The file_group.textsearch tsvector indexes all four text columns;
d_text (captions, document text, zip contents) is ~90% of the data,
making ts_rank searches take seconds on large libraries. Add a
textsearch_abc generated column (title/author/description only, ~8x
smaller) and default all searches to it, with deep=true opting into the
full column.
Database:
- Migration adds the generated column and a GIN index (rewrites the
table; a few minutes on large libraries).
Backend:
- FilesSearchRequest, VideoSearchRequest, ArchiveSearchRequest, and
DocSearchRequest accept `deep`; each search function picks the column.
- /api/search_file_estimates returns both counts in one index-bound
query (COUNT FILTER on the abc subset): `file_groups` (fast) and
`file_groups_deep`.
UI (one coherent story across the count surfaces):
- The typing dropdown always suggests the deep count (what is findable).
- The tab badge shows the active mode's count so it matches the result
list.
- The gap is reconciled by a hint offering deep search: count-aware on
the global search ("No results in fast search. N results available in
deep search." / "N more results available..."), generic on the module
pages (no deep estimate there). The hint resets pagination.
- "Search Depth" toggle in the search filter modals, persisted as
?deep=true.
Tests that relied on d_text matches now pass deep=True, with fast-mode
assertions alongside. Also skip flaky test_download_channel on
CircleCI.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Browser-based ESP32/ESP8266 flasher (esptool-js + Web Serial); nothing is
uploaded to WROLPi, the flash is written straight to the device over USB.
Firmware sources (themed, tabbed): a saved set, the WROLPi media directory,
or a local file. Local files can be dropped anywhere on the page, which jumps
to the "Add from computer" dropzone. The media picker searches .bin files by
path and can detect the connected chip to filter firmware to matching images.
- Indexed file_group.suffix + path filter for fast .bin search (alembic
migration backfills via split_path_stem_and_suffix)
- modules/flasher: chip detection from ESP image headers, /search endpoint,
and flasher.yaml saved configurations (/configs CRUD)
- Flash progress modal with ETA, connect timeout, boot-mode hint, auto-
scrolling logs
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
getSettings() in api.js remaps the /settings response through a fixed field
whitelist, which dropped download_daily_limit_global/per_domain before they
reached the SettingsContext. The values saved correctly but always rendered
empty. Add both fields to the mapping.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Restore the censored-video search filter
Censored videos (no longer publicly available — detected when a channel
refresh drops them from the channel list, or when comment downloads fail
with "video unavailable/private/removed") are already flagged on
FileGroup.censored, but there was no way to find them since the search
rework dropped the old filter.
Add a "Censored" toggle to the reworked search filter modal, shown only on
the Videos page (showCensored). When enabled the video search returns only
videos with FileGroup.censored = TRUE, tracked via a ?censored=true query
param (useSearchCensored).
* Move Availability filter below the Tags/Results row
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Inventories are a config-only MultiFileConfig (config/inventory/<slug>.yaml) and
were not reachable by the admin config Restore UI, so there was no way to re-import
a hand-edited/copied-in file or restore an inventory from its daily backup.
Backend:
- MultiFileConfig gains generic helpers: get_backup_dates(slug)/_get_backup_file
(surface the daily config/backup/<subdir>/<slug>-DATE.yaml files) and reimport()
(rebuild the in-memory store from disk; picks up hand-edits and copied-in files,
drops files deleted on disk). backup_date is constrained to YYYYMMDD so it cannot
traverse out of the backup directory.
- InventoriesConfig.preview_restore/apply_restore: overwrite replaces the whole
inventory; merge unions items by id (keeps current fields and existing items).
The pre-restore state is backed up automatically, so a restore is undoable.
- API: POST /inventory/reimport, GET /inventory/<slug>/backups,
POST /inventory/<slug>/restore/preview, POST /inventory/<slug>/restore.
