| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Expose invite_ttl_hours, pair_ttl_hours, device_request_ttl_minutes,
max_concurrent_streams and transcode_incompatible_video in the Node
management panel. Changes are applied immediately via roster.db and
written back to node.toml so they survive a DB wipe. On startup,
roster overrides take precedence over node.toml defaults.
Draft v6 §2.11 documents the design; MNP gains node_settings_set /
node_settings_set_ack for the browser path.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
| |\ \
| |/
|/| |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
1. The "Search Files" <h2> is removed from the search page — it duplicated
the nav and cost a line of vertical space on mobile. `search.title` is
dropped from all ten locales (its only use).
2. A monochrome refresh button sits left of the Files breadcrumb home
icon. It calls `onRefreshIndex` (re-fetch the group index from the
node — the file list is a view over a cached copy with no other way to
pull a fresh one), spins while in flight, and only renders where an
`onRefreshIndex` is wired (the real group view, not the cross-group
search results). New `refresh` icon; `group.refresh_index` in all ten
locales.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018BMLQjqFGCize2KtNBT79v
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Root of the 2026-08-29 demo35 storms. `_do_media_meta_request` could not
tell "this is a movie" from "this video isn't enriched yet" — both have
season/episode None — so during a slow initial scan with a browser on the
Videos tab, every show episode requested was run through the *movie*
search path with its raw filename as the query
(`search/movie?query=Show S01E01 1080p WEB DL ...`), hundreds per second,
until TMDB rate-limited and posters stopped loading. Worse, an
un-enriched show episode's own valid cached "tv" match was treated as
stale (its provisional kind was "movie"), so a file already resolved got
re-queried anyway.
- While a video is un-enriched (no display_title — enrich.py always sets
one), never run a TMDB *search*. Serve the cached match if the content
hash has one, honouring the cached kind ("tv"/"movie") rather than the
provisional split; otherwise answer confidence 0.
- Once enriched, the strict `cached_media_type == media_type` check is
unchanged: an enrichment fix that reclassifies a folder movie->tv still
drops the stale match and re-resolves.
- video-app.js: `useMediaMeta` gains an `enrichSig` dependency
(`entry.display_title`) so the client refetches once the enriched fields
arrive on an index delta — the fileId is a content hash and never
changes, so nothing else would retrigger it.
Not caused by the V8-V13 work; it raised the per-file call count so the
pre-existing race became a visible storm.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018BMLQjqFGCize2KtNBT79v
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A one-click alternative to the full "Fix match" search-and-pick flow, and
reachable without SSH (`meshbay-node video rematch` clears a whole group).
- MNP 0.13: tmdb_rematch / tmdb_rematch_ack (additive — an older node
logs "unknown type", the button just does nothing). OP_TMDB_REMATCH,
signed like tmdb_override (media_cache is shared node-wide).
- media_cache.drop_tmdb_match(file_id): forgets the match AND the
override marker — deliberately stronger than clear_file_tmdb, since the
operator is explicitly asking for a fresh resolution.
- webrtc_server: _do_tmdb_rematch / _admin_exec_tmdb_rematch, dispatch +
admin-response routing, broadcasts tmdb_rematch_ack.
- transport.js: rematchTmdbMatch(fileId, signFn); 'tmdb_rematch' in the
admin-op allowlist; tmdb_rematch_ack handled like tmdb_override_ack.
- video-app.js: a "Re-match" button beside "Fix match" in the detail
modal (isNodeAdmin), then bumpMediaMetaGeneration(). video.rematch_one
key in all ten locales.
docs/mediacenter.md §10.1: V8–V13 marked done.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018BMLQjqFGCize2KtNBT79v
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Two movie files that TMDB resolves to the same id (the same film at two
resolutions, or the same rip in two folders) now collapse to one poster
card, mirroring the existing show merge — same keying discipline so an
unmerged movie keeps its card and a merge updates props rather than
remounting. The detail modal lists the versions (resolution · duration ·
size), each a Play button, when there is more than one; a single-file
movie is unchanged. New `video.versions` key in all ten locales.
Known edge, noted in §10.1: "Fix match" on a merged movie corrects only
the representative file; the other version un-merges and can be corrected
on its own.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018BMLQjqFGCize2KtNBT79v
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
A batch of wrong poster-grid matches found live on a real library
(2026-08-29): a two-volume film's second part matched the first; a
numbered sequel matched a same-year making-of documentary; several
entries of one franchise matched a single early entry whose localized
TMDB title is the franchise name; one matched nothing. One mechanism:
_tmdb_search returned the first candidate query whose title-similarity
ratio merely cleared 0.6, before alternative_title / the Roman-numeral
variant was ever tried.
Matching:
- title_parse: fold guessit's volume/part number back into display_title
so the parts of a multi-part film stay distinct in the query, the card
and the override.
