Completes the 13 keys the obtainability feature left as [TODO: Translate], in all
9 locales: the two gate-event counts, the sale / Blue Buzz / "Free Now" / early
access end-date badge strings, and the six settings.priceTracking strings plus the
section header.
- Buzz and Blue Buzz stay as-is everywhere: they are CivitAI currency names, not
translatable words (R3)
- register follows each file's existing norm (你 zh-CN, 您 zh-TW, Sie de, tú es,
вы ru); fr keeps the file's ASCII apostrophe style; no full-width punctuation
leaked into the Latin/Cyrillic/Hebrew locales
- two source fixes came with the pass: the unused settings.priceTracking.label key
is removed (no template renders it - the toggle uses enabled/enabledHelp), and
settings.sections.priceTracking now reads "Buzz Download Prices" so the header
matches what the feature does: prices are displayed, nothing is tracked for
alerts
- the locale files were edited by exact-value replacement rather than re-serialised,
so each diff is 13 lines and the per-file indentation is untouched (R1)
Live model 958009 has 37 gated versions across 6 price points, and the owner's
own library has a model whose two early access versions end on different dates,
so an 'Update - 500 Buzz' badge would be fabricated rather than summarised. The
uniformity that made it look plausible is an artefact of floor pricing. Prices
stay in the version list; only a deadline is well defined at model level.
The owner could not tell from the UI what "Buzz Price Tracking" enabled, what the
"Price alert threshold" number meant, or what "Price Alerts" was alerting about.
That was not a copy problem: the implementation exposed our mechanism (a page
scrape) and our SQL predicates as the user's concepts. Two concrete defects came
from the same root:
- the alert population included versions the user already owns (neither the event
generator nor the query filtered on is_in_library; in the owner's library 28 of
52 gated versions were already downloaded, so most "alerts" were about files
already on disk, which cannot become cheaper *for them*);
- a threshold-filtered state list lived in a notification surface, so an empty
panel had three indistinguishable causes and read as a broken feature.
The information model is now the version plus ownership: cost is shown only where
a decision exists. Owned -> nothing. Not owned and free -> nothing. Not owned and
gated -> the price when it is known, `Paid` without a number when it is not, and
early access keeps its countdown because "free on <date>" decides between waiting
and paying. The numeric threshold has no place in that model: every decision is
categorical (wait / pay / skip), so the setting, the comparison and the whole
alert-state machine are gone.
- both alert-state columns are removed from the schema rather than left dead; a
database created by an unreleased build has them dropped on open (native
ALTER TABLE ... DROP COLUMN, guarded), which is a no-op for everyone else
- gate events are emitted only for versions the user does not have, and the
price-drop event goes with the threshold it belonged to
- both alert endpoints, PriceAlertsHandler and the service-registry adapter field
it needed are removed: events already reach the UI through the refresh response
- the bell tab, panel, CSS, both entry points, the unread watermark and their
locales are removed; the setting keeps only the enable flag and the refresh
interval and is framed as plumbing
- "Price unavailable" is replaced by `Paid`: the gate is certain from the public
API, only the number is best-effort, and that is our plumbing, not the user's
problem
Verified against a copy of the owner's real database: 52 gated versions ->
28 owned (now silent) + 24 the feature is actually about; the drop migration ran
and both removed endpoints 404.
The default threshold is 0 ("only tell me when a version becomes free") and the
settings copy says so, but the panel did not: a real instance with 52 priced paid
versions and an untouched threshold showed "Nothing is under your price threshold
right now" with only a small "Alert threshold: 0 Buzz" in the corner, which reads
as a broken feature.
- the payload now carries pricedCount, so the empty state can say how many paid
versions already have a known price
- when the threshold is 0 and prices are known, the empty state says so and
points at Settings - Library instead of implying there is nothing to show
- the read-time threshold comparison means setting one takes effect immediately;
measured on a copy of that instance: 0 Buzz -> 0 alerts, 100 -> 44,
500 -> 48, 5000 -> 51
Found in a real instance: after enabling price tracking, a normal "Check updates"
captured exactly one price out of 718 models, so the alerts panel looked broken
while the log said the refresh completed.
Price capture only ran when the version list was re-fetched, so it inherited the
metadata TTL: with 24 h metadata and 24 h price TTLs, only the handful of models
whose metadata happened to be stale that round were ever priced.
- the cached record already carries the gate, so the price pass now runs off
whichever version list is available (freshly fetched or stored) and applies the
result without touching last_checked_at, so a price-only pass cannot silently
extend the metadata TTL
- a failed attempt now satisfies the price TTL, so a mature model whose page no
host will serve is not retried on every single update check
- an explicitly forced check re-prices within the TTL
Verified by copying a real instance's update DB into a sandbox and running a
non-forced check: bulk metadata fetches 0 (version lists entirely from cache)
while priced versions went 1 -> 20 and the panel listed 19 alerts.
End-to-end verification against the live site found the price capture broken for
a whole class of users: the civitai page hosts are not interchangeable, and the
user's civitai_host preference was silently fatal. With civitai_host=civitai.red
the update DB held zero prices even with tracking enabled.
- civitai.red refuses non-browser HTTP clients outright (Cloudflare challenge,
403 for any User-Agent, aiohttp and httpx alike), while civitai.com and
civitai.green answer normally for anonymously visible models and 404 for
mature ones. An earlier manual check with curl passed on TLS fingerprint luck,
which is why this was missed.
- get_model_prices now tries the configured host first, then the others, and
takes the first parseable payload. The host that worked is remembered, and a
host that refuses outright is parked for 15 minutes so a library full of
mature models does not pay three requests each; a 404 is model-specific and
does not park the host. Links keep using the configured host, which is where
the user's own browser has clearance.
- Mature models still have no price source anywhere, so that is now stated
instead of silent: price_check_attempted_at separates "tried and unreadable"
from "never looked", gated versions show a muted "Price unavailable" badge,
and the alerts panel reports unavailableCount.
- Failures are logged at warning level, once per host per TTL, with the
per-host reason, instead of only at debug level.
- The recorded alternatives (internal tRPC with the user's API key, or an
extension-assisted fetch from the user's browser) and the strengthened
upstream ask for a public price field are documented in the plan.
P5a of docs/plans/paid-model-price-tracking.md: one surface that answers "what
got cheaper / became free", without a permanent button (the grid filter was
dropped by owner decision, so the panel carries the actions itself).
- price_alert_since records when an alert started, so the panel can say
"dropped 3 d ago" and count what is new since the user last looked; it is set
on the first sight of an already-cheap version, preserved while the alert
stands, and cleared when the price rises back above the threshold
- get_price_alerts() compares the threshold at read time (editing it takes
effect immediately, no refresh needed) and returns both kinds in one list;
model_type=None covers every type, which the shared update DB makes a single
query. "became free" needs no price data, so it is reported even while price
tracking is off
- GET /api/lm/price-alerts, registered once in MiscRoutes rather than per model
type, decorating rows best-effort with the local model name and file path from
the scanner indexes (a cold cache just omits them)
- a third tab in the notification bell: segments for under-threshold and
became-free, the three states (tracking off / nothing matching / stale), and
per-row actions (CivitAI always, Open when the model is local)
- two non-permanent entry points share one helper: the controls-bar updates
dropdown and the global context menu, whose label carries the unread count
- unread state stays client-side (localStorage watermark); the count is fetched
once on init and only when price tracking is enabled
- the per-type frontend client method is removed as dead code; the per-type
backend route stays for the companion extension
CivitAI's public API deliberately omits prices — paidAccess is trimmed to
{permanent, endsAt} because "pricing belongs to the purchase flow" — but the
public model page embeds the site's own model.getById result, including
paidAccess.terms, in its server-rendered payload. That is read anonymously
(no API key, no internal endpoint, no forged Origin), one request per gated
model, so only the ~2% of models that actually carry a gate pay for it.
- optional capture, off by default: price_tracking_enabled,
price_alert_threshold_buzz (0 = alert on "became free" only) and
price_check_ttl_hours; prices refresh on their own TTL and immediately when a
gate changes, and a failed fetch keeps the stored price instead of blanking it
- versions that stop carrying a gate are marked free (persisted gate_lapsed_at)
and gate transitions are reported as events on the refresh response, so a
version already in the library can announce that it became free
- price_alert_state plus a price_drop edge event; new
GET /api/lm/{type}/updates/price-alerts lists what is under the threshold
- versions tab shows the price (effective, with the list price struck through
and a Blue Buzz note) and a Free Now badge; an update check toasts the
transitions in one message
- the parser and the alerts query are unit-tested against a trimmed page
fixture, and every route definition is now asserted to resolve to a handler
Plan, verification notes and the deviations from it are in
docs/plans/paid-model-price-tracking.md.
CivitAI only returns a non-null paidAccess for an *active* gate: a lapsed gate
stays in the database as a tombstone and is filtered out server-side, so
{"permanent": false, "endsAt": null} — a timed gate whose window end has not
been recorded yet — is still enforced. Verified live: on model 1802980 that
version reports canDownload: false while its lapsed siblings report true.
Both the update service and the download gate dropped that shape, so such
versions read as free and "Hide Early Access Updates" missed them — the class
of bug reported in #1060.
The interpretation now lives in py/utils/paid_access.py and is shared, so the
badge, the update filter and the download warning cannot disagree.
Add OpenModelDB (openmodeldb.info) as a metadata and download source for
the existing upscaler model type.
Metadata:
- New OpenModelDBClient: fetches the site's bulk JSON dumps, caches them
on disk (24h TTL + ETag revalidation), and builds a local sha256 index
- New OpenModelDBModelMetadataProvider adapts catalogue entries to the
CivitAI-shaped version dict contract; registered in the fallback chain
behind the enable_openmodeldb_api setting (default on), gated to the
upscaler sub-type so other model types never trigger the dump download
- Persisted provenance uses metadata_source "openmodeldb" plus a nested
openmodeldb block (page URL, architecture, scale, license)
Images: paired-image LR/SR URLs are ephemeral imgdiff.net sessions, so
displayable images come from the site-hosted auto-generated thumbnails
(model-level cover leads images[], per-image thumbs for the rest); the
original comparison URL is kept in meta.comparisonUrl.
Downloads:
- New OpenModelDBSource (flat model ids, omdb: group prefix) with
resource filename derivation that recovers names hidden mid-path
(mediafire) or synthesizes {id}.{type} for folder links
- HTML-gateway mirrors (mediafire/mega/drive) are rejected with a clear
manual-download hint instead of silently saving an HTML page as .pth
- ModelSource base gains is_valid_source_id / default_subdir_parts /
resolve_download_url hooks so flat-id sources need no platform branches
UI: "View on OpenModelDB" link in the model modal (downloaded and
hash-enriched models), settings toggle next to the CivArchive one.
Applying a filename template to a large library re-did O(library) work for
every renamed file: a full natsort resort plus whole-table SQLite rewrite and
download-history resync after each rename, and a full scan plus resort of the
entire recipe collection per renamed LoRA. On a 20k-model library with 300k
recipes on a HDD this pushed "Apply to Library" into multi-day runs.
- ModelScanner.defer_cache_persist(): bulk loops update the in-memory entry
and indexes only; resort + persist + download-history sync run once at
context exit, forced even on cancellation/error since files are already
renamed on disk. Single-rename callers keep immediate per-call behavior.
- RecipeScanner.build_lora_hash_index(): one-shot hash -> recipes index so
per-file lookups are O(1); update_lora_filename_by_hash gains hash_index /
defer_maintenance params, with a single finalize_bulk_filename_updates()
resort at the end of a bulk session.
- ModelLifecycleService.bulk_rename_session() / BulkRenameContext wire the
deferred path through rename_model (hash index built lazily on first
recipe-touching rename).
- Blocking os.rename sequence offloaded via asyncio.to_thread so one file's
HDD I/O no longer stalls the event loop (no cross-file parallelism).
- Skip logic, per-batch WebSocket progress, cancellation, and result
counters unchanged.
- Model modal hash footnote now shows Civitai model id and version id
(right-aligned, quick-copy buttons); hidden for non-Civitai models
- Hash/id exact search (sha256/autov2/autov3/civitai ids) is now always
on: the search-options "hash" toggle is removed and the search_hash
query param is silently ignored for API compatibility
- Footnote render condition relaxed so autov3-only and id-only models
still show the line
- i18n: 4 new keys translated in all 9 locales; filters.hash key removed
On a CivitAI/CivArchive 429, download-model and download-model-get now
return HTTP 429 with {"reason": "rate_limited", "retry_after": N}
instead of a generic 500 string, and the queue row goes back to
"queued" rather than history as failed — so queue drivers can
auto-pause and retry later instead of burning through the queue.
- new DownloadRateLimitError carrying retry_after/host (opt-in via
raise_on_rate_limit on Downloader; other call sites keep the legacy
string behavior)
- fail-fast pre-flight gate in DownloadManager consults
RateLimitCoordinator before acquiring the semaphore slot: hosts in
cooldown get an immediate structured 429, no HTTP request attempted
- best-effort 429 detection for the aria2 backend
Fills the 4 settings.unknownBaseModelRouting.* placeholders plus the
leftover doctor.issues.sidecar_mirror_orphans.title placeholder per the
feature owner's request, following docs/i18n-translation-guidelines.md:
option labels reuse each locale's checkpoints.modelTypes renderings,
base model follows the §5 matrix, and punctuation/register match each
file's conventions. Adds the feature's Status note and a §2 term-map
subsection for 'routing'.
The setting decides which library an unknown-model download lands in, so
it belongs next to the default-root selects rather than General >
Downloads. Element id, settings key and i18n keys are unchanged, and
loadSettingsToUI() populates it by getElementById on every modal open,
so no JS changes are needed.
Also drop "(recommended)" from the diffusion-models option label; the
default is already conveyed by pre-selection.
Cross-checked both baseModel lists against CivitAI's official
baseModelRecords (packages/civitai-shared src/basemodel.constants.ts):
- CHECKPOINT_BASE_MODELS gains SD 2.0/2.1 768, SD 2.1 Unclip, SDXL 0.9 /
1.0 LCM / Turbo / Distilled, Playground v2 and Stable Cascade
(unCLIP-style but CheckpointLoader-loaded).
- DIFFUSION_MODEL_BASE_MODELS gains SVD XT, LTXV 2.5, Flux 3 Video,
Wan Image 2.7, Wan Video 2.7 / 3.0, HiDream-O1, Boogu and the Ming
Image Design families. API-only (Kling/Sora/Veo/Imagen...), 3D and
audio baseModels are intentionally skipped.
- Pony V7 exclusion now backed by live-API evidence (model 1901521 is
AuraFlow-architecture shipping .gguf variants).
CivitAI has no model-level diffusion ModelType: DiT models are uploaded
as "Checkpoint" or "UNet", with only uploader-chosen file types to tell
them apart. model.type "unet" previously fell through type derivation
and failed with 'not supported for download'; it now goes through the
checkpoint branch in both the download manager and the download routing
endpoint, so the standard chain (file type -> baseModel lists -> unknown
default) applies.
CivitAI labels new DiT architectures (MiniMax H3, future Flux/Wan/Qwen
variants) as model.type "Checkpoint" with plain "Model" file entries,
so the DIFFUSION_MODEL_BASE_MODELS allowlist could never keep up and
such downloads were mis-routed to the checkpoint roots (e.g. model
2877206 / version 3374439). The set of true full-checkpoint families is
closed, so the baseModel fallback is inverted:
1. file type UNet/Diffusion Model -> unet (unchanged)
2. baseModel in DIFFUSION_MODEL_BASE_MODELS (now incl. MiniMax H3) -> unet
3. baseModel in new CHECKPOINT_BASE_MODELS (SD 1.x/2.x/3.x, SDXL, Pony,
Illustrious, NoobAI) -> checkpoint
4. unknown/empty baseModel -> new unknown_base_model_routing setting,
defaulting to diffusion models
The setting is exposed under Settings > Downloads, validated in
SettingsManager, and threaded into both the download manager and the
download routing endpoint so they keep agreeing.
Two gaps kept 'import example images' from working on the Other page:
- import_images/delete_custom_image/set_example_image_nsfw_level only
searched the lora/checkpoint/embedding scanners, so Other-category
models were never found ('Model with hash ... not found in cache').
All three now go through a shared scanner list that includes the
Other scanner.
- Other models (and fresh checkpoints) carry hash_status=pending with
an empty sha256, so the frontend sent an empty model_hash and the
import was rejected with 'Missing model_hash parameter'. The modal
now also sends the model's file path, and the import use case
resolves the hash on demand via the scanner's lazy-hash calculation.
The resolved hash is returned to the UI and persisted on the
showcase element so follow-up operations target the same
hash-keyed folder.
Concurrent or repeated move requests for the same model raced each other:
the first move succeeded, the rest failed with FileNotFoundError, leaving
the model card pointing at stale/empty paths.
- Serialize moves per source file with an asyncio.Lock keyed on the
normalized source path
- When the source file is already gone, reconcile instead of failing:
locate the model via the hash index or the expected target paths, repair
the metadata sidecar and cache entry, and reuse the stale cache entry
when no sidecar exists at the new location
- Avoid duplicate cache entries when the cache already tracks the moved
file; only drop the stale source entry
- Move via business paths (abspath) instead of realpath, matching every
other file mutation and the containment check; realpath stays reserved
for scanner dedup per project convention
The frontend's right-side Properties Panel falls back to WidgetLegacy for
unregistered widget types, and WidgetLegacy.draw() writes widget.width
(≈panel width) onto the real widget object. The canvas DOM overlay honors
widget.width ?? node.width, so clicking a node with the panel open squashes
the widget content to the left until undo/recreate (ComfyUI_frontend #11574).
Setting hideInPanel: true on all addDOMWidget options keeps LM widgets out
of the panel entirely. Older frontends without the option ignore it safely.
Refs #979
model_name, sampler_name and scheduler were declared as COMBO outputs, so the
documented wiring failed at queue time with "Return type mismatch between linked
nodes". ComfyUI only accepts a COMBO output into a node that declares its
dropdown as COMBO/IO.Combo, while Load Checkpoint, KSampler and the LoRA Manager
loaders expose their options as a plain list; comfy_execution.validation rejects
any non-string input type there, and a STRING output is rejected the same way.
Declare the three sockets untyped ("*"), the type ComfyUI's own Primitive node
uses to feed widget inputs. Verified with execution.validate_inputs that they now
link into both classic list dropdowns and IO.Combo inputs.
Also corrects the wiring guide, which claimed COMBO was the supported type.
Add a "Save Recipe with Workflow" action next to "Save Recipe" in the LoRA
widget context menu. It posts the current UI-format graph alongside the save
request so the stored preview embeds it and the recipe can send the graph back
to ComfyUI. Embedding stays opt-in rather than folded into "Save Recipe": the
workflow is by far the largest metadata field and its widget values may carry
sensitive data.
- web/comfyui: new menu entry; saveRecipeDirectly({ embedWorkflow }) posts the
UI graph and reports the outcome (embedded / skipped) via toasts.
- save_recipe_from_widget handler: reads an optional JSON workflow field so the
long-standing body-less POST keeps working, including from cached clients.
- RecipePersistenceService.save_recipe_from_widget: embeds the graph through
the existing optimize_image workflow path, derives has_workflow by detection,
and skips graphs above MAX_WORKFLOW_EMBED_BYTES with workflow_skipped.
CivitAI serves a re-encoded, metadata-free optimized rendition as the recipe
preview, so the ComfyUI workflow embedded in the original image was dropped:
imported recipes reported has_workflow=false and never offered "Send Workflow
to ComfyUI" even when the source image carried one.
Recover the workflow from the original rendition and carry it to the save step
as data, so the stored preview stays the small optimized image:
- ExifUtils: embed a caller-supplied workflow during optimize_image's single
encode pass, and add embed_workflow() to patch WebP EXIF in place (used by
the verbatim skip_optimize branch and as a safety net).
- RecipePersistenceService.save_recipe: embed metadata["workflow"] before
detecting has_workflow.
- analyze_remote_image: return the workflow recovered from the original
rendition it already downloads for EXIF parsing.
- RecipeManagementHandler: add _fetch_original_media() and workflow helpers;
_do_import_from_url reuses them, and _do_import_remote_recipe fetches the
original only when CivitAI reports a ComfyUI payload (meta.comfy) so
workflow-less images pay no extra bandwidth.
- Batch URL imports and the import modal forward the recovered workflow.
Verified against the reported image: has_workflow flips from false to true and
the recovered workflow matches the original (25 nodes, same graph id).
A model root that moved earlier is mirrored under a pinned name derived
from its old path. If anything resolved against the new sidecar path before
the relocation ran, the destination got a map naming the mirror after the
*current* path. migrate_root's keep-newer transfer then dropped the source
map, so the moved metadata stayed orphaned under the pinned component while
reads followed the new name and rebuilt defaults — losing favorites, notes
and tags a second time.
The identity map is now relocated by relocate_root_map() instead of the
generic transfer: entries recorded under the old sidecar root win for the
roots they describe, destination-only entries are preserved, and the cache
is dropped so the next resolution reloads the merged map. A merge that
cannot be written is reported as a migration error rather than silently
stranding the moved metadata.
Reported by the Codex review on #1131.
Centralized sidecars were addressed by a hash of the model root's absolute
path, so moving or renaming a root produced a new mirror directory. The
scanner then found no sidecar there, rebuilt default metadata, and silently
lost favorites, notes, tags and usage tips for every model under that root,
leaving the old metadata orphaned on disk.
Mirrors are now addressed by a root identity pinned in
<sidecar_root>/.lm-sidecar-roots.json. The identity starts as the existing
deterministic <basename>-<path digest> -- so pre-existing mirrors keep
resolving even if the map is lost, and a relocated sidecar root keeps its
names -- and is re-anchored to the root's new path when it moves, matched by
basename and recorded sample directories. Ambiguous matches are never
guessed: the mirror is left untouched and reported.
Also:
- drop the <library> path segment (single-library direction); a legacy
library prefix is only recognised while adopting an existing mirror, which
also keeps the unreleased centralized layout usable
- surface stranded mirrors in Doctor and in the log instead of silently
rebuilding sidecars
- keep the default alongside mode untouched: the identity map is loaded and
reconciled lazily, only while centralized storage is in use
applyTheme() mirrors the active mode onto <body> as data-theme="dark", but
the theme preset is only ever written to <html>. The palette token blocks in
tokens/colors.css and base.css used the bare attribute selector
[data-theme="dark"], so <body> matched them on its own and re-declared the
default dark palette (#1a1a1a / #2d2d2d / ...) directly on the body, where it
shadowed the preset values inherited from <html>. Every non-default preset
therefore painted the selected accent over the default preset's background,
surface, text and border tokens, and flipped into that state ~200ms after
load, when initTheme() first touched <body> — the accent-tinted background
flash seen on reload and nav-tab switches. Reached only in dark mode, since
light mode has no [data-theme="light"] token block.
Scope the palette token blocks to :root so a data-theme attribute on any
descendant (only <body> has one) can no longer re-declare them; descendant
rules such as [data-theme="dark"] .foo still match through <html>. Add a
regression guard that fails on bare attribute token blocks.
set_hf_url rejected files under config.other_roots (VAEs, text encoders,
upscalers, ...) with 'File is not within any configured model directory'
because neither _find_matching_root nor _infer_model_type knew about the
Other category. Include other_roots in both, so linking routes the cache
update to the Other scanner instead of falling back to the LoRA one, and
downloads into Other roots keep the lazy-hash metadata path.
Also make the root prefix match boundary-aware so /models/vae no longer
swallows a sibling like /models/vae-old.
Fill the [TODO: Translate] placeholders for the 7 settings open*/path
keys, the confirm-dialog destination line and the 13-key migration
summary block in all 9 locales, per the sidecar storage terminology in
docs/i18n-translation-guidelines.md (new term rows added there for
storage location / Open Folder / installation folder).
Follow the app's existing operation-summary convention (Metadata Fetch
Summary / Batch Download Summary): a self-managed modal appended to
document.body with a 3-state summary header, stat cards (moved /
models / skipped / conflicts / errors), and a failure table listing
per-model errors that were previously swallowed into a single count.
The storage location line and Open Folder action move into the modal
actions; the page reload still happens only when the modal is
dismissed. ESC is captured so it never reaches the settings modal
underneath. The obsolete migrateSuccess toast key is dropped — the
modal is the success feedback now.
Addresses PR review: on a native Linux/SSH session with neither DISPLAY
nor WAYLAND_DISPLAY (and not Docker/WSL), _open_path unconditionally
launched xdg-open and reported success even though no file manager can
open. Mirror open_settings_location: hand the path to the browser for
copying instead. Fixes open_backup_location, open_wildcards_location
and open_sidecar_location together.
After migrating to centralized sidecar storage users had no indication
where their files went, and portable-mode installs silently placed the
sidecar root inside the plugin folder where a reinstall or git clean
would delete it.
- Migration now also covers models excluded from the library view and
returns the resolved sidecar root in its result payload
- get_settings exposes the resolved sidecar root, whether it is the
default, and whether it lives inside the installation folder
- New POST /api/lm/sidecars/open-location endpoint opens (or copies)
the sidecar storage folder
- Settings UI always shows the effective storage path with an
open-folder button, and warns when the root is inside the
installation folder (portable-mode hazard)
- Migration confirmation shows the destination; on completion a result
dialog summarizes moved/skipped/conflict counts with the storage
location and an open-folder action
- Ignore /sidecars/ at the repository root so portable-mode sidecars
are never committed
Refs #1045
Fills in the 23 keys that came with optional centralized sidecar storage
(settings.sections.sidecarStorage, the 18 settings.sidecarStorage.*
labels/help/status/confirm strings and the 4
modals.sidecarMigrationConfirm.* titles/button), which the feature PR merged
as [TODO: Translate] copies. No placeholder remains in any locale.
Each locale reuses its own storage-relocation verb rather than a
transliteration of "migration" (ja 移動, ko 이동, ru перенос, matching
settings.folderSettings.recipesPathMigrating), its existing "preview images"
and nav-path renderings, and quotes the migrate button label the way its
other UI-label references do. `.metadata.json`, `.civitai.info` and the
`(<settings dir>/sidecars)` literal stay verbatim.
docs/i18n-translation-guidelines.md gains the "Sidecar storage feature" term
table and pass note, and refreshes the leaf-key count to 2128.
Codex review on #1124:
- P1: mirror layout root component is now <basename>-<roothash>
(sha256 of the normalized root path), so two roots sharing a basename
no longer map to the same mirror directory and overwrite each other's
sidecars
- P1: changing sidecar_storage_path while centralized no longer strands
assets in the old root — new relocate_root migration direction moves
the whole mirror tree, rewrites preview_url prefixes inside sidecars,
reconciles scanner caches, and prunes the emptied old tree; the
settings UI detects the path change and offers the relocation
- P2: migration enumerates the same preview candidates as
find_preview_file — case-insensitive variants (model.WEBP) and the
legacy .example.0.jpeg suffix — instead of exact lowercase
PREVIEW_EXTENSIONS only
- P2: _rollback_model_staging restores staged files with the
EXDEV-tolerant mover, so a failed undoable-delete staging no longer
strands a cross-filesystem centralized sidecar copy
Tests: same-basename root injectivity, mixed-case/example preview
migration, relocate_root happy path + guards + route 400, frontend
relocation prompt flow. Verified end-to-end in a sandboxed standalone
server: uppercase/legacy previews migrate, root relocation moves the
tree and the list API serves the new locations immediately without a
rescan.
Fills in the 9 locale renderings for the 5 keys added with the folder delete
verification (`sidebar.deleteFolderModal.notEmptyMessageCount`,
`.notEmptyMessageExcluded`, `.busyTitle`, `.checking` and
`sidebar.deleteFolderResult.notEmptyWithCount`), so no `[TODO: Translate]`
placeholder remains anywhere.
Each locale reuses its existing `notEmptyMessage` cascade clause verbatim
(punctuation included) and its help-text style for quoting the "Manage
Excluded Models" label; `{count}` / `{excluded}` match en.json. The Hebrew
model-file noun stays `קובצי מודלים` across all four keys.
docs/i18n-translation-guidelines.md records the new copy in §2 and refreshes
its stale `{count}` and leaf-key counts.
The sidebar derives "empty folder" from the models-only list, which omits
models flagged `exclude: true`, while the delete guard walks the folder on
disk and refuses on any weight file. A folder whose models are all excluded
therefore looked empty, offered the confirmation, and then failed with
"still contains models".
The delete modal still opens on that prediction, but is now corrected by a
dry run of the very delete the user is about to confirm, so the button state
cannot contradict the backend. The confirm button stays disabled while the
check runs, and a late answer is discarded once the modal is dismissed or
retargeted. The dry run also covers weight files no scanner indexes (a lora
folder holding only a `.gguf`, say) and files that appeared after the last
scan.
`_collect_folder_manifest()` now reports `excluded_model_count`, and the
refusal names the excluded models, so the message explains the mismatch
instead of reading like a bug. Locale files carry the sync placeholders in
this commit; the translations follow.
Sandbox E2E showed that after a migration the list API kept serving
pre-migration preview_url values; the first request to a stale URL made
the preview route's stale-URL cleanup wipe the reference from the cache
entirely, recoverable only by a full rebuild rescan.
The use case now records each migrated model's final preview location
(from the destination directory, covering conflict-keep cases), updates
the owning scanner's cache entries via ModelCache.update_preview_url,
and persists the cache. Per-scanner reconcile failures are logged and
skipped; per-model migration errors no longer prevent reconciliation of
the healthy models.
Verified end-to-end in a sandboxed standalone server: after
to_centralized and to_alongside migrations the list endpoint immediately
returns the correct preview URLs with no rescan, previews serve with
HTTP 200 in both layouts, and the mirror tree is empty after migrating
back.
Add an opt-in 'centralized' sidecar storage mode alongside the default
'alongside' layout. In centralized mode, .metadata.json sidecars and
preview assets live under a configurable root (sidecar_storage_path,
default <settings_dir>/sidecars), mirroring the library-relative
directory structure: <root>/<library>/<root_basename>/<rel_dir>/.
Backend:
- settings: sidecar_storage_mode / sidecar_storage_path with validation;
changing either refreshes the preview allowlist
- config: centralized root added to preview-serving allowlist
- lifecycle: delete / move / rename / folder-rename / folder-delete and
undoable-delete staging all operate on the mirror tree in centralized
mode (model files themselves never move); EXDEV-tolerant cross-
filesystem moves
- scanners: pending-hash filesystem scan walks the mirror tree in
centralized mode; preview discovery reads from the sidecar dir;
.civitai.info stays co-located in both modes
- migration: SidecarMigrationUseCase moves sidecars+previews between
layouts both directions (keep-newer conflict resolution, preview_url
rewriting, WebSocket progress), exposed as POST+GET
/api/lm/sidecars/migrate with a mode guard (force=true for the
settings-first flow)
Frontend:
- settings modal: sidecar storage section (mode select + path input with
browse/validation), mode-change confirmation offering immediate
migration (force=true), and a 'Migrate Sidecars Now' action
- i18n keys synced to all locales ([TODO: Translate] placeholders)
Docs: metadata-json-schema.md gains a storage-location section;
AGENTS.md records the sidecar_paths helper convention.
Phase 1 of #1045 (optional centralized sidecar storage): introduce
py/utils/sidecar_paths.py as the single place that resolves .metadata.json
and preview locations, and replace all inline splitext-based derivations
across scanners, services, download manager, and route handlers.
No behavior change: the default 'alongside' storage mode resolves every
path exactly as before. .civitai.info (third-party sidecar) derivation is
intentionally left co-located.
format_response whitelists fields explicitly, so the new ModelScope
identity fields never reached the frontend: group badges rendered, but
card.dataset.modelId stayed empty and clicking 'N versions' silently
no-oped (handleViewLocalVersionsFromCard early-returns without it).
Found via browser E2E against a sandboxed standalone server.
A repository is not a model identity: collection repos on Hugging Face
and ModelScope host many unrelated models, which were wrongly shown as
versions of each other.
- Hugging Face models no longer auto-group (the Hub exposes no
site-native model id)
- ModelScope models group by the site's native published-model id
(MuseInfo modelVersion.modelId), extracted during enrichment and
persisted on the sidecar as source_model_id/source_version_id;
unenriched models stay standalone instead of collapsing a whole repo
into one group
- TensorArt grouping unchanged (its URL id is already model-level)
- Frontend group-key derivation mirrors the new backend semantics
Fill the six settings.huggingfaceApiKey* placeholders left by the HF
access-token feature in all 9 locales, reusing each locale's existing
civitaiApiKey* status renderings; document the new terminology
(access token, gated repository) in the translation guidelines.
Add a huggingface_api_key setting (Settings UI, HF_TOKEN /
HUGGING_FACE_HUB_TOKEN env override) and attach it as a Bearer token
to Hugging Face file listing, model card fetching and downloads, so
gated and private repositories can be downloaded once the user has
accepted the repo terms.
- fetch_json/fetch_text accept custom headers; ModelSource gains an
auth_headers() hook so handlers stay platform-agnostic
- 401/403 from the tree API now explain how to fix (configure token /
accept gated terms)
- aria2 pre-resolves huggingface.co redirects and strips credentials
before handing the signed CDN URL to aria2, mirroring the CivitAI
handling so the token never leaks to the CDN
- settings API exposes huggingface_api_key_set only; the raw key joins
_NO_SYNC_KEYS
ComfyUI frontend 1.53 turned LGraphNode.mode into a prototype accessor
backed by node._state, and serialize() now reads that state directly.
Redefining mode on the instance shadowed the setter, so bypass/mute
never reached the serialized workflow and silently reverted to Always
on save/reload or workflow tab switch.
Add interceptModeChange() in web/comfyui/utils.js: it delegates to the
prototype accessor when one exists (observing changes only), and falls
back to the legacy closure accessor on older frontends. Use it in
lora_loader.js and in the Vue widgets' setupModeChangeHandler, which
covers the LoRA provider/aggregator nodes with the same latent bug.
IS_CHANGED only receives constant inputs, so a linked text always
arrived as None and the node kept serving its cached first expansion.
Declare hidden PROMPT/UNIQUE_ID inputs and walk the prompt graph to
the upstream node: rerun only when its constants contain dynamic
syntax or cannot be statically resolved, keep caching for static
linked text.
Fixes#1120
When metadata extraction succeeds but no recognized latent source
provides dimensions (e.g. img2img via VAEEncode), width/height now fall
back to the source image size from the loaded pixels instead of the
synthetic 1024x1024 starter preset. The starter preset for metadata-free
images keeps its fixed size, and explicit overrides still win.
Follow-up to the keyword-dump guard. The reported model's tag list is
["lora, character, ... face", "base model"], so skipping the dump left the
"base model" label to be picked as the folder name. That label describes
Civitai's listing rather than the model's content, which makes it as
meaningless as a folder as the blob was.
Add CIVITAI_META_TAGS and is_civitai_meta_tag(), and skip those labels in
the automatic fallback. An explicit priority entry still matches them, so a
user who does want a "base model" folder can configure one.
The reported model now resolves to "Krea 2/no tags" instead of
"Krea 2/base model".
Also correct a comment that listed ".civitai.info" among the files sitting
next to a model. LoRA Manager only reads that sidecar -- other tools write
it -- and writes ".metadata.json" itself.
CivitAI tags are normally short single-concept labels, but some uploaders
pack their entire keyword list into one tag. The model in #1119 carries
"lora, character, rosie, irish, ... face" as a single 181-character tag.
Priority resolution matches aliases by exact equality, so that tag matched
nothing and resolve_priority_tag_for_model fell back to tags[0] -- the blob.
With the default "{base_model}/{first_tag}" template the model was filed
under "Krea 2/<181-character blob>/", and the full path plus the
".civitai.info" sidecar and the preview images next to it ran into the
Windows MAX_PATH limit.
Tags also bypassed sanitization on the way into a path: both
calculate_relative_path_for_model and DownloadManager._calculate_relative_path
sanitized model_name and version_name but interpolated {first_tag} verbatim,
so a tag containing "/" or ":" silently produced nested or illegal folders.
Two changes:
- The fallback skips tags that cannot serve as a folder name.
is_usable_path_tag rejects comma-separated keyword dumps and tags longer
than MAX_PATH_TAG_LENGTH; the resolver returns "" when nothing usable is
left, which callers already render as "no tags". Whole-tag priority
matching is untouched, so existing priority configurations behave the
same.
- sanitize_folder_name gains an optional max_length, and every tag-derived
segment now goes through it. Tags are capped at MAX_PATH_TAG_LENGTH, model
and version names at MAX_FOLDER_NAME_LENGTH, and rendered filename stems at
MAX_FILENAME_STEM_LENGTH.
For the reported model the folder becomes "Krea 2/base model" instead of the
blob, and the full path drops from 235 to 64 characters.
Existing libraries are not migrated up front: a path is only recomputed on
download, on an auto-organize run or when a filename template is applied, and
values already inside the caps are left byte-identical. Models previously
filed under a keyword-dump folder move on the next auto-organize run.
26 keys (settings.filenameTemplates.*, filenameTemplateProgress,
modals.filenameTemplateConfirm, related toasts) translated into the 9
non-English locales, reusing each locale's autoOrganizeProgress /
downloadPathTemplates renderings. Terminology recorded in
docs/i18n-translation-guidelines.md.
Redefine the empty download filename template from a no-op to a bulk
revert: FilenameTemplateUseCase resolves the target from each model's
recorded original_file_name sidecar entry (skipping models without one),
which resolves follow-ups 1 and 2 with a single coherent semantic shared
by the download and bulk-apply paths.
Also replace the browser-native confirm() with a self-managed
confirmation modal (filenameTemplateConfirmModal) that stacks above the
settings modal, since ModalManager would close the settings modal when
opening a registered one.
Add per-model-type filename templates ({model_name}, {version_name},
{base_model}, {author}, {first_tag}, {hash_short}, {original_name}) so
downloaded files get informative names instead of e.g. V1.safetensors.
Empty template keeps the current filename (opt-in, off by default).
- apply template automatically after downloads; rename conflicts keep
the original name and never fail the download
- record original_file_name in metadata on rename for traceability
- bulk apply via GET|POST /api/lm/{prefix}/apply-filename-template with
WebSocket progress, sharing the auto-organize lock
- settings UI lives in the new Organization tab with validation, live
preview, and per-type 'apply to library' actions
Move download path templates, priority tags, and auto-organize
exclusions out of the Library settings section into a dedicated
Organization section, so Library keeps location-focused settings
(roots, extra paths, example images, metadata) and Organization holds
file-arrangement rules. Translated settings.nav.organization for all
locales.
The cycler-list payload now carries usage_tips, and the LORAS widget
parses strength_min/strength_max/strength_range into a cached lookup.
Strength inputs (model and clip) turn amber with an explanatory tooltip
when dragged, typed, or stepped outside the recommended range.
Related: https://github.com/willmiao/ComfyUI-Lora-Manager/issues/1090
Stacking every active banner vertically ate header height when several
were active at once. Only the highest-priority banner renders now; a
‹ 1/N › pager cycles through the rest, and all active banners are still
recorded in the notification-center history so cycled-away ones stay
reachable. Newly registered banners preempt the displayed one only when
they outrank it.
Also fix the startup flow: the restart-required banner (now priority 80)
outranks the model-folders setup warning (60), and the setup banner is
retired once a non-empty folder path is saved.
New banners.pager.* keys translated in all 9 locales.
Add a reusable directory-picker modal backed by a new generic
POST /api/lm/browse-directory endpoint (browse logic extracted from the
recipe batch-import handler into py/utils/directory_browser.py) and wire
a browse button plus advisory validate-path feedback (POST
/api/lm/validate-path) into the settings path inputs: recipes path,
example images path/local root, and the extra-folder/model-path rows.
The browse button insets into the right edge of static inputs so narrow
settings rows keep their single-control layout.
Translations for the new settings.directoryPicker and
settings.pathValidation keys are filled in for all 9 locales.
Standalone users previously had to hand-edit settings.json to configure
primary folder_paths. Add a standalone-only Model Paths section to the
settings modal:
- Backend exposes standalone_mode, folder_paths (with template placeholder
values filtered out) and a data-driven folder_path_schema derived from
OTHER_MODEL_FOLDER_SUBTYPES via GET /api/lm/settings
- The new section renders multi-path editors per model type from the
schema, with inline enable_other_models / sub-type controls so other
model types are configured without leaving the tab
- Persistent restart-required cues after a save: nav dot, inline notice
and a global banner (unique id per change so dismissals don't mute
future reminders)
- The missing-model-paths startup banner and the Other Models no-paths
empty state now deep-link into the new section instead of pointing at
settings.json
The prune guard intentionally leaves the in-memory view empty while the
stored cache keeps the user's recipes, so there has to be a documented
way to accept the on-disk truth. That route is an explicit rebuild, which
clears the stored cache before a full directory scan. Cover it so the
FAQ recovery steps stay true.
Setting LORA_MANAGER_PORTABLE=1 once wrote use_portable_settings: true
into the plugin's own settings.json, and every later run of every
instance sharing that plugin folder then read and wrote the portable
settings directory. There was no way back except editing the file by
hand, which is exactly the trap a user hit while following the FAQ's
instructions for isolating a second instance (#1114).
LORA_MANAGER_PORTABLE=0 is now the explicit exit:
- _should_use_portable_settings honours "0" as a forced off, so the
resolved settings directory no longer depends on the persisted flag.
- SettingsManager clears the persisted flag in that case, so later runs
without the variable stay on the shared settings directory.
Unset or unrecognised values keep the previous behaviour: the persisted
flag decides, so existing portable installs are unaffected.
LORA_MANAGER_SETTINGS_DIR still takes precedence over both.
Installing a second LoRA Manager instance (standalone or a second
ComfyUI install) that shares the settings directory puts two processes
on the same cache databases. Three things made that unsafe.
- The updater preserved cache/ and model_cache/ but not a legacy
recipe_cache/ directory, so a portable install predating the cache/
move lost its recipe database on a git-based update. Add it to
_PRESERVE_DIRS and to .gitignore.
- Cache connections used the sqlite3 default 5s timeout, which a
scanning instance can exceed, turning a concurrent write into
"database is locked". Route every shared cache connection through
connect_cache_db(), which raises the timeout to 30s and sets
busy_timeout + synchronous=NORMAL to match the existing WAL mode.
App-private databases (download queue, update history) are unchanged.
- A full-table cache replace is a read-modify-write that SQLite cannot
make atomic across processes, so two instances could interleave and
one snapshot could overwrite the other. Guard the recipe and model
save_cache paths with a cross-process advisory lock (flock on POSIX,
msvcrt on Windows). Locking is best-effort: if it is unavailable the
call proceeds and the SQLite busy timeout is the fallback.
The lock file is a hidden sibling of the database and is deliberately
never unlinked, so a second process cannot lock a fresh inode.
A scan that finds no recipe files at all is not a reliable deletion
signal: an unmounted drive, a recipes_path that silently falls back to
another LoRA root, or a cache shared with a second instance all look
exactly like a real wipe. The reconcile step treated them all as
deletions and overwrote the persistent cache with an empty one, so
DELETE FROM recipes destroyed the user's only record of their recipes
and the FTS index was rebuilt from the empty view (#1116).
Guard the prune:
- _reconcile_recipe_cache reports an all-missing result when every
persisted recipe file is gone AND the stored rows match the recorded
file stats. An internally inconsistent cache (leftover orphans) is
stale, not evidence of a fresh disappearance, and still prunes.
- The caller keeps the stored cache and logs a warning naming the
directory it scanned and the number of recipes it preserved, instead
of writing the empty result. It also skips the FTS rebuild so the
index stays aligned with the stored rows.
- save_cache gains skip_if_empty as a storage-level backstop: refuse to
empty a populated cache. Intentional clears (manual rebuild) keep the
default behaviour.
- Log the resolved scan directory per run so a support reader can tell a
real wipe apart from a scan that looked elsewhere.
Partial orphans (ordinary manual deletions) keep pruning as before.
Example videos added through the "Add examples" flow were stored with a
hardcoded 720x1280 entry. The dimension probe next to it only ran for
images (PIL cannot open .mp4/.webm files), so every video entry stayed
portrait regardless of the source. The showcase viewer then sizes its
container straight from that value (--media-aspect in showcase.css), so
landscape clips were letterboxed inside a 9:16 box. CivitAI-sourced
examples were unaffected because their dimensions come from the API.
PIL cannot read video containers, so add a dependency-free reader that
parses the container headers instead: moov/trak/tkhd for ISO base media
(with the sample description as a fallback), Segment/Tracks/Pixel* for
WebM/Matroska, and RIFF/WebP for animated examples saved with a video
extension. The sniffed signature decides which reader runs, so a .mp4
that is really WebM still reports the right size; the extension is only
a fallback. Both readers seek past mdat rather than reading it, so a
large file costs the same as a small one.
Imported entries now record the file's real size and keep the previous
placeholder only when the file cannot be parsed.
Existing libraries keep their wrong entries, so backfill them once via
the existing naming migration: bump CURRENT_NAMING_VERSION to 3 and
repair each model's empty-url entries from the files on disk, then sync
the scanner cache. Only entries with no remote url are touched -- those
have no other source, which makes the rewrite lossless -- and entries
already carrying the right size are left byte-identical, so the pass is
idempotent and a no-op for libraries that never imported a video.
Civitai types with no scanner at all (Wildcards, Workflows, Hypernetwork,
Poses, AestheticGradient) used to get a 400 'Model type "x" is not
supported', which hid the Civitai version list from clients.
The handler now answers 200 with supported:false, a machine-readable
reason (model_type_unsupported, or other_models_disabled when the opt-in
master switch is off) and the versions marked read-only. The interactive
payload gains an explicit supported:true. Legacy clients only read
success/versions, so they are unaffected.
The standalone empty state showed the folder_paths keys but not where to
put them, and its Open Settings button led to a modal that cannot edit
primary folder paths. Now the page shows the real settings.json path and
an Open Settings Folder button backed by the existing open-location API.
Also stop open_settings_location from claiming success on headless Linux
sessions: with no DISPLAY/WAYLAND_DISPLAY, xdg-open cannot work, so the
handler now returns clipboard mode and the browser copies/shows the path
instead.
start_server.py computed the project root three levels up from scripts/,
assuming it lived under .agents/skills/<skill>/scripts/. After moving to
scripts/e2e/ that resolved to the ComfyUI root, so the launcher failed
with "can't open file 'standalone.py'".
updateOtherModelsControls() synced the sub-type checkboxes and default-root
selects but never set the master toggle's checked state, and the
setting_toggle macro renders no checked attribute, so after a page refresh
the toggle always appeared off regardless of the saved setting.
The reorder shortcut cannot be made reliable in this UI. `Alt + Arrow` is
the browser's tab-history / back-forward gesture on several platforms,
and the modal already binds bare `ArrowLeft`/`ArrowRight` to model
navigation, so the binding either did nothing — a keypress with nothing
focused never reaches a listener on the tag list — or fought the browser.
An affordance that occasionally navigates the page away is worse than
having no keyboard path at all, so drop it.
Reordering is pointer-only again: drag the chip (tags) or its `⠿` grip
(trigger words, whose chip body is click-to-edit). Everything that existed
only to serve the shortcut goes with it — the keydown listener, the hover
tracking used to resolve the target chip, the aria-live announcements, the
per-grip position labels and `moveItemWithinContainer`. The grip becomes a
decorative, non-focusable `<span>` (`aria-hidden`, behind a 5px drag
threshold) instead of a `<button>`, so it no longer promises a keyboard
action it cannot perform.
The tooltip and hint drop the shortcut mention in all 10 locales
(`common.reorder.dragHandle` = "Drag to reorder" and the localised
equivalents); `common.reorder.ariaLabel` and `common.reorder.announcement`
are pruned from every locale by the sync script. The i18n guidelines
record the decision so no shortcut is re-added without re-adding the keys.
Two stacked defects cut the side edges off the URL textareas in the
download and batch-import modals.
`#modelUrl` and `#batchUrlInput` are `width: 100%` with padding and a
border but no `box-sizing: border-box`, so the border box was wider than
the containing block and its right edge landed in the region the modal
clips: the right border column is missing in both screenshots while the
corner pixels of the top/bottom borders are drawn, and the batch
textarea's resize handle sits a padding-width to the right of the mode
toggle above it.
The download modal's `#downloadModal .download-step` additionally
scrolls with `overflow-x: hidden` and has no horizontal padding, so the
global `:focus-visible { outline-offset: 2px }` lost both vertical edges
there and only the top and bottom lines survived. Draw that ring inset
inside `#downloadModal`, mirroring the existing `#importModal` fix in
import-modal.css.
`.input-group input, .input-group select` gets the same border-box
treatment, which also repairs the standing clipped right border on the
other full-width fields the shared rule styles (the import modal's URL,
recipe-name and tag inputs, the batch directory and tags inputs, the
model root select and the target folder path).
Verified: `npx vitest run` 130 files / 1259 tests passed.
Fill in the 4 `modals.download.progress.*` keys added by the previous
commit across all 9 locales, so no `[TODO: Translate]` placeholder remains
and the "no remaining placeholders" claim in the guidelines holds again.
No new terminology: `metadata` reuses the §5 row (fr métadonnées, de
Metadaten, es metadatos, ru метаданные, he מטא-נתונים, ja メタデータ,
ko 메타데이터, zh-CN 元数据, zh-TW 中繼資料) and the fetching phrasing
mirrors each locale's existing `download.fetchingRepoFiles` /
`fetchingVersions` (de passive "werden abgerufen", es "Obteniendo", fr
"Récupération des", ru "Получение", he "מביא", ja "取得中", ko "가져오는
중"). "model file" follows `errors.noModelFiles` in each file.
`{name}` and `{source}` are verbatim §1-R2 placeholders. `{source}` is
replaced at runtime with the *untranslated* platform name, so its
surrounding spacing follows each locale's `modelCard.actions.viewOnSource`
precedent — ja `{source} から`, ko `{source}에서`, zh `从 {source}` /
`從 {source}`, he `מ-{source}` (as in the existing `מ-CivitAI`), ru
`из {source}` (as in `из Workflow`) — and no brand ever appears inside the
translated text.
Punctuation: ASCII `:` for the Latin / Cyrillic / Hebrew locales and for
ja / ko, whose four sibling keys in the same `progress` block already use
ASCII; French keeps this file's ` : `; zh-CN / zh-TW use full-width `：`
like their siblings.
The guidelines gain a status block recording the pass and those spacing
precedents, so a future source added to the same slot does not have to
re-derive them.
Verified: `pytest tests/i18n/test_i18n.py` 20 passed,
`sync_translation_keys.py --dry-run` reports no drift, `npm test` exits 0
(1259 JS + 91 Vue). Each locale file gains exactly 4 lines — the values
were substituted as literals rather than re-serialising the JSON, so no
formatting churn.
A ModelScope or Hugging Face download landed as a bare filename, hash and
source link; the model card stayed empty until the user ran "Enrich
Metadata with AI" by hand. But everything that makes a CivitAI download
useful — the display name, the description, the tags, the trigger words,
the example images, the preview — is already published by those sites'
public APIs, so asking for it at download time is deterministic work, not
model work.
Add `py/services/model_sources/hydration.py`, called by
`_save_source_metadata()` once the sidecar exists and the file is in the
scanner cache. It fetches the model card plus the site's card extras and
hands them to the same `PostProcessor` the AI skill uses, with an empty
`llm_output`, so the two paths cannot drift apart. What lands:
* `model_name` from the site's own display name (ModelScope's `Name`), so
the card stops showing the local filename — written only while the value
still equals the file stem, since once a user renames a model that
choice is theirs to keep
* `civitai.name` from the matched version's label (`showName`), which the
card renders as the version chip
* `civitai.description` / `modelDescription` from the author summary plus
the README as HTML
* `civitai.images` / `preview_url` from the per-file example images
* `civitai.trainedWords` from the per-file trigger words
* `base_model`, `tags` and `usage_tips` as before
Provenance stays honest: the pass records
`metadata_source = "source:<platform>"` rather than the skill's
`agent:enrich_hf_metadata`, and — because no provider ran — it no longer
stamps `llm_enriched_at`; that stamp is now conditional on the LLM
actually answering, which is what the field means. The five hand-rolled
`civitai` dict merges in the post-processor collapse into one
`_merge_civitai()` helper.
Two guards keep it safe. Only a model whose stored
`source_platform`/`source_url` match the repository being downloaded is
updated, so a local file that merely shares a name never receives another
model's card; and a file already on disk is topped up too, which
back-fills models downloaded before this existed. READMEs and detail
payloads describe the repository rather than the file, so a short-lived
process-wide `ModelSourceCache` (300 s, 32 entries) keeps a batch over one
repository to two HTTP requests. Every failure is logged and swallowed:
hydration can never fail a download.
Fix the hash policy while here. `_save_source_metadata()` went straight to
`MetadataManager.create_default_metadata()`, bypassing the per-type
factory on the owning scanner, so a checkpoint paid a full SHA256 inside
the download request — `CheckpointScanner`/`OtherScanner` deliberately
record `hash_status="pending"` with an empty `sha256` for their multi-GB
files. Metadata is now created through `scanner._create_default_metadata()`.
Hydration copes with the empty hash: `_matching_versions()` falls back to
the repository basename, which is exactly what the download just wrote.
Report both post-transfer stages, which advance no byte counter and so
read as a stall: the bar sat at 100% showing `0 B/s` for the seconds spent
hashing and fetching. `_report_phase()` broadcasts
`{"status": "metadata", "stage": "indexing" | "source", "platform": ...}`,
and `LoadingManager` names the stage in the status line (keeping the batch
position), retitles the item line, replaces the dead speed figure and runs
a sheen over the bar. `stage`/`platform` are machine-readable; the wording
is localised in the frontend.
Finally, `modelscope.ai` is its own catalogue rather than an alias of
`modelscope.cn` — `referall13/EM1` exists only on `.ai` and
`jj3550945163/Krea-2-LORA` only on `.cn` — so its URLs were rejected with
"Invalid model URL format". Register it as `ModelScopeIntlSource`
(`platform="modelscope-ai"`, `msai:` group prefix, its own default
download directory) and derive every URL either deployment builds from a
per-class `base_url`. `modelscope.com` stays an alias of `.cn`, which is
what it redirects to. The frontend source table, the link dialog hints and
the docs mirror the split.
Verified against the live APIs: both reported `.ai` repositories list
their files, read their READMEs and yield name / version / base model /
trigger words / example images. Backend 3092 passed; frontend 1259 JS +
91 Vue passed. The nine locales carry the new progress copy in the next
commit.
Adds a base model pill at the front of the recipe modal's tags row,
showing the full base model name (cards keep the abbreviation since
their overlay width is constrained). Falls back to a dimmed Unknown so
the header layout does not shift when hydration fills the value in.
Hydration now also merges base_model. Translated in all 9 locales.
`civitai.trainedWords` is an ordered array, and the order is what gets
pasted into a prompt: "Copy Trigger Words" and the insert-into-node
action join it as-is. The refresh merge unioned the stored words with the
freshly fetched ones via `list(set(...))`, so any metadata refresh
silently shuffled a user's ordering into an arbitrary one. Now that the
UI exposes reordering, that would look like the feature losing the change
at random.
Merge in order instead: stored words first (in their saved order), then
newly discovered ones, duplicates dropped. `_merge_ordered_unique` keeps
the behaviour easy to assert, and the existing merge test keeps passing
because it compares the result as a set.
The three `common.reorder.*` keys (grip tooltip and hint, the per-grip
aria label, and the aria-live announcement) are rendered in all 9
locales. They sit under `common` rather than in a feature namespace
because both the tag editor and the trigger-word editor render them, and
only `dragHandle` is visible copy — the other two are screen-reader text.
`Alt` and the `↑/↓` glyphs stay verbatim everywhere, the same precedent
as `Shift+Enter` in `modals.model.metadata.notesHint`, because they name
the keys rather than an action. "position X of Y" reuses each locale's
existing counting phrasing (ja `{total} 件中 … 番目`, ko
`총 {total}개 중 …번째`, fr `sur {total}`, ru `из {total}`), and
parentheses follow each file's own convention: full-width in zh-CN /
zh-TW / ja, ASCII in ko and the Latin/Cyrillic locales. No
`[TODO: Translate]` placeholder is left in any locale.
docs/i18n-translation-guidelines.md gains the matching §2 subsection and
status note so a later terminology sweep preserves these renderings; the
leaf-key count in its header is corrected to 2025 at the same time.
Model tags could already be reordered by dragging a chip, but the only
hint was a `cursor: grab` on `.metadata-item` — a hover-only, mouse-only
signal that also leaked into the bulk add-tags modal, where the chips are
not sortable at all. Trigger words could not be reordered, and their
order matters: "Copy Trigger Words" and the insert-into-node action join
the array as-is to build a prompt.
Both editors now share one vocabulary: a `⠿` grip that appears whenever
the list has something to order, plus `Alt + arrow` keyboard moves with
an aria-live announcement. Whether the chip body is draggable is a
property of the item rather than of the feature:
- tags have no click action of their own, so the whole chip stays
draggable (`handleSelector: null`), with a 5px threshold so a click on
the grip only focuses it
- trigger words keep click-to-edit on the body, so a drag starts from the
grip only
- the grip is the element that opts out of touch scrolling
(`touch-action: none`), so touch users drag by the grip in both editors
- reordering is offered only while editing: trigger words reveal the grip
from `.edit-mode`, tags from the edit container, which stays hidden
outside edit mode
The drag engine moves out of ModelTags.js into shared/pointerSort.js,
which now marks sortable containers with `pointer-sort-enabled` so only
lists that really sort show the grab cursor. Labels, the sortable flag,
the keyboard handler and the live region live in
shared/reorderSupport.js, and both editors render the same three
`common.reorder.*` keys (translations follow in the next commit).
Two inherited engine bugs are fixed on the way: the drop position was
only settled when an animation frame was still pending, so a fast drag
(or one that started by crossing the threshold) fell back into its
original slot; and the guard that stops a drop from triggering the chip's
own click handler was removed on a timer, swallowing unrelated clicks
until the next task.
The sidebar used the models-only folder list while the download and move
destination pickers list every directory, so a model downloaded into an
empty category folder did not appear in the sidebar at all. Empty folders
are also deliberate organization on disk, and a file-manager-shaped tree
that hides them is surprising. Default the preference to on.
Because the preference no longer gates fetching, both folder lists are
always loaded: the full list is the tree's single source of truth and the
empty-folder count, the models-only list is what "empty" is measured
against. That makes the view-options toggle a pure re-render, and lets the
new count decorate the "..." menu so the preference's effect is visible
without scanning the tree:
- empty-folder count shown next to the menu label, cleared when unknown
- the toggle is hidden entirely when there are no empty folders
- list view filters empty entries itself (the tree gets that for free
from the backend, the flat list is now always loaded in full)
- creating a folder still re-enables the preference, since the folder the
user just asked for would otherwise be invisible
Dim styling is decided by _isRenderedEmptyFolder() at render time; the
models-only set keeps its ancestor-expanded semantics for the delete
guard.
`Delete folder` and `Delete Model` rendered as plain menu text: the rule
was `color: var(--danger-color)`, and that token is defined nowhere in
the stylesheet tree. A var() reference to an undefined custom property is
invalid at computed-value time, so the declaration does not fall back to
a default — `color` inherits, and it silently matched the surrounding
menu text.
Points both the label and its icon (which inherits the colour) at
`--lora-error`, the themed error token the delete buttons already use.
The shared hover paints the accent background, which a red label does not
read against, so destructive entries also get their own `--lora-error-bg`
wash.
The same dead token was masked by a hardcoded fallback in the settings
priority-tags validation state; those now use the themed token too.
Fixes all seven destructive entries at once — the four model-card menus,
the exclude/duplicates `delete-all` entry and the folder sidebar menu.
Guard: tests/frontend/regression/contextMenuTokens.test.js fails if any
custom property used by menu.css stops resolving (verified by reverting
the token), and if var(--danger-color) ever comes back.
The folder context menu now reads: the content action (check for updates)
on top, then the folder operations as one group (new subfolder, rename),
then the destructive entry behind its own divider.
Gating the three folder entries per page could leave the menu with
dangling separators — the recipes sidebar hides all of them and keeps
only the update check, which already rendered one stray divider before
this change and would have rendered two after it. Add
_updateContextMenuSeparators: a divider survives only when a visible
entry sits on both sides, and a run of consecutive ones collapses to a
single line. The three per-item display toggles fold into one loop.
The template order is guarded by a regression test that parses
templates/components/context_menu.html, plus behaviour tests for the
divider collapsing.
Fills in the 35 `sidebar.*` placeholders the folder-sidebar feature series
left behind — view options (tree/list, empty folders), folder creation,
the delete confirmation modal and its result toasts, and folder
renaming, including the undo copy and the "deleting a folder never
cascades over model files" rule the backend enforces.
All nine locales are translated, so no `[TODO: Translate]` placeholder
remains anywhere: the state §7 of the translation guidelines describes
holds again. Terminology reuses the existing §2 maps (folder, model
root, sidebar) with tree/list view added, recorded in a new §2
subsection; the stale leaf-count in the header is refreshed too.
Follows the folder create/delete work: a typo'd directory could be
removed but not corrected, and for a folder holding models the only fix
was to move every model out by hand.
Adds POST /api/lm/{prefix}/rename-folder. Unlike the delete path this one
deliberately works on folders that hold models — a rename keeps every
file, so nothing is cascaded over: the directory is renamed on disk and
the scanner re-keys the records that pointed at the old prefix (recorded
folder list, cache file_path/folder/preview_url, hash and autov3 index
paths, excluded-model paths, and the metadata sidecars that travelled
with the directory). Ancestors are never touched, and only the leaf name
is accepted so a rename can never escape its parent.
Library roots, top-level symlinks and folders holding a staged delete are
refused; the last because a staging manifest records absolute
original/staged paths, so moving it would break undo and purge. A name
collision is a 409 target_exists conflict.
The sidebar reuses the inline-row idiom from folder creation: prefilled
with the current name, inserted in place of the node with that node
hidden while editing, Enter confirms and Escape/blur cancels. The
persisted selection and the expanded set are re-keyed across the rename
so the user keeps their place in the refreshed tree.
Folders created from the sidebar had no in-app way back out: the only
removal path was to leave ComfyUI, delete the directory by hand and
rescan. A typo'd folder also polluted the move/download destination
picker permanently, since it reads the same all_folders source.
Adds POST /api/lm/{prefix}/delete-folder, restricted to directories
whose subtree holds no model weight files — a folder-level cascade would
bypass the per-model lifecycle bookkeeping (metadata sidecars, previews,
cache entries, pending-delete staging, recipe references). The service
walks the directory itself instead of trusting the possibly stale cache,
reports what it would remove (models / files / subfolders / symlinks),
and refuses library roots, top-level symlinks (shutil.rmtree rejects
those) and folders holding a staged delete, whose manifest would be
invalidated by the move. Symbolic links inside the subtree are counted
but never followed.
ModelScanner.remove_known_folder mirrors add_known_folder: the removed
subtree leaves all_folders while ancestors are kept (every recorded
ancestor exists on disk in its own right), stale cache entries under the
prefix are purged and the folder list recomputed. The handler broadcasts
models_changed so destination pickers drop the folder too.
The sidebar entry is a destructive context-menu item. The modal opens in
a confirm state for model-free folders and an explanatory one when the
subtree still holds models, decided from the models-only set that already
dims empty nodes; a stale tree is caught by the 409 not_empty/busy
conflict. Truly empty folders get the existing 20s undo affordance,
implemented by re-creating the directory.
- Render the new-folder input as a temporary tree row at the creation
location (file-explorer style): full-width input confirmed with Enter
and canceled with Escape/blur; the parent folder auto-expands, and in
list mode the row is inserted after the parent item
- Remove the drag-to-blank-area folder creation (drop-zone strip,
sidebar-level drag handlers, performDragMoveWithState); dropping models
onto folder nodes still moves them
- Update empty-state hints and locale keys accordingly
Empty folders (tracked in the scan-recorded all_folders list, same source
the move/download destination picker uses) can now be surfaced in the
folder sidebar via a view-options toggle, dimmed when their subtree holds
no models. Folders can be created directly from the sidebar through a new
POST /api/lm/{prefix}/create-folder endpoint with library-root
containment checks; the scanner records the new directory incrementally
so the tree reflects it without a rescan.
The sidebar header moves its view toggles (tree/list, recursive, empty
folders) into a "..." menu to fit the new create-folder button.
A LoRA named `lora-sd1.5-backlight_slider_v10.safetensors` showed up in the
manager as `lora-sd1`, hid itself from searches for the rest of its name, and
collapsed into the same lora syntax tag as every sibling sharing the prefix.
The name was cut twice. `_process_model_file()` imports a third-party
`.civitai.info` sidecar by handing `from_civitai_info()` the local stem with
the extension already stripped, and the builder then stripped a second
"extension" from it -- `os.path.splitext` reads everything after the last dot
as one, so the version dot in `1.5` ended the name. The download path never
hit this because API filenames keep their extension and only need one strip.
Pass the real basename from the migration site, and make the builder strip
only a recognized model extension (`strip_model_extension`), so both input
shapes resolve to the same stem. The `model_name` fallback that reused the
same expression is fixed with it: on a sidecar without `model.name` the
display name was truncated too.
Libraries already corrupted do not heal on their own: the incremental Refresh
skips paths already in the cache (only a full rebuild reloads metadata) and
startup hydrates rows from SQLite as-is, so the wrong name survives restarts.
Reconcile now compares each cached row against the stem of its file path --
one string compare per file and no extra syscall, so a clean library pays
nothing -- and repairs mismatching rows through `load_metadata()` (which
normalizes the sidecar) and the existing in-place `_sync_cache_from_metadata_impl()`
path, which writes a targeted single-row SQL delta instead of a full save.
Repairs are one-shot, and a missing or corrupt sidecar keeps its row so a full
rebuild can recreate it without losing tags or civitai data.
Tests: the builder keeps dotted stems for all four model classes and still
strips real extensions; the migration writes the full local name to the
sidecar; and reconcile repairs memory, sidecar and SQLite row, runs exactly
once, and never reads metadata on a clean library.
Will Miao