Frontend:
- New InventoryImportModal (opened from a Restore button in the Inventory toolbar):
Re-import from disk, plus a dated-backup table with per-row Merge/Overwrite that
previews the change before applying.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the DB-backed inventory module with a config-only design and build out
food-storage tooling around it.
Inventory (config-only, no database)
- Each inventory is a YAML file at config/inventory/<slug>.yaml, loaded once
into memory and the single source of truth. New reusable MultiFileConfig
(wrolpi/common.py) manages a directory of per-entity files.
- Inventory types (food/fuel/tool) seed a customizable field schema
(text/number/quantity/date/select/location/calories). Unit math/aggregation
moved to the frontend (mathjs); Pint dropped from the backend.
- Minimal API: GET / (all inventories full), POST /, PUT /<slug> (whole-
inventory save with version-based optimistic concurrency; 404 for missing),
DELETE. Legacy inventories.yaml / old tables migrate to per-inventory files
(pound->lb), and an alembic migration drops the tables.
Frontend
- New app/src/components/inventory/ package: spreadsheet-style table with
Tab/Enter keyboard entry, sortable columns, expired-row highlighting,
customizable field editor, and pooled cross-inventory location autocomplete.
- A single useInventories store loads everything once and persists whole
inventories optimistically; list/items/summary/locations derive client-side.
- Summary groups items (default by Category), totals size x count, shows a
calories column, and embeds a ration estimate (how long the stock lasts).
Food catalog
- Shared, editable catalog (config/inventory/catalog.yaml) seeded from
catalog_defaults.yaml (~60 generic items, total calories per package) with
merge-keep-edits + tombstones. GET/PUT /api/inventory/catalog, a management
modal, and entry-row pre-fill of category/size/calories from a name match.
Calculators
- Generic Food Storage planner (like the Water calculator): household +
editable per-person/year amounts -> total to store over a chosen duration,
with women/children weighted by DGA calorie shares and an imperial/metric
toggle.
Also fixes ConfigFile.save() to accept nested file_names. Backend pytest and
frontend jest suites updated/added throughout; alembic check clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Move One-Time Pad encryption to the browser; delete backend OTP module
OTP encryption/decryption is now performed entirely in the browser so
messages, keys, and generated pads never leave the originating device.
- Add app/src/components/otp.js: a JS port of the former modules/otp/lib.py
(encrypt/decrypt, message formatting, validation, and secure pad
generation via crypto.getRandomValues with rejection sampling).
- Encrypt/Decrypt now run live as the user types or pastes, with inline
error messages instead of submit buttons.
- Generate New Pad builds the printable HTML locally (blob URL) instead of
hitting /api/otp/html.
- Migrate all backend OTP tests to app/src/components/otp.test.js with
identical data.
- Delete the modules/otp backend module and its blueprint registration.
* Revoke the One-Time Pad blob URL after the new tab loads
Addresses a minor blob-handle leak: handleGenerateNewPad created an object
URL on every click but never released it.
* Let users choose the One-Time Pad character set
Add an "Advanced" accordion under the buttons with three options: A-Z,
A-Z/0-9 (default, unchanged), and a custom character set. The chosen set is
shared by encrypt, decrypt, and pad generation.
- otp.js: parameterize encrypt/decrypt/generate on the alphabet (default
OTP_CHARS for back-compat); add OTP_CHARS_ALPHA and validateCharset; input
is auto-uppercased only when the set is all-uppercase, so custom sets with
lowercase are case-sensitive; the printable pad is labeled with its set.
- Apps.js: lift charset state into OTP(); gate encrypt/decrypt/generate on a
valid set with inline validation errors.
- otp.test.js: add A-Z, custom round-trip, auto-uppercase, validateCharset,
and pad-label tests.