- _tmdb_search: keep a strong PASS 1 fast path (ratio >= 0.85, one
request), otherwise score every candidate query and pick the best. A
year-exact rescue lifts a sub-0.6 top hit to the confidence floor only
when TMDB's own year-filtered result lands exactly on the filename's
year. No local re-ranking of any single result list; no tmdb.py change.
Fix match / rematch:
- _admin_exec_tmdb_override: a movie override touches its own file only
(guessit gives a whole franchise one display_title); a show override
still fans out. Corrected files are marked in media_cache.tmdb_override.
- media_cache: tmdb_override table; clear_file_tmdb / clear_tmdb_matches
drop auto-resolved matches while sparing manual corrections.
- ops.rematch_video + `meshbay-node video rematch` (loopback endpoint +
CLI verb): re-resolve a group's video matches after a matcher fix.
file_tmdb is keyed by content hash and otherwise only pruned on
deletion, so nothing dislodged a cached match before.
- a rename now drops the stale auto match too (daemon
_reenrich_renamed_video_entries).
UI:
- VideoDetailModal shows the source filename and resolved TMDB id; an
unmatched poster gets a badge (3 new video.* i18n keys x 10 locales).
So a wrong match can actually be identified before hitting Fix match.
docs/mediacenter.md 10.1 records this and the V8-V13 follow-up backlog
(show-branch ladder, year-aware _best_match, wider sequel_variants, the
0.6-0.85 extra calls, movie grid merge, per-card rematch).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018BMLQjqFGCize2KtNBT79v
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
spinner
Three regressions fixed in ChatPanel:
- Scroll pinning: layout effect now depends on [messages, hasMore] so the
"load older" button appearance triggers a re-pin; setHasMore is called
before setMessages to avoid an intermediate render without the button;
fit() is wrapped in fitAndPin() so panel resizing re-pins the scroll.
- Message loading on group switch: fetch effect depends on the status prop
instead of transportRef.current?.connected to avoid racing with
GroupPage's cleanup; loadedRef resets on fetch failure so a retry works.
- Connecting spinner restored in the Chat tab empty state.
Adds structural regression tests (test_chat_scroll_bottom.py) that lock
the dependency arrays and setState ordering so these invariants break
loudly in CI.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
| |/
|
|
|
|
|
|
|
| |
App components were keyed by app name only (e.g. 'chat'), so switching
groups while staying on the same tab reused the component instance —
loadedRef stayed true and messages state was never cleared. Now keyed
by app name + groupId, forcing a clean remount on group change.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Wizard (Electron):
- Auto-provisions node config (hub URL + username) from logged-in user
- node:start handles both cold start and restart of misconfigured daemon
- Waits for daemon to reach 'running', auto-links node key on hub
- probeNode accepts intermediate states for wizard progress feedback
Reset (meshbay-node reset):
- Unlinks node key from hub (DELETE /me/node_key, best-effort)
- Stops and disables daemon (systemctl --user disable --now)
- Erases ~/.config/meshbay, ~/.local/share/meshbay, ~/.local/state/meshbay
MusicBrainz contact:
- Resolved from owner's hub email instead of per-node roster config
- Removed musicbrainz_contact UI and WebRTC handshake field
- Removed set_musicbrainz_contact/musicbrainz_contact from roster
Node pairing:
- Added operator pairing banner on NodePage
- Added operator_paired flag to list_groups
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
chat_hist_resp checked only the single oldest pending entry across all
types. When switching tabs, keyed requests (media_meta_req, file_req)
from Videos/Music/Photos stayed in _pending and were older than the
chat_hist entry, so the response was dropped and the chat appeared empty
on return. Now scans for the first chat_hist entry instead. Same fix
applied to index_sync dispatch for consistency.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
| |
Search-page connections (indexing and pool) now time out after 10s
instead of the transport's default 30s, so down nodes are skipped
faster. Downloads show a user-facing error when a group is unreachable.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Downloads from search: FilesPanel accepts a getTransport callback for
cross-group file downloads. Select/Download are available in search;
delete and zip are hidden (readOnly). Each download connects to the
entry's group via the pool on demand.
Photos view: PhotosApp reused in search with the same pattern as
Videos/Music — synthetic root, per-entry transport refs, group badges
on album cards. Lightbox uses per-entry transport for full-resolution
image fetches. Zip download hidden in search context.
Pre-connecting: connections to all groups start as soon as indexing
finishes, regardless of the active view. Thumbnails cache across pool
evictions, so they're ready when the user switches views.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Search page fetches indexes from all groups, then renders consolidated
entries through the existing FilesPanel, VideoApp, and MusicApp
components — no reimplemented views. Groups appear as top-level
directories in the file browser; video/music entries use a synthetic
root with per-entry transport refs for thumbnails and metadata across
groups. Music player lifted to app.js with getConnection(groupId) for
cross-group playback. Connection pool (max 3, LRU eviction) manages
lazy WebRTC connections. All 10 locales updated with search keys.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
The button detaches a group from this node's config. For a group that still
exists on the hub that strands it with no host; deleting such a group is the
group Settings tab's job (detach + hub delete together). It now shows only
when the group is `stale` — present on the node, gone from the hub — which
the page already computes for the "not on hub" badge.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |
|
|
|
|
|
|
|
|
|
| |
The @owner line turned the sidebar name into a two-line block, and the
status dot (a flex sibling, align-items:center) floated to its vertical
middle. A group entry is now a flex column: dot+name on one row (.si-head),
@owner on a subordinate line indented past the dot. Non-group sidebar links
are untouched (.sidebar-item stays a row; only .sidebar-group flips).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A group's identity stays its UUID. What changes is that "the name is
unique" — until now an unenforced expectation — becomes real, scoped to the
owner account, and the owner's username is surfaced so two groups called
"photos" on different nodes can be told apart.
