, so
they animate. Guarded with shouldServeGifOriginal(): a matrix:image:size cap
(10 MB) keeps a huge self-hosted GIF on the frozen thumbnail, and the generic
card's eager <img> gains loading="lazy" (it was the one preview image missing
it) so originals stay off the wire until near the viewport.

Also adds Mixcloud + Deezer inline media embeds (iframe widgets via
parseMediaEmbed/MediaEmbedCard, matching the existing click-to-play pattern),
and fixes Deezer podcast links: they live at /show/<id>, not /podcast/<id>
(the latter 404s on Deezer's own oEmbed) — verified against the live API.

Reviewed by two agents; both findings (Deezer /show, GIF eager-load) fixed
and covered by tests. Desktop Tauri frame-src CSP updated separately.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-24 00:21:10 -04:00
jaredandClaude Opus 4.8 ef82650cf7 feat(embeds): detailed Steam store / news / app-widget embeds
Recognize store.steampowered.com content URLs and render each richly, within
the existing privacy-first facade. 2-agent reviewed (both SHIP).

- getSteamTarget / steamWidgetEmbedUrl (videoEmbed.ts, +tests): classify
  /app/{id}, /news/app/{id}/view/{gid}, /(bundle|sub|dlc)/{id}; non-content
  pages (home/search/wishlist) and other hosts fall through to the generic card.
- SteamCard now dispatches:
  - app → OG capsule header + click-to-play facade → Steam's OFFICIAL store
    widget iframe (store.steampowered.com/widget/{id}): live region-aware price,
    discount %, Buy on Steam. Nothing loads from Steam until "Show price &
    store" is pressed; gated by the inlineMediaEmbeds setting. App pages use the
    wide card so the ~646px widget has room.
  - news → rich announcement card (banner + headline + body preview + link) —
    your example URL previously fell through to the plain generic card.
  - bundle/sub/dlc → the OG store card.

Grounded in our CSP: the widget works via frame-src https: (no infra change),
images route through the homeserver (img-src excludes Steam), and there is NO
client-side Steam API call (connect-src + Steam CORS both block it) — which is
also the honest ceiling: no review scores/genres client-side, price/buy come
from the official widget.

Runtime QA still needed: the live widget iframe rendering (height/fit) can't be
verified headlessly.

Gates: tsc 0, eslint 0, prettier clean, 912 tests, build ok.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-21 21:41:06 -04:00
jaredandClaude Opus 4.8 f2673effe4 fix(embeds): parsing over/under-match + broken thumbnails + wide layout
Bugs found by a 3-agent audit of the inline-embed system (core posture —
sandbox, postMessage origin+source, XSS, noreferrer, oEmbed — verified sound);
fixes reviewed by 2 agents on the staged diff (both SHIP).

Parsing (videoEmbed.ts, + tests):
- Twitch/Kick/SoundCloud/Streamable reserved-path exclusion — their own utility
  pages (twitch.tv/directory, kick.com/browse, soundcloud.com/discover/…,
  streamable.com/login, bare /videos) no longer render as broken player embeds.
- SoundCloud: `/<artist>/<tab>` profile-tab listings excluded; `/<artist>/sets/<slug>`
  real sets still detected.
- Vimeo: unlisted-hash capture constrained to lowercase-hex, so a normal video's
  trailing segment (/likes, /settings, a slug) isn't captured as a bogus `h=`
  param that Vimeo then rejects.

Rendering (UrlPreviewCard.tsx, RenderMessageContent.tsx):
- Spotify/Steam/Discord/IMDb route og:image through mxcUrlToHttp like every other
  card — a raw og:image is an mxc:// URI (broken <img> on standard Synapse) or an
  off-homeserver request that defeats the click-to-play facade.
- `wide` card class now follows the RESOLVED embed (incl. the og:url short-link
  fallback), so an og:url-resolved player gets the wide layout, not a cramped one.
- Twitter host detection (isTwitter/isTwitterTweet) aligned with getTweetId —
  mobile.twitter.com and legacy /statuses/ now route to the Twitter card/embed.
- De-dupe preview URLs so a message repeating a link doesn't render sibling
  cards with identical React keys.