* Scope inverted-accordion dark background to direct children
The rule forcing #171616 on .ui.inverted.accordion .title/.content used a
descendant selector, so it leaked onto any nested element using the `content`
class — e.g. an ErrorMessage's Message.Content rendered inside SimpleAccordion
got a dark band behind its text. Scope to direct children so the accordion
still styles its own title/content but not nested components.
* Add optional error-detection checksum to the One-Time Pad
Adapts the position-weighted checksum from otp-checksum.md to the
frontend-only OTP. checksum = sum(position * value) mod N, parameterized on
the selected character set (N = chars.length), so it follows whichever pad
alphabet is chosen. Position weighting catches transposition errors.
- otp.js: add calculateChecksum/appendChecksum/verifyChecksum.
- Apps.js: opt-in "Append error-detection checksum" toggle in Advanced
(default off). When on, encrypt appends a checksum char; decrypt verifies
it, warns on mismatch, and still shows the decrypted text.
- otp.test.js: checksum unit tests, charset-variation, and end-to-end vectors.
* Show compact checksum status instead of a large warning box
Replace the bulky semantic-ui WarningMessage on decrypt with a small inline
status line: a green check + "Checksum is valid" or a red x + "Checksum is
invalid", shown whenever the checksum option is on and a message is being
decrypted.
* Fix checksum-mode decrypt truncating messages without a checksum
When the "Append error-detection checksum" option was on, Decrypt always
stripped the last ciphertext character before decrypting, so a message that
carried no trailing checksum (wrong sender setting, body pasted alone, etc.)
lost its final character and decrypted to garbage.
Add stripChecksum(received, chars), which removes the trailing character only
when it actually verifies as the checksum of the preceding characters; a
message without a valid checksum is returned intact and flagged invalid. The
decrypt UI now uses it.
Also broaden test coverage:
- stripChecksum tests, including a regression for the truncation bug.
- validation error classes (InvalidOTP/Plaintext/Ciphertext) and the
longer-than-OTP length guards.
- deterministic encrypt/decrypt roundtrip fuzz across several character sets.
- generateMessage character-count and charset-membership checks.
- end-to-end checksummed encrypt -> format -> strip -> decrypt flow.
Use a dedicated InvalidMessage error for the checksum helpers instead of the
misleading InvalidPlaintext.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
* Add Playlists: ordered, mixed-type collections
A playlist (Collection kind='playlist') is a manually-curated, ordered list that
mixes item types. Squash of the playlists feature branch (24 commits).
Data model & API:
- CollectionItem is polymorphic via item_kind: 'file' (FileGroup), 'zim'
(zim_id + zim_entry), or 'url' (link to anything the WROLPi browser opens,
e.g. /map?lat=&lon=&z=). CHECK constraint enforces exactly one kind; per-kind
unique constraints; migration a7b1c2d3e4f5.
- Endpoints: create playlist; add/remove/reorder items (idempotent adds: 201 new,
200 existing); GET returns the ordered items. All mutations wrol_mode_check'd.
Config (source of truth):
- playlists.yaml (PlaylistsConfig) stores each playlist's metadata AND ordered
items (file by relative path, zim by path+entry, url inline); round-trips a
DB wipe. Supports backup import (DB_CONFIG_FILE_NAMES).
- The redundant generic CollectionsConfig (collections.yaml) is deleted:
channels/domains own their configs, authors/subjects are re-derived.
On-disk sync (sync_playlists_directory switch):
- Each playlist materializes into the configurable Playlists Directory
(playlists_destination, Settings UI input, excluded from indexing) with
position-prefixed entries: NNNN_<file> hard links (copy fallback across
filesystems) and NNNN_<title>.html stubs for zim/url items.
- Tagged playlists are namespaced like channels: <playlists>/<tag>/<name>;
a tag change moves the directory and prunes the empty old tag dir.
- Deletion safety: managed files are only removed when they are our marked
stubs or still have another hard link (st_nlink > 1) -- never a last copy.
Stray loose files at the playlists root are removed; stub URLs are
scheme-allowlisted (no javascript:/data:).