Hub:
- `groups` gains a functional unique index `uq_groups_owner_name` on
`(admin_id, lower(name))` (model + migration c3d4e5f6a7b8). The migration
pre-flights: if the data already clashes it aborts and lists the
offending (admin_id, name) pairs rather than renaming anyone's group.
meshbay.org checked clean.
- `create_group` trims the name, rejects blank (422) and an owner-scoped
case-insensitive clash (409), with an IntegrityError backstop for the
race, and returns `owner_username`.
- `owner_username` added to `/v1/groups/mine`, `GET /v1/groups` (local
rows), `POST /v1/groups/{id}/join`, `GET /v1/admin/groups`.
SPA:
- new `static/group-name.js` — `<GroupName name owner [inline]>` renders the
name with the `@owner` handle on a smaller grey line under it.
- used in the sidebar, the group-page header, Explore cards, the Admin
groups table, and cross-group Search (via a widened `cacheGroupIndex`
carrying the owner). Delete/leave confirmations show `name@owner` inline.
- federated Explore rows show `@<source_hub>` instead of an account.
Design record and the locked decisions: ~/next/groupnames.md (out of repo).
MNP unchanged. Tests: test_group_name_unique.py, test_group_name_migration.py.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |
|
|
|
|
|
|
|
|
| |
When the hub has public groups disabled, /explore showed an empty "Public
Groups" screen with a search box and a Create button that do nothing. It now
renders a single line — "This hub does not have public groups." — and the
server already returns an empty directory regardless.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bump the three packages to 0.8.0 (released together) and realign
meshbay-common's __version__, which had drifted to 0.7.0 while the
pyproject stayed at 0.6.0. Dependency pins updated to meshbay-common>=0.8.0.
Protocol versions are independent and unchanged: MNP 0.12, MHP 0.1. The
Electron client stays on its own 0.1.0 track (hub MIN_CLIENT_VERSION).
Since 0.6/0.7: public-groups admin switch with full server-side enforcement
and a group Revoke action; chat link previews (node-side URL unfurl, SSRF-
gated); chat composer focus + scroll-to-bottom on tab entry; whole-group
"Filter files" search.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The filter only ever matched files whose folder was exactly the current
path, so at the top of a group — where every row is a root, never a loose
file — typing in the field did nothing at all.
With text in the field the panel now searches every file in the group by
name or containing folder, wherever it lives, and hides the folder rows
(you are searching, not browsing). Each result shows its folder as a
sub-line; clicking it clears the filter and opens that folder. Empty field
restores the normal folder-by-folder view unchanged.
Also hardened the sort comparators against an entry missing `name`/`type`.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |
|
|
|
|
|
|
|
|
| |
ChatPanel remounts on every tab switch, so a mount effect is the tab-entry
hook. `focus({ preventScroll: true })` so it does not fight the panel-sizing
and scroll-to-bottom effects still settling, and so a phone opens the
keyboard without the view jumping.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Paste an http(s) link in a group's chat and it unfurls into an OpenGraph
card — title, description, site name, and image — the way WhatsApp/Signal/
Slack do it.
The fetch is the node's, never the browser's or the hub's. The browser
cannot: a strict img-src/connect-src and CORS block it, and a direct fetch
would leak every reader's IP to the linked host on each render. The hub must
not touch group content (draft-v6 §2.5). The node already fetches third-party
metadata for the Videos and Music apps, over the same authorised path.
Flow mirrors media_meta_req: the client sends `link_preview_req {url}`, the
node replies `link_preview_resp` with the card fields (or `ok: false`), and
any OG image is stored under its blake3 in the existing media_cache thumb
store — the client then fetches it via the normal file_req path, exactly like
a poster. Nothing durable is added: the card text lives in a bounded in-memory
TTL cache on the node (draft-v6 §2.7 — enrichment on demand, the asking device
caches), and MNP goes 0.11 → 0.12 (additive: an older node logs "unknown type"
and the client shows the bare link).
Because the URL is chosen by a *member* and triggers an outbound request from
the operator's machine, `linkpreview.safe_url` is an SSRF gate: http(s) only,
no credentials, and every resolved address must be globally routable — no
loopback, private, link-local, multicast or reserved range, cloud-metadata
included. Redirects are followed by hand so each hop is re-checked. Residual,
documented in the module: DNS rebinding between the check and connect, closed
properly by pinning the checked IP — a follow-up.
