- Add deleted_at column to AssetReference model and migration
- soft_delete_reference_by_id sets deleted_at instead of removing rows
- DELETE /api/assets/{id} defaults to soft-delete; delete_content=true
for hard-delete
- Add deleted_at IS NULL filters to read queries, tag queries, and
scanner queries so soft-deleted refs are invisible
- restore_references_by_paths skips soft-deleted refs
- upsert_reference clears deleted_at on explicit re-ingest
- Add tests for soft-delete API behavior, scanner persistence, bulk
insert, enrichment exclusion, and seed asset garbage collection
Amp-Thread-ID: https://ampcode.com/threads/T-019cb6fc-c05c-761f-b855-6d5d1c9defa2
Co-authored-by: Amp <amp@ampcode.com>
The previous commit acquired the exclusive lock before Alembic migrations,
but Alembic opens its own connection — which was then blocked by our lock.
Move lock acquisition to after migrations complete in a dedicated connection.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Content-Disposition: drop raw filename= parameter, use only RFC 5987
filename*=UTF-8'' to prevent header injection via ; and special chars
- delete_asset: default delete_content to False (non-destructive) when
query parameter is omitted
- create_asset_from_hash: return 400 MISSING_INPUT instead of 404 when
hash not found and no file uploaded (client input error, not missing resource)
- seeder: clear _progress when returning to IDLE so get_status() does not
return stale progress after scan completion
- hashing: handle non-seekable streams in _hash_file_obj by checking
seekable() before attempting tell/seek
- bulk_ingest: filter lost_paths to only include paths tied to actually
inserted asset IDs, preventing inflated counts from ON CONFLICT drops
Amp-Thread-ID: https://ampcode.com/threads/T-019cb67a-9822-7438-ab05-d09991a9f7f3
Co-authored-by: Amp <amp@ampcode.com>
When two ComfyUI processes share the same database file but point to
different input/output/model directories, each process's scan marks
the other's assets as missing, causing unreliable asset visibility.
This adds an exclusive lock so the second process fails fast at startup
with a clear message to use --database-url.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Extract validate_blake3_hash() into helpers.py, used by upload, schemas, routes
- Extract get_reference_with_owner_check() into queries, used by 4 service functions
- Extract build_prefix_like_conditions() into queries/common.py, used by 3 queries
- Replace 3 inlined tag queries with get_reference_tags() calls
- Consolidate AddTagsDict/RemoveTagsDict TypedDicts into AddTagsResult/RemoveTagsResult
dataclasses, eliminating manual field copying in tagging.py
- Make iter_row_chunks delegate to iter_chunks
- Inline trivial compute_filename_for_reference wrapper (unused session param)
- Remove mark_assets_missing_outside_prefixes pass-through in bulk_ingest.py
- Clean up unused imports (os, time, dependencies_available)
- Disable assets routes on DB init failure in main.py
Amp-Thread-ID: https://ampcode.com/threads/T-019cb649-dd4e-71ff-9a0e-ae517365207b
Co-authored-by: Amp <amp@ampcode.com>
- Fix missing import for compute_filename_for_reference in ingest.py
- Apply code review fixes across routes, queries, scanner, seeder,
hashing, ingest, path_utils, main, and server
- Update and add tests for sync references and seeder
Amp-Thread-ID: https://ampcode.com/threads/T-019cb61a-ed54-738c-a05f-9b5242e513f3
Co-authored-by: Amp <amp@ampcode.com>
Replace --disable-assets-autoscan with --enable-assets so the assets
system (API routes, database sync, background scanning) is off by
default and must be explicitly opted into. Expose the flag as an
"assets" entry in SERVER_FEATURE_FLAGS so the frontend can read it
from GET /features.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Filter hidden files/directories (dot-prefixed) in collect_models_files()
using is_visible(), matching the existing behavior for input/output roots
- Exclude the 'custom_nodes' folder name from get_comfy_models_folders();
custom nodes that register their own paths under other folder names
will still be scanned as expected
Amp-Thread-ID: https://ampcode.com/threads/T-019c924b-591a-725e-b8b7-0d49ba1a5591
Co-authored-by: Amp <amp@ampcode.com>
commonpath raises ValueError on Windows when comparing paths on different
drives (e.g. C:\models vs D:\extra_models). Replace all usages in the
asset scanner with Path.is_relative_to() which handles cross-drive paths,
case-insensitivity, and prefix traps natively without try/except.
Amp-Thread-ID: https://ampcode.com/threads/T-019c9224-d83c-7797-8c02-e1e1ae2ee452
Co-authored-by: Amp <amp@ampcode.com>
get_comfy_models_folders() previously filtered by startswith(models_root),
excluding extra model paths outside the main models directory. Now includes
every category with non-empty paths from folder_names_and_paths.
Amp-Thread-ID: https://ampcode.com/threads/T-019c9224-d83c-7797-8c02-e1e1ae2ee452
Co-authored-by: Amp <amp@ampcode.com>
list_files_recursively now uses followlinks=True so symlinked
directories under input/ and output/ roots are traversed, matching
the existing behavior of folder_paths.recursive_search for models.
Tracks (st_dev, st_ino) pairs of visited directories to detect and
break circular symlink loops safely.
Amp-Thread-ID: https://ampcode.com/threads/T-019c9220-21b8-7678-b428-9215ff1bb011
Co-authored-by: Amp <amp@ampcode.com>
* model_management: Remove non-comfy dynamic _v caster
* Force pre-load non-comfy weights to GPU in ModelPatcherDynamic
Non-comfy weights may expect to be pre-cast to the target
device without in-model casting. Previously they were allocated in
the vbar with _v which required the _v fault path in cast_to.
Instead, back up the original CPU weight and move it directly to GPU
at load time.
* draft zeta (z-image pixel space)
* revert gitignore
* model loaded and able to run however vector direction still wrong tho
* flip the vector direction to original again this time
* Move wrongly positioned Z image pixel space class
* inherit Radiance LatentFormat class
* Fix parameters in classes for Zeta x0 dino
* remove arbitrary nn.init instances
* Remove unused import of lru_cache
---------
Co-authored-by: silveroxides <ishimarukaito@gmail.com>
Comfy Aimdo 0.2.4 fixes a VRAM buffer alignment issue that happens in
someworkflows where action is able to bypass the pytorch allocator
and go straight to the cuda hook.
This was previously considering the pool of dynamic models as one giant
entity for the sake of smart memory, but that isnt really the useful
or what a user would reasonably expect. Make Dynamic VRAM properly purge
its models just like the old --disable-smart-memory but conditioning
the dynamic-for-dynamic bypass on smart memory.
Re-enable dynamic smart memory.
Multi-step samplers (eg. dpmpp_2s_ancestral) call the model at intermediate sigma values not present in the schedule. This caused set_step to crash with "No sample_sigmas matched current timestep" when context windows were enabled.
The fix is to keep self._step from the last exact match when a substep sigma is encountered, since substeps are still logically part of their parent step and should use the same context windows.
Co-authored-by: ozbayb <17261091+ozbayb@users.noreply.github.com>
* sd: add support for clip model reconstruction
* nodes: SetClipHooks: Demote the dynamic model patcher
* mp: Make dynamic_disable more robust
The backup need to not be cloned. In addition add a delegate object
to ModelPatcherDynamic so that non-cloning code can do
ModelPatcherDynamic demotion
* sampler_helpers: Demote to non-dynamic model patcher when hooking
* code rabbit review comments
Allow non QuantizedTensor layer to set want_requant to get the post lora
calculation stochastic cast down to the original input dtype.
This is then used by the legacy fp8 Linear implementation to set the
compute_dtype to the preferred lora dtype but then want_requant it back
down to fp8.
This fixes the issue with --fast fp8_matrix_mult is combined with
--fast dynamic_vram which doing a lora on an fp8_ non QT model.
