Commit Graph

15 Commits

Author SHA1 Message Date
d46ec73261 Per-user read state and subscriptions
Read, starred and position move to entry_state; subscriptions carry each
person's keywords, auto-download, explicit and per-scan limit. The feed
list and unread counts are per person, and the existing library is
adopted by the admin on first start.

The feed URL, folder and schedule stay shared and admin-only: one file
serves everyone, so they describe the file rather than a preference.
Scanning merges subscribers' wants -- anyone wanting an item is enough --
via merge_policy, which is pure and tested.

Also: the test fixture wiped its data directory from every Playwright
worker, deleting the database out from under the running daemon.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-11 02:17:16 +00:00
06f182b555 Accounts and sign-in, and fix the read toggle
epAction's redraw closure called itself when handed a row, so Mark read
recursed until the stack blew; it now swaps that row in place. Opening an
item also marks it read, redrawn where it stands so nothing vanishes from
under the pointer on the Unread tab.

Step A of multi-user: users and sessions tables, Argon2id, a session
cookie, ipx user subcommands, and a trusted proxy header for Cloudflare
Zero Trust -- honoured only from a trusted_proxies address. The shared
token still works and is the admin. A new database starts with
admin/ipodderx.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 23:11:20 +00:00
772dda5154 Call them items, not episodes
Half the library is text feeds, so the UI no longer assumes a podcast:
counts, search, the empty detail pane, the phone back button, retention
and per-feed settings, and the download dialog all say item. S1E1 badges
and the episode column stay -- those are the itunes:episode field, which
only appears when a feed publishes it.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 19:12:58 +00:00
d157c88ba9 Make the UI work on a phone
The ☰ button lived in the player bar, which is hidden until something
plays, so the feed list was unreachable on a phone. It moves to a bar
that is always present, and the sidebar gets a scrim.

The reading pane takes the whole screen over the list with a back
button, the player stacks into two rows above it, and the page no longer
scrolls sideways -- a grid column is min-content wide by default, so one
long headline dragged everything off the right edge.

Covered by a Playwright case at 390x844.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 18:58:10 +00:00
7473d5b7fb Mark all read on an OPML subscription
Marking the subscription read did nothing: its own row holds no entries.
read-all now resolves the feeds grouped under the id -- via
subscriptions(), so a child promoted to config is included -- and marks
those. Button sits before Unsubscribe, where every other feed keeps it.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 18:45:46 +00:00
dfcdf47143 Line up the feed sidebar
Every row reserves the chevron slot, so artwork and titles share one
column instead of stepping left when a feed has no children. Children
keep a single icon size and read as nested from the indent alone. Labels
stack on one line-height, and the unread count has a min-width so a
three-digit feed doesn't shove its own title.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 18:39:25 +00:00
dd9c0881ee Keep every enclosure on an item, and view without downloading
The rss crate keeps at most one enclosure per item and, when a feed ships
several, silently keeps the last -- so a two-file item lost its first
file. enclosures_by_item reads them from the XML in document order,
unescaping attributes so a URL's &amp; survives. The row summarises the
one you would act on and counts the rest; the pane below lists them all.

Non-media enclosures gain a View link opening in a new tab: the
publisher's URL, or the local copy once downloaded. A direct link, not a
proxy, so the daemon does not become a fetch-anything relay.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 18:01:25 +00:00
6d6b4dd1e4 Do not offer a player for a file that is not audio or video
The media-type filter stopped new image enclosures being fetched, but ones
already on disk still got a play button and an <audio> element, because
the UI tested for a path rather than for a playable type. Four places did
this, including play() itself, which picked the first downloaded enclosure
whatever it was.

isPlayable() checks audio/* or video/*, falling back to the extension when
a feed declares no type. A downloaded non-media file now shows as its kind
with Save and Delete, so it stays available without posing as an episode.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 17:50:39 +00:00
ddb9dbb7e1 Three-pane layout: feeds, items, item text
Feeds beside, the feed's items above, and the selected item's text with
its enclosures below -- the shape iPodderX used. Selecting a row fills the
pane below instead of expanding inline; enclosures render there as a
player when the file is present and a labelled download when it is not.
The divider drags and its position is remembered.