Also fixes a long-standing chat annoyance the preview cards made worse:
opening the Chat tab landed a screen or two above the newest message because
the scroll-to-bottom ran before attachment thumbnails and (now) preview cards
had loaded and grown the content. A ResizeObserver keeps the view pinned to
the bottom through late content growth, and does nothing once the reader
scrolls up.
Tests: test_linkpreview.py (the SSRF gate and the OpenGraph parse, incl.
redirect re-validation and image downscaling) and test_link_preview_request.py
(reply shape, the media_cache image round-trip, the result cache).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A new General tab in Administration carries one switch, allow_public_groups,
stored in a hub_settings key/value table (runtime-editable, unlike hub.toml).
Default is on; an absent row means on, so an upgrade changes nothing.
Enforcement is server-side on every hub-mediated path, not just the SPA:
- create_group refuses visibility=public (403), staff included
- list_public_groups the directory returns nothing (local + federated)
- join_group open-joining a public group is refused
- group_online_nodes a non-member of a public group is handed no node
- signaling.webrtc_offer drops the "node hosts an open group" fallback
- federation.export_directory advertises nothing to peer hubs
The switch is read live, so flipping it back restores every path. Existing
members of a group that predates the switch keep their membership row and
their access — this is plan A, not a purge. GET /v1/hub/info exposes the
flag (unauthenticated) so the create-group form and the sidebar's "Public
groups" link render correctly.
Also in the admin Groups tab: a Revoke action beside Suspend. Suspend is the
reversible hub flag; Revoke calls POST /v1/admin/revoke, which sets
status=revoked and broadcasts a signed revocation every node enforces
(denylist + dropped live sessions). It is confirm-guarded and names the group.
And a message fix the revoke work surfaced: group_online_nodes, join_group and
webrtc_offer answered "Group is suspended" for any non-active status. They now
report the real state, so a member of a revoked group is told "Group is
revoked" rather than something reversible-sounding.
Tests: test_public_groups_toggle.py (10) covers the switch end to end and the
five enforcement paths; test_revocation.py gains the status-message assertion.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018gKJ85aZyvEwarXMFzFEwi
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
merge various-artists compilations into one album
Two real, confirmed bugs in a large flat music library:
- enrich_audio.py's sibling-cover fallback assumed one folder is one
release. A large flat "chart ranking" folder mixing dozens of unrelated
artists carried several distinct WMP AlbumArt-cache guids (one per
original album a track was ripped from), and the fallback picked
whichever one WMP had copied to Folder.jpg — attaching one unrelated
release's cover to every other track in the folder. Now refuses to pick
a cover at all once 2+ distinct guids show up, rather than guess.
- music-app.js's groupMusicEntries grouped by artist first, album second,
so a various-artists compilation (many genuinely different per-track
artists, one shared album tag, no album-artist tag at all — a real
~20-track soundtrack rip has exactly this shape) could never be
recognized as one release: every track landed alone in its own artist's
bucket and got folded into a singleton pile. Now detects an album key
shared across 2+ distinct artist keys and merges those tracks into one
compilation card under a "Various" heading instead.
Both verified against real, previously-affected files and live in the
browser: the shared wrong cover is gone, and the compilation renders as
one card with all its tracks in order.
|
| |
|
|
|
|
|
|
| |
video-app.js's toolbar isn't built from .toolbar-group, so it never got the
existing row-per-group mobile treatment. Mode buttons plus the new
All/Movies/Series control left nothing for the search field on a phone
width. Wraps now, with the filter field taking its own full-width line
like .toolbar-group's search field already does elsewhere.
|
| |
|
|
|
|
|
| |
A segmented control to the left of the search box, defaulting to "All".
Applied before the text filter — a title match within a type nobody asked
to see still isn't shown. Resets to "All" on group change, matching the
text filter's own reset.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Prefetch depth (music-player.js): 5 tracks ahead on Wi-Fi, 3 on cellular —
more runway through a screen-lock network gap when the connection is cheap
and fast, less when it's metered. navigator.connection is Chromium-only;
Firefox/Safari (where it's undefined) get the same conservative tier as an
unrecognized connection type, never assumed fast. MAX_CACHED_BLOBS raised
to 6 to hold the largest case (current + 5).