Frontend:
- /playlists list page (CollectionTable, name filter, New Playlist modal, Tag
column) between Files and Zim in the nav.
- View-only detail page (Name link) and edit page (Edit button): themed table
with model icons/colors + poster thumbnails, items link to their model page
(video/archive/doc) or preview, up/down reorder, remove, add-URL form,
single-tag selector; all editing disabled in WROL Mode.
- Reusable green "Add to Playlist" button on Video/Archive/Doc detail pages,
the FilePreview toolbar, Zim search results, and Map pins (added as a /map
URL item with a map-marker icon).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Playlist edit page parity: rename, custom directory, tag modal move suggestion
Match the ChannelEditPage/DomainEditPage UX on the playlist edit page:
Back/View buttons, an edit form (name, description, directory), and the
shared CollectionTagModal with the violet "Tag" button. The tag modal now
suggests moving the playlist to its managed tag location
(playlists/<tag>/<name>); accepting the suggestion stores directory=None so
the sync keeps auto-managing the location. Tag buttons on the Channel and
Domain edit pages are now violet to match.
Backend: update_collection learns rename (with per-kind validation and
duplicate checks) and playlist custom directories (cleanup of the abandoned
directory, config round-trip, commit-before-switch to avoid the background
sync racing pre-commit state). tag_collection's duplicated untag/tag
branches are merged, and the per-kind config-save dispatch is consolidated
into _activate_config_save().
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Deduplicate Collection error handling and Tag creation
Delete the duplicate UnknownCollection in wrolpi/collections/errors.py and
re-export the canonical one from wrolpi/errors.py, which carries
status_code=NOT_FOUND (the local copy inherited 500). Replace the repeated
query/raise pattern in reorganize.py with Collection.find_by_id.
Add Tag.get_or_create_tag and use it for collection tagging; flush the new
Tag before returning so callers can read .id immediately -- previously
tagging a collection with a brand-new tag name set tag_id=None. Regression
test included.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Map viewer has simple search.
Map search databases can be downloaded from CDN, or built locally. Map searches can be built offline, but will not have enriched things like state names downloaded from Wikidata.
Create a new Docs module that consolidates PDF and eBook handling into a
unified system with subjects and authors support. This replaces the
separate PDF/eBook modelers with a single Doc model backed by an
alembic migration (rename ebook to doc).
Key changes:
- New modules/docs package with models, API, extractors, and lib
- Subject and author metadata extraction with case-insensitive
normalization for cleaner PDF metadata
- Frontend Docs pages with AuthorsPage, SubjectsPage using
CollectionTable, and doc card deletion
- Global search integration for subjects and authors
- Generic collection item count support
- Frontend file card metadata adjusted to match collection sort
- Bump yt-dlp version
* Added encrypted cookies feature for use with yt-dlp. Decrypted cookies never touch the disk.
* Adding sleep between all downloads to prevent rapid requests.
* Channels are fully refreshed before downloading.
* Removing superseded video browser profile.
* Batch reorganization is supported for both Domain and Channel Collections.
* Batch reorganization happens when any Collection does not match the respective file format. For example, a Channel Collection file format that does not match the `yt_dlp_options.file_name_format` will show `needs_reorganization`.
* Individual Collections can always be reorganized. The reorganization is a little more aggressive. Useful for when some files may have been missed in the batch reorganization.
Introduces FileWorker for background file operations with dual-queue
architecture (public/private). Handles refresh, move, and rename tasks
with progress tracking, job completion monitoring, and batch processing.
Compares DB FileGroups to filesystem state using FileGroupDiff to detect
new/modified/deleted files. Designed for memory efficiency on Raspberry Pi.
Replacing Channels and Domains with Collection.
Moving downloading relationships from Channel to Collection.
Created domains.yaml config to manage tagged domain collections.
Adding turn-key upgrade solution in the UI.