The archived site kept no usable screenshot of the original window, only
marketing panels, so this follows the description rather than reference
art.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 17:45:18 +00:00
8b21d19365 Log tabs with a daemon I/O view, and Playwright UI tests
The log view splits into All / Daemon I/O / Scans / HTTP. Daemon I/O is
the control protocol itself, logged where every command funnels through
so it covers socket clients, the CLI and the web UI alike. stderr and the
in-app buffer now have separate filters, so the UI can keep debug detail
the terminal should not carry.

Playwright drives a real browser against a daemon on fixture feeds. Eight
tests, each mapping to a bug that reached a user -- the Rust tests and the
stub-DOM smoke test cannot see a wrong selector or a dead handler.

It immediately found one: OPML folders rendered expanded by default,
because the code stored closed groups, so any folder never toggled counted
as open.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 17:07:19 +00:00
665d5b8ecb Keep OPML feeds out of config, cap downloads, log daemon work
Writing 82 derived feeds into a hand-edited config.toml made it
unreadable. The OPML is the source of truth, so its feeds are re-derived
each scan and held in the database, inheriting the subscription's
settings; editing one promotes it to a real entry. A migration moves
existing children out -- 611 lines to 38 -- keeping all entries and files.

max_new_per_check defaulted to unlimited, so subscribing to an OPML of 82
feeds pulled whole back catalogues. It now defaults to 3 via [general],
capping every feed that does not set its own, and the pending queue orders
by publish date so a cap of 3 means the three newest.

Scans and downloads travelled as socket events only, so the log view
showed no daemon activity. They are mirrored into tracing, with routine
skips at debug -- at 82 feeds those alone would flush the buffer.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 15:15:54 +00:00
5f6e2a8dc1 Add a log view, and Docker packaging
The Log button shows the running daemon live: feed scans, downloads,
torrents and every HTTP request. It reads a ring buffer filled by a
tracing layer rather than tailing a file, so it works under Docker where
logs go to stdout. The access-log middleware skips /api/logs, or the
panel's poll would log itself forever.

Detached torrents could leave a row stuck in 'downloading' across a
restart, where nothing would ever revisit it; those are requeued at
startup.

Dockerfile, entrypoint and compose: 114 MB runtime, config bound to
0.0.0.0 on first run since container loopback is unreachable, drops to
PUID:PGID for Unraid, and a healthcheck that goes through the control
socket so a wedged worker reads as unhealthy.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 12:06:54 +00:00
e97f2b9c2f Schedule pickers, and target progress at one row
"Check feeds every" becomes a number plus a unit dropdown in both global
and per-feed settings; parse_interval gained weeks to back it. The
per-feed dropdown can select the global default, clearing the override.

Fixes progress painting every pending row: the event carried no enclosure
id, so the handler had nothing to target and set the width on all of them.
Adding a feed looked like it was downloading everything. Progress,
DownloadDone and DownloadError now carry the enclosure id.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 02:38:14 +00:00
9960befed5 Fix the UI dying at load, and add a page smoke test
$('#prefs').onclick referenced a prefsModal that was never defined, and an
uncaught ReferenceError stops the whole script -- taking the theme toggle,
the feed filter, the event stream and loadFeeds() down with it, so the app
rendered an empty shell.

The scheduling patch had anchored on a function the rewrite already
deleted; str.replace matched nothing and said nothing.

tests/page-smoke.js executes the page against a stub DOM so this class of
failure is visible, since every server-side check passed while the UI was
completely dead. Handler wiring now skips a bad reference instead of
throwing.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AdXho5tTkjFLeUXKbEjKBh
2026-09-10 02:31:18 +00:00
ab2583fc64 Step 3: feed fetch and parse
Conditional GET plus an RSS-first, Atom-fallback parser normalising both
into one entry model, with iTunes explicit and ttl handling carried over
from the Python. Enclosures are recorded as pending; nothing downloads yet.

GUID falls back guid -> link -> enclosure url -> title rather than hashing
the description as the original did.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RPyeapneuXrCdojsaiXGbe
2026-09-09 20:34:09 +00:00