- Write channel_name (and stream_name fallback) into Redis metadata at
channel init time; read directly from Redis in get_basic_channel_info,
removing Stream and M3UAccountProfile DB queries on every stats tick
- Pass channel.name from views.py into ChannelService.initialize_channel
via new optional channel_name param, skipping the redundant Channel DB
lookup on the normal streaming path
- Pass stream_name from get_stream_info_for_switch through change_stream_url
into _update_channel_metadata, skipping the Stream DB lookup on switches
- Resolve m3u_profile_name on the frontend from the playlists store instead
of querying the DB per tick; StreamConnectionCard builds an id→profile map
- Fix channel start notification showing no name and stop notification
showing UUID: both now use channel_name from the stats WebSocket payload
- Adding streams to a channel (per-row "Add to Channel" button and bulk "Add selected to Channel") now completes in a single PATCH request instead of three. Previously each operation called `API.updateChannel` (which internally triggered `requeryStreams`), then `requeryChannels()`. Both paths now use a dedicated `API.addStreamsToChannel()` method that issues only the PATCH, merges the existing channel streams with the newly added stream objects locally, and updates the store in-place, no `requeryStreams` or `requeryChannels` round-trips.
- Removed an unnecessary `requeryStreams()` call from `API.createChannelFromStream()`. Stream data does not change when a channel is created from it; the caller already calls `requeryChannels()` to display the new channel.
- Applied the same lightweight store-update approach from stream reorder to stream removal. `removeStream` previously called `API.updateChannel` (which internally triggered `requeryStreams`), then also explicitly called `requeryChannels()` and `requeryStreams()`. Removal now calls `API.reorderChannelStreams` with the remaining stream list (optimistic local `setData` first), matching the one-request pattern used by drag reorder.
- `removeStream` in `ChannelTableStreams` was capturing `data` and `channel` in its closure, causing the `columns` `useMemo` to recreate the entire column array (and new TanStack table instance) on every reorder or remove. Both values are now read through refs (`channelRef`, `dataRef`) so `removeStream` has no dependencies and is stable for the lifetime of the component. Removed `removeStream` and `playlists` from the `columns` dep array.
- `DraggableRow` (stream rows inside the expanded channel) is now wrapped in `React.memo` with a comparator on `row.original` identity and `index`, so rows whose stream data and position haven't changed are skipped entirely during re-renders caused by a sibling row moving.
- Removed dead code in the `name` column cell: `playlists[stream.m3u_account]?.name` was indexing an array by an integer ID, which always returns `undefined`, the value was immediately overridden by `m3uAccountsMap`. The dead access and its fallback variable are gone; `m3uAccountsMap` is now the sole lookup.
- Removed spurious `state: { data }` from the `useReactTable` call. `data` is a root-level TanStack Table option, not a controlled-state entry; passing it in `state` was a no-op but misleading.
- **Channel table performance**:
- Removed unused Zustand store subscriptions (`channels`, `selectedProfileChannels`, `selectedProfileChannelIds`) and an unused `channelIds` array subscription from `ChannelsTable` to reduce unnecessary re-renders on unrelated store updates. Cleaned up associated dead code and unused imports.
- Optimized `ChannelTableStreams` (the expanded stream list inside each channel row) to reduce mount cost: moved pure helper functions and static values (`getCoreRowModel`, `defaultColumn`, stat categorization/formatting) outside the component so they're created once; stabilized the TanStack column definitions by removing `data` and `expandedAdvancedStats` from the `useMemo` dependency array (cell renderers receive the row at render time); switched advanced-stats toggle tracking from `useState` to a `useRef` + per-cell local state so toggling one stream's stats doesn't recreate the entire column array and table instance; memoized `dataIds`, `removeStream`, `handleDragEnd`, and `handleWatchStream` with `useMemo`/`useCallback`; extracted `StreamInfoCell` as a `React.memo` component with its own memoized stat categorization.
- Fixed `getChannelStreams` store selector to return a stable empty-array reference instead of creating a new `[]` on every call for channels without streams, preventing unnecessary re-renders via the `shallow` comparator.
- Memoized individual rows in `CustomTableBody` so that expanding/collapsing a channel only re-renders the 1-2 affected rows instead of all rows on the page. Callback functions (`renderBodyCell`, `expandedRowRenderer`) are stored in refs so memoized rows always use the latest version without the function references themselves defeating the memo comparator.
Enhancement:
- Decoupled row expansion from row selection in `CustomTable`, expanding a channel row no longer also selects it. Added an `onRowExpansionChange` callback to `useTable` so callers can react to expansion changes independently of selection state.
- Fixed a stale-closure bug in memoized channel row checkboxes, `selectedTableIdsRef` now ensures the checkbox `onChange` handler and `handleShiftSelect` always read the current selection set rather than the stale set captured at render time. Without this, clicking any checkbox after the first would silently deselect previously checked rows.
- Fixed the "Add to Channel" per-row and bulk buttons in the Streams table not activating when a channel row is expanded. `StreamRowActions` now subscribes to the Zustand store directly (bypassing `React.memo`) so button state updates when `expandedChannelId` or `selectedChannelIds` change without any parent row props changing. Added `targetChannelId` (expanded channel takes priority; falls back to a single selected channel) used by both the per-row and bulk add paths.
- Removed dead props `getExpandedRowHeight` and `tableBodyProps` from `CustomTableBody` and their corresponding pass-throughs in `CustomTable`, both were accepted but never consumed.