Keep-screen-on-during-audio (new user preference, off by default): a
Settings toggle, backed by a new whitelisted key on
/v1/users/me/preferences (music_keep_screen_on). music-player.js holds a
Screen Wake Lock only while a track is playing and only when the user has
opted in — unlike the video player's unconditional lock, this must not
fight the ordinary expectation (matching Spotify/Deezer) that the phone
locks on its own while listening.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Found live: every reconnect attempt failed "Signaling failed: 401 Invalid
or expired token", looping for 4+ minutes with no chance of ever
succeeding. connect() takes one token but uses it in two places — the
handshake sent to the node, and the Authorization header on the signaling
POST to the hub — and only the constructor's original `this._accessToken`
was ever used for the latter. onNeedToken correctly fetches a fresh token
for each reconnect attempt, but it only ever reached the handshake; the
signaling call kept sending whatever token the transport was constructed
with, no matter how many minutes had passed or how many attempts fetched
a new one.
connect() now updates this._accessToken on every call, reconnects included,
so both places use the same current token.
|
| |
|
|
|
|
|
|
|
|
| |
Complements the WebRTC auto-reconnect: a track already sitting in
blobCacheRef needs no live connection to play, so whatever was fetched
before a screen lock started plays through it regardless of what the
transport is doing during the lock. One track of runway was often shorter
than the lock itself; two buys more of it. MAX_CACHED_BLOBS (3) already
covers the currently-playing track plus these two, so no cache-size change
needed.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
listener leak
Confirmed live by trace: during a screen lock, every reconnect attempt
failed with "Failed to fetch" (the browser grants no network access to a
locked/backgrounded tab, no code can change that) — expected. But the
backoff timer itself was also throttled while locked: an attempt scheduled
30s out took ~3 minutes of wall clock to fire, because a backgrounded tab's
timers run only when the OS lets them. Recovery after unlocking was
correspondingly delayed rather than prompt.
Fix: an always-on visibilitychange listener (separate from the diagnostic
one, and unlike it not gated on trace mode) resolves the current backoff
wait immediately once the page is visible again, instead of waiting out
whatever of it is left. The actual reconnect this enables is fast (~1.2s in
the trace that showed the "Failed to fetch" run) — the wait was the
throttled part.
Also fixes a real bug the same trace exposed: connect() re-arms the
diagnostic visibility listener and health-ping interval on every attempt
without ever removing the previous instance's — 8 failed attempts during
one lock left 8 duplicate `visibility` trace lines per real event, and (more
than a cosmetic issue) 8 concurrent health-ping intervals once reconnected.
|
| |
|
|
|
|
|
|
|
|
|
| |
Likely root cause of a very confusing test result: the /app HTML response
had no Cache-Control at all, so a browser that decided to cache it
heuristically could keep re-serving the SAME old page (old ASSET_V, old JS)
indefinitely — a normal reload, pull-to-refresh included, has no reason to
override a cache entry it still considers fresh. Every static asset already
gets a fresh URL from ASSET_V precisely so a change is visible, but that
only matters if the HTML naming that URL is itself refetched. no-store
forces every navigation here to hit the network.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Found live: >5min screen lock while music played, then "Transport not
connected" at the 2nd track's end (~8min in) despite the auto-reconnect
from the previous commit. Root cause: fetchTrackBlob's own
`!transport.connected` pre-flight check ran and threw before the
already-in-progress reconnect got the few seconds it needed — it never
reached _sendAndWait, which is the only place the previous fix taught the
transport to wait.
Adds transport.waitForReconnect(), factored out of _sendAndWait's existing
gate, and calls it from fetchTrackBlob before giving up. No-op when nothing
is being reconnected, so the ordinary path is unchanged.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Purely additive and isolated from the streaming/transport code: touches no
DataChannel, SourceBuffer, or playback state, so it cannot itself cause a
stall or regress existing playback (PC/Electron included) — feature-detected
and try/caught, a no-op wherever unsupported or refused.
Sidesteps the commonest real-world trigger for the WebRTC-drop recovery
(the phone auto-locking on its own idle timer while someone just watches).
Does nothing for a deliberate power-button lock or backgrounding the tab —
released automatically in both cases per spec — so the reconnect path is
still what handles those.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
in-flight streams/downloads
Confirmed live (client trace + node logs, mobile screen-lock ~5min): ICE
goes disconnected -> failed within ~10s on both ends, but the DataChannel's
readyState stays "open" throughout, so nothing failed fast — every request
just sat out its own 8s/30s timeout, matching the reported symptom (poster
spinners, blocked chat, dead new streams).
transport.js: on connectionState "failed", reject pending requests
immediately (TransportLostError) and start a self-contained reconnect loop
(capped exponential backoff, redoes the full signaling handshake — the node
already discards the old session on its own "failed"/"closed", so there is
nothing lower-level to resume). New hooks: onNeedToken (fetch a fresh JWT,
since the captured one may have expired during the outage) and
onReconnected (let a consumer resume something that was mid-flight).
file-utils.js: pipelinedDownload retries a lost chunk instead of aborting
the whole transfer — covers Files downloads, poster/thumbnail fetches, and
music-player.js's blob-based track download, all of which go through it.
video-player.js: onReconnected reissues the existing seek-to-current-time
path, which already knows how to land a new stream_init on the live
SourceBuffer without resetting playback.
Playing audio is unaffected either way — musicbay.md's design downloads a
track to a blob before playing it, so a dead transport was never a network
dependency for what is already playing.
Stays on this branch until confirmed by real-device testing.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Client (transport.js): a localStorage ring buffer of connection/ICE/
DataChannel state transitions, visibility changes, request timeouts, and
periodic health pings — enabled once via ?trace=1 (persists), read back at
any time via #mb-debug without devtools. Off by default, zero behavior
change unless enabled.
Node (webrtc_server.py): MESHBAY_WEBRTC_TRACE=1 gates ICE-state-change
logging and a per-session heartbeat (message count, seconds since last
message, ICE/connection state) every 30s.
Debugging aid for the "stuck after several minutes of mobile screen lock"
report — not a fix. Stays on this branch until confirmed useful/resolved.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The :hover/:focus fix in 909505f didn't fully clear the "stuck pressed"
look reported on mobile Chrome: screenshots showed the music player's
"next" button (which carries no state class in the JS at all — not
.active, not .music-player-play) filled with a solid accent-blue
circle after a tap, and the same on "repeat". Neither :hover (already
scoped to `hover: hover`) nor the default focus outline (already
cleared for non-:focus-visible) produces a filled background, so
neither explained it.
Third, separate mechanism: -webkit-tap-highlight-color, Android's own
touch-feedback flash, never disabled anywhere in this stylesheet. Left
at its default, many Android builds tint it from the OS accent colour,
which is why it read as "the app's own blue" rather than an obviously
foreign highlight. Unrelated to real :hover/:focus, so disabling it
carries no keyboard-accessibility tradeoff at all.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013XSohfUQQiaE77qyFLgSv3
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A tap on a touchscreen fires a synthetic hover with no "pointer left"
to end it, and leaves the tapped element genuinely focused (unlike a
desktop click) — so any :hover-styled button looked permanently
"pressed" until something else was tapped, and any plain <button> with
no custom hover at all (the nav hamburger) showed the browser's default
focus ring for the same reason. Not reproducible on desktop, which
actually has a mouse to move away with. Found live on mobile Chrome,
2026-08-26 — most visible on the music player's next/prev/shuffle
buttons, but the same mechanism affects any button in the app.
Two small, general fixes rather than a per-button patch:
- `:focus:not(:focus-visible) { outline: none; }` in the global reset —
clears a lingering focus ring from a pointer/touch interaction while
leaving real keyboard-navigation focus untouched.
- `.music-player-btn`/`.music-player-play`'s :hover rules scoped inside
`@media (hover: hover)`, so a touch tap never triggers them at all.
Only the two reported, confirmed cases are touched here — the same
`@media (hover: hover)` wrap applies to any of the stylesheet's other
:hover rules if the same stickiness shows up elsewhere.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013XSohfUQQiaE77qyFLgSv3
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
IndexEntry.path is the *folder* a file is in (indexer.py's
_virtual_dir docstring: "the directory a file appears in"), not the
file itself. GroupIndex.get_entry_by_path() treated it as if it named
one file, and every one of its four callers did too:
_do_music_meta_request, _do_media_meta_request, _do_tmdb_override, and
_admin_exec_tmdb_override. Any two files sharing a folder — an album is
one folder with many tracks, a season is one folder with many episodes
— collided: a lookup by path silently returned whichever entry the
index happened to iterate to first, regardless of which file the
client actually asked about.
Found live (2026-08-25): three unrelated albums ("High Tone - Various",
two "Le Peuple de l'Herbe" albums) all showed the same MusicBrainz
cover, because all their representative tracks happened to sit in one
"high_tone" folder alongside a track that legitimately matched that
cover. A force-reload didn't help — the bug is server-side, not a
stale client state.
Fixed by keying these four request/response pairs by `file_id` (the
entry's own content hash — already unique, already how every other
lookup in the system identifies a file) instead of `path`, both in the
wire messages (music_meta_req/resp, media_meta_req/resp, tmdb_override)
and in music-app.js/video-app.js's own hooks. GroupIndex.get_entry_by_path
is now unused and removed — GroupIndex.get_entry(file_id) already did
the right thing.
No test previously exercised either handler with two entries sharing a
folder — the only existing coverage (test_tmdb_override_policy.py) gave
each entry its own folder, so the bug never had a chance to show up.
Added that scenario there and in two new test files, all confirmed
failing against the pre-fix code before being confirmed green against
the fix.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013XSohfUQQiaE77qyFLgSv3
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
scanning progress
Two real-world bugs found together while testing multi-root group
creation:
- CreateGroupWizard only sent the enabled-apps PUT when the operator had
*unchecked* something, assuming "every box left checked" already matched
the node's own default (Roster.DEFAULT_APPS = chat, files). It doesn't —
so leaving every app checked, the common case, silently left
Videos/Music/Photos disabled on the node. Now sent unconditionally.
- The wizard's "add extra roots" step never polled index-status, so once
step 3 (which only watches the first/upload root) finished, the
progress bar froze while the node kept scanning the remaining roots for
minutes, unwatched. Added waitForRootsIndexed (platform.js), mirroring
waitForGroupHosted's own race handling.
That fix exposed a deeper one: indexer.py's _scan_root() only flipped
`progress.scanning` on *after* walking the directory and stat()-ing every
file — both off-loop, but slow enough on a large root that a poller's
grace period (waitForRootsIndexed's 5s) could expire before ever
observing `scanning: true` (confirmed against production logs: a GEK-init
step fired 5.058s after a root started scanning, matching the grace
period almost exactly). The stat() pass was also a synchronous loop
directly on the asyncio event loop — blocking the whole daemon (WebRTC,
chat, admin UI) for as long as it took on a root with many files. Both
fixed: `scanning` now flips on before the walk starts, and stat()-ing is
now off-loop too (_size_files).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013XSohfUQQiaE77qyFLgSv3
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A new group application (docs/apps.md's plug-in mechanism), following the
plan in docs/photos.md. Unlike Videos/Music: several photo roots per group
instead of one (photo_roots is a set, one signed op replaces it whole),
a single album-grid view with no third-party matching step, and per-photo
info read from the file's own EXIF at index time — no metadata service,
no credential, no outbound network call at all.
Protocol (meshbay-common, MNP 0.10 -> 0.11, additive): `taken_at`/`camera`
on IndexEntry; `photo_roots`/`photo_roots_ack`; `OP_PHOTO_ROOTS`.
Node: roster.py stores photo_roots as a group_settings entry (JSON list,
same shape as enabled_apps); ops.py/webrtc_server.py validate and sign the
whole set in one op, same pattern as apps_enabled; a new PhotoEnricher
(indexer/enrich_photo.py) runs Pillow in its own small bounded pool,
separate from the video/audio pools, producing a resized thumbnail plus
the two EXIF fields — never GPS, checked by a grep-based regression test.
Client: photos-app.js — one album card per directory containing images,
a per-album photo grid, and a lightbox with next/previous (keyboard and
buttons), zoom in/out/fit/100% starting from the actual on-screen fit
percentage, and a "zip this album" button reusing files-app.js's own zip
mechanism (lifted into file-utils.js's downloadDirectory so both call the
same implementation). group-settings.js gets an add/remove multi-root
picker, distinct from Videos/Music's single-value one.
Bugs found and fixed before this ever shipped, worth keeping the story of:
- enrich_photo.py read width/height from the raw image *before* applying
EXIF orientation correction, and read DateTimeOriginal off the plain
0th-IFD Exif object — a real camera stores it in the Exif sub-IFD, which
Pillow only exposes via get_ifd(Exif). A flat, hand-built EXIF dict
round-trips through Pillow either way, which is exactly what would have
hidden both bugs; the regression test builds EXIF with piexif instead,
matching what real hardware produces.
- photos-app.js's album grouping stripped a trailing path segment from
entry.path under the assumption it still carried a filename — it
doesn't (files-app.js's own convention: e.path is already the
containing directory), so every album collapsed one level into its
parent. Found live against a real multi-folder library.
- transport.js's ADMIN_OP_TYPES allowlist (already the fix for an
identical bug on video_root/apps_enabled, see 4783d81) was missing
photo_roots: its admin_challenge matched no pending request and was
silently dropped, so saving a photo root just timed out after 30s with
no error.
- daemon.py pruned a thumbnail when its file left the index (root removed
or reconfigured) but never forgot the content hash was "already
attempted" — the same bytes reappearing under a renamed/relocated root
(an operator's real workflow) were then permanently skipped, forever,
with nothing to indicate why. Discarding the attempt alongside the
cache entry on prune is what makes pruning actually reversible.
- packages/meshbay-client's app:// protocol handler served every file
with no Cache-Control header, so Chromium was free to serve a stale
cached copy indefinitely — none of several `npm run sync-ui` + reload
cycles during development actually picked up the new code until the
renderer's disk cache was cleared by hand. Now sends Cache-Control:
no-store.
- the lightbox's zoomed image used flex centering (align-items/
justify-content: center) combined with overflow: auto — a well-known
trap where the browser centers overflowing content by shifting it, and
the leading half of that overflow (here, the top of a zoomed photo)
sits outside what the scrollport can actually reach. Reported live as
"unusable". Fixed by switching to top/left alignment once zoomed.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TiZG4AuSnxHohQMpwTHTyL
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Point 1 (directory list not updating without a full page reload) turned
out to still reproduce after the earlier onRefreshIndex fix — that one
addressed the Videos/Music root pickers (nodeDirs), but this section's own
list reads a different field entirely (ops.list_groups returns the
*runtime* root set, groups_ctx[gid]["roots"]) that only gets replaced once
_reload_config_inner's retarget actually finishes. /api/reload itself is
fire-and-forget on the node by design (ops.start_reload's own docstring —
a brand-new group's initial scan can take minutes, the caller must not
block on it), so a single loadNodeInfo() call right after can land in the
gap before that replacement happens and show the pre-change count.