Gates: tsc 0, eslint 0, prettier clean, 910 tests, build ok.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-19 23:35:59 -04:00
jaredandClaude Opus 4.8 a267e9e960 fix: low-tail correctness — thread notifs, call audio, OIDC expiry
Verify-then-fix batch of minor bugs; each staged diff reviewed by 2 agents
(both SHIP). Two listed items (N6 receipt-avatar refresh, H10 room-name
length reject) were already handled and left unchanged.

Threads:
- T5: a just-sent reply no longer under-notifies — `participated` also checks
  the local thread timeline for our own events, since the server-bundle
  `hasCurrentUserParticipated` lags.
- T6: a room set to "Mentions & Keywords only" no longer over-notifies Default
  thread replies — new `roomMentionsOnly` gate (behavior-identical when false;
  +4 unit tests).
- T7: thread-mode account-data writes are serialized with content carried
  forward (setAccountData is a bare PUT whose result lags the /sync echo, so
  plain serialization wouldn't stop the lost update); carry only on success.

Calls / audio:
- C-L2: a real incoming ring cancels a lingering Settings ringtone preview.
- C-L3: the ringtone AudioContext is primed on the first page gesture (via the
  always-mounted CallEmbedProvider) so the first ring after a cold load isn't
  silent.
- C-L5: useCallSpeakers depends on a stable boolean, so the tile MutationObserver
  + io.lotus.call_state subscription aren't rebuilt on every membership change.

Crypto:
- F5: the OIDC refresher forwards the freshly-refreshed token expiry
  (passed on the tokens object at runtime) as expiresInMs, so the persisted
  expiresAt no longer goes stale across reloads.

Gates: tsc 0, eslint 0, prettier clean, 860/860 tests, build ok.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-19 02:47:27 -04:00
jaredandClaude Opus 4.8 72e7447d28 fix(mobile): stack embed cards + secondary 44px touch targets (r2)
Mobile follow-ups round 2 (survey findings deferred from the mobile audit),
reviewed by 2 agents on the staged diff (both SHIP).

- URL-preview cards: the Twitch / Twitter / TikTok-fallback cards render
  their thumbnail/header BESIDE the content as direct children of the
  UrlPreview flex row, which squeezes both on a phone. Add `StackOnMobile`
  (@media max-width:750px -> flex-direction:column) scoped to those variants
  via cardClass. folds Box has no default `direction`, so the override wins
  uncontested; desktop (>750px) is unchanged. No-op for the single-column
  embed cards (MediaEmbedCard/TikTokEmbedCard).
- 44px touch targets (MobileTouchTarget, @media max-width:750px) on the
  otherwise ~28px controls: embed-player Close/Collapse/Fullscreen/View-post
  buttons; image-viewer close/zoom/download; the read-receipt "seen by" pill.

Deferred (rationale, not built): PiP resize handles + fullscreen button —
enlarging four 24px corners to 44px would swallow a ~160px mobile PiP and
block "Return to call"; presence dot is a non-interactive status indicator.

Gates: tsc 0, eslint 0, prettier clean, 856/856 tests, build ok.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-19 01:56:55 -04:00
jaredandClaude Opus 4.8 c3e1fbfff5 fix(a11y): honor prefers-reduced-motion for avatar decorations (P2)
Avatar decorations are animated APNGs and were the only motion feature not
gated on prefers-reduced-motion (chat backgrounds / seasonal overlays all
suppress motion under it). Since there's no static-frame asset to freeze to,
render just the avatar (no decoration overlay) when the user prefers reduced
motion — the only motion-respecting option. Users without the preference are
unaffected; live OS-toggle is reactive via useReducedMotion. Also relieves the
mobile perf drain of dozens of live APNGs in scrolling lists.

Reviewed: correct a11y behavior, hooks-safe, no layout dependency on the overlay.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-19 00:00:07 -04:00