feat(links): support ModelScope and TensorArt as model sources

A model file could only ever be linked to huggingface.co: `set_hf_url`
validated the URL with a huggingface-only regex, the agent fetched the card
from a hardcoded HF URL, and the readme processor built every relative image
path off `https://huggingface.co/{repo}/resolve/main`. ModelScope publishes the
same model-card convention (README.md + YAML frontmatter, often carrying
`base_model:` and `trigger_words:`) behind a public, key-less API, so the
enrichment pipeline could already serve it - it was the plumbing that was
HF-shaped, not the idea.

Make the external source a first-class, provider-driven concept:

- New `py/services/model_sources/` registry. A `ModelSource` owns URL
  recognition (lenient for stored values, strict for user input), the
  canonical page URL, model-card fetching, the asset base URL and the
  capability flags. `HuggingFaceSource` is the previous logic relocated;
  `ModelScopeSource` reads `/models/{o}/{n}/resolve/{master|main}/README.md`
  and falls back to `/api/v1/models/{o}/{n}/repo`. `TensorArtSource` is
  link-only on purpose: tensor.art answers plain HTTP clients with a
  Cloudflare challenge and its internal API (ap-east-1.tensorart.cloud /
  cn.tensorart.net) rejects every /v1/model/* route with "invalid
  authorization header", so it declares supports_enrichment=False rather than
  failing silently later.
- Metadata gains `source_platform` + `source_url`; `hf_url` stays as a
  read/write alias, written only for Hugging Face, so existing sidecars,
  cached rows and third-party consumers keep working. Normalisation runs at
  the scanner, the persistent cache (both directions, plus two new columns
  behind an ALTER migration) and the linking handler - which is what stops a
  user who switches sources from leaving a stale `hf_url` on a ModelScope
  model.
- The agent pipeline keys off the provider instead of `hf_url`: the fast-fail
  gate now explains *why* a model is skipped (no source / unknown source /
  source without a reachable card), the prompt context exposes
  source_url/source_id/source_label/asset_base_url while still filling the
  legacy hf_url/repo aliases, and the four README image extractors take a
  base_url (defaulting to HF) so relative paths resolve against the right
  site. Version grouping generalises to hf: / ms: / ta: keys.
- `POST /api/lm/set-hf-url` keeps its path and its legacy payload keys but
  accepts `source_url`, validates against every provider and returns the
  platform. `GