fix(scanner): compare model roots by identity, not by spelling (#1146)

ComfyUI hands its model folders over in the host's own spelling --
`utils/extra_config.py` runs every configured path through `os.path.normpath()`,
which is all backslashes on Windows, and `main.py` registers `<output>/loras`
unconditionally for its save nodes. `Config` kept that spelling verbatim while
`_prepare_*_paths()` normalized the live root lists to forward slashes, and the
scanner compared the two as raw strings. On Windows every configured root
therefore looked like a root that was not in the live list: the Refresh menu
listed each directory twice (the twin showing "0 models"), the completion toast
claimed reachable roots were "currently inaccessible" and kept their whole cache
count as `kept_unreachable`, and the settings sync persisted host-registered
directories that were never created.

The duplication was not only cosmetic. `_UnreachablePaths.add()` normalizes the
prefix it is given, so the configured twin of a live root registered exactly the
live root's prefix as unreadable: in the prune phase every cached-but-missing
entry under it matched and was "kept", so models deleted from disk were never
removed from the cache again.

* `py/utils/root_paths.py` holds the single `normalize_root_identity()`
  (separator style, redundant/trailing separators, and case where the filesystem
  folds it: whole paths on Windows, drive letters everywhere). `config.py` and
  `settings_manager.py` each carried a private copy and `model_scanner.py`
  compared raw strings.
* `ModelScanner` compares root identities in `get_configured_model_roots()`,
  `describe_model_roots()` (dedupe + `available`), the reconcile's
  `configured_only` list and the `ReconcileScope.roots` membership test, so one
  directory is one root whatever the host called it. `_normalized_root_prefix()`
  is built on the same identity, and an empty prefix no longer matches every
  path.
* A configured root that never made it into the live list is reported only while
  it still holds cached entries -- in the menu and in `skipped_roots` /
  `unavailable_paths` alike. A directory that neither exists nor was ever scanned
  has nothing to keep, so `<output>/loras` stops being an unavailable root.
* The startup sync drops saved roots that neither exist nor hold cached entries
  (`PersistentModelCache.roots_with_cached_entries()`, one scan per model type,
  separator-agnostic and wildcard-escaped). A root named by the library's
  `default_*_root` is kept even when empty: dropping it would let the next boot
  "repair" the default onto another root. When the cache cannot be consulted,
  every root is kept.
* `GET /scan?roots=` validates requested roots by identity too, so the browser
  extension or a script passing the spelling settings.json holds selects the root
  instead of answering 400 "Unknown model root(s)".

Verified: 8 of the new tests fail against the pre-fix source (checked with the
fix stashed); 3798 passed, 7 skipped, 11 deselected. The probe costs 71ms for a
100k-model library with seven missing roots (357ms when probing root by root).
This commit is contained in:
Will Miao
2026-10-12 08:43:06 +08:00
parent e056142669
commit 6d0d7cb15b
11 changed files with 822 additions and 57 deletions
+104 -15
View File
@@ -4,7 +4,6 @@
# import cycles. Breaking them would require an architectural refactor.
import os
import platform
import posixpath
import threading
from pathlib import Path
import folder_paths # pyright: ignore[reportMissingImports]
@@ -20,6 +19,7 @@ from .utils.cache_paths import CacheType, get_cache_file_path, get_legacy_cache_
from .utils.constants import (
OTHER_MODEL_FOLDER_SUBTYPES,
)
from .utils.root_paths import normalize_root_identity
from .utils.settings_paths import (
ensure_settings_file,
get_settings_dir,
@@ -34,14 +34,15 @@ standalone_mode = (
logger = logging.getLogger(__name__)
def _normalize_root_identity(path: str) -> str:
"""Normalize a root path for comparisons across slash styles."""
normalized = posixpath.normpath(path.strip().replace("\\", "/"))
if len(normalized) >= 2 and normalized[1] == ":":
return normalized.lower()
return normalized
# Saved folder-path keys mapped to the model type whose cache can tell whether a
# root that is missing right now is still worth keeping (issue #1146). The
# checkpoint page holds both checkpoints and unet roots.
_CACHE_MODEL_TYPE_BY_FOLDER_KEY = {
"loras": "lora",
"checkpoints": "checkpoint",
"unet": "checkpoint",
"embeddings": "embedding",
}
def _append_new_paths(current: List[str], candidates: Iterable[str]) -> List[str]:
@@ -52,11 +53,11 @@ def _append_new_paths(current: List[str], candidates: Iterable[str]) -> List[str
to the root *order* would move user data.
"""
merged = list(current)
seen = {_normalize_root_identity(path) for path in merged}
seen = {normalize_root_identity(path) for path in merged}
for path in candidates:
if not isinstance(path, str) or not path.strip():
continue
identity = _normalize_root_identity(path)
identity = normalize_root_identity(path)
if identity in seen:
continue
seen.add(identity)
@@ -78,15 +79,15 @@ def _resolve_valid_default_root(
stripped = path.strip()
if not stripped:
continue
identity = _normalize_root_identity(stripped)
identity = normalize_root_identity(stripped)
if identity in seen:
continue
seen.add(identity)
fallback_paths.append(stripped)
allowed = {_normalize_root_identity(path) for path in fallback_paths}
allowed = {normalize_root_identity(path) for path in fallback_paths}
if current and _normalize_root_identity(current) in allowed:
if current and normalize_root_identity(current) in allowed:
return current
if not valid_paths:
@@ -320,6 +321,77 @@ class Config:
"Could not load extra paths from library settings: %s", exc
)
def _drop_missing_roots_without_cache(
self,
folder_paths_map: Dict[str, List[str]],
default_roots: Mapping[str, str],
) -> Dict[str, List[str]]:
"""Drop saved roots that neither exist nor hold cached entries (issue #1146).
The startup sync persists what is *configured* rather than what is readable,
so a switched-off drive keeps its path (see ``save_folder_paths_to_settings``).
A directory the host registered without ever creating it -- ComfyUI adds
``<output>/loras`` and friends for its save nodes whether or not they exist --
came along for the ride and stayed there as a permanently "unavailable" root.
A root that still holds cached entries is exactly the switched-off drive the
persistence exists for, so it is kept. A root named by the library's
``default_*_root`` is kept even when it is empty: dropping it would let the
next boot repair the default onto another root.
Best effort: when the cache cannot be consulted, every root is kept.
"""
try:
from .services.persistent_model_cache import get_persistent_cache
cache = get_persistent_cache()
except Exception as exc: # pragma: no cover - defensive logging
logger.debug("Keeping every configured model root: %s", exc)
return folder_paths_map
protected = {
identity
for identity in (
normalize_root_identity(path) for path in default_roots.values()
)
if identity
}
filtered: Dict[str, List[str]] = {}
for key, paths in folder_paths_map.items():
missing = [path for path in paths if not os.path.exists(path)]
if not missing:
filtered[key] = list(paths)
continue
probe = [
path
for path in missing
if normalize_root_identity(path) not in protected
]
try:
cached = cache.roots_with_cached_entries(
_CACHE_MODEL_TYPE_BY_FOLDER_KEY.get(key, "other"), probe
)
except Exception as exc:
logger.debug(
"Keeping every configured %s root (%s): %s", key, missing, exc
)
filtered[key] = list(paths)
continue
cached_ids = {normalize_root_identity(path) for path in cached}
filtered[key] = [
path
for path in paths
if os.path.exists(path)
or normalize_root_identity(path) in protected
or normalize_root_identity(path) in cached_ids
]
return filtered
def save_folder_paths_to_settings(self):
"""Persist ComfyUI-derived folder paths to the multi-library settings."""
try:
@@ -533,9 +605,26 @@ class Config:
or self._should_activate_comfy_library(libraries, libraries_changed)
)
# A root that the host registered but never created (and that holds no
# cached models) is not a library path: persisting it would bring it
# back as an unavailable root on every boot, and in standalone mode the
# saved paths are the only source of truth (issue #1146).
default_roots = {
"loras": default_lora_root
or comfy_library.get("default_lora_root", ""),
"checkpoints": default_checkpoint_root
or comfy_library.get("default_checkpoint_root", ""),
"unet": comfy_library.get("default_unet_root", ""),
"embeddings": default_embedding_root
or comfy_library.get("default_embedding_root", ""),
}
persisted_folder_paths = self._drop_missing_roots_without_cache(
target_folder_paths, default_roots
)
settings_service.upsert_library(
"comfyui",
folder_paths=target_folder_paths,
folder_paths=persisted_folder_paths,
extra_folder_paths=extra_folder_paths,
default_lora_root=default_lora_root,
default_checkpoint_root=default_checkpoint_root,