Both the add and remove handlers now poll /api/groups (up to ~4s, every
400ms) until the root count actually matches what the action should have
produced, rather than fetching once and hoping the timing worked out.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A real report showed audio_root timing out with *nothing* logged in
between the connection handshake and the timeout — no admin_challenge, no
error, nothing. The previous fix made an unmatched admin_challenge return
silently (correctly, to stop it stealing an unrelated pending request —
see the earlier commit), but that silence is indistinguishable from "the
request never reached the node at all", which is exactly the ambiguity
blocking this investigation. An unmatched admin_challenge is now logged
(op, op_id, and the full set of currently-pending keys) instead of
dropped quietly, and setAudioRoot/_authorizeAdminOp trace both hops of
the round trip explicitly. Node-side, _do_audio_root gets a debug log at
entry — cheap, and the only way to know from server logs alone whether
the request was ever received if the client-side trail comes up empty.
Diagnostic only: no routing behavior changed from the previous fix,
verified against the same reproduction script.
|
| |
|
|
|
|
|
|
|
|
| |
Flagged directly: "Saving" ran for a few seconds then just stopped, with
nothing telling the operator whether it had worked. Both outcomes used
the same dim .settings-hint styling, so a real failure and a real success
looked identical at a glance. Success and failure are now tracked
separately (previously one plain string held either) and rendered with
the same success-msg/error-msg styling already used elsewhere on this
page, so which one happened is unambiguous.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Reproduced from a real report: enabling the Music app and saving its root
folder in the same Settings visit (the new merged Directories section
makes this a fast, natural back-to-back sequence) fired two signed admin
ops within milliseconds. Neither the admin_challenge reply nor the
admin_response ack two steps later was keyed by anything — both were
matched purely by "whichever request happens to be oldest pending"
(transport.js's own documented last-resort guess). apps_enabled's
challenge stole audio_root's pending slot; audio_root's own request never
received a challenge at all and just sat there until its 30s timeout.
Both hops are now keyed by op name: admin_challenge already carries `op`
from the node, and admin_response is given one client-side purely for
this (the node's _do_admin_response never reads it — only op_id and
signature). A stray admin_challenge with no matching request is dropped
outright rather than guessed at — it is never a broadcast (one
`self._send`, no peer loop, docs/webrtc_server.py), so a session with no
matching key genuinely has nothing to do with it. A domain ack (an actual
broadcast — every connected client gets audio_root_ack, not just the
requester) still falls through to the existing per-type handling when
nobody here is waiting on it, unchanged.
Verified against a standalone reproduction of the exact race (two admin
ops racing, reordered replies) — this codebase has no browser-JS test
runner to add as a real regression test, so the repro lived in a scratch
script rather than the suite.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Folders (unlike files) only ever arrive over MNP as part of a full
index_sync — the ongoing index_delta push has no `dirs` field at all
(daemon.py never puts one there for incremental updates) — so a directory
added or removed via the Electron-local add/remove flow never showed up
in the Videos/Music root pickers until the whole page was reloaded. The
merged Directories section made this easy to hit: add a shared folder,
then immediately try to pick it as a root, in the same visit. Both
actions now call the same onRefreshIndex a chat upload already uses to
pick up its own effect on the index.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Five related pieces of polish against the Music app and Settings, all
from the same conversation:
- Music app now requires audio_root, same as Videos requires video_root:
an empty-state message until one is set, and grouping filtered to only
what's under it (underAudioRoot, mirroring video-app.js's
underVideoRoot). Wires the new audio_root/audio_root_ack pair through
transport.js and group-page.js state the same way video_root already
flows.
- Starting one player now stops the other — opening a film closes the
music queue, starting a track closes the video modal. Both used to run
at once, found live.
- Group Settings reworked: every section but a bare form (invite,
pair-operator, approve-device) is now collapsible (CollapsibleSection);
the uploads on/off button is a real toggle switch (ToggleSwitch,
reused for TMDB/MusicBrainz's enabled switches too, each now with an
icon + status badge in its header instead of a plain checkbox row);
and shared directories, the Videos root picker, and the new Music root
picker are merged into one "Directories" section (RootFolderRow) instead
of three separate ones scattered down the page — the root pickers only
show once their app is actually enabled.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Two complaints against real use: the artist -> album -> track hierarchy
was invisible (every depth sat flush left, distinguishable only by which
chevron happened to be open — Videos' own flat list never needed more
than one level, so there was nothing to reuse for this), and a filled-in
album unfolded into a wall of identical little icon-box squares, one per
track, carrying no information a track row can actually use (unlike
Videos' per-episode thumbnail).
Track rows now reuse Mode A's own numbered tracklist style
(.music-track-row: number, title, duration, no icon box) instead of
Videos' boxy thumb-slot row. A folder's expanded contents get wrapped in a
new .music-flat-children indent + rule line, so nesting reads as visible
steps into the tree rather than same-level siblings. Folder rows
(artist/album headers) still reuse Videos' flat-row style, which fits them
fine — this is not a wholesale rewrite, only what didn't actually work.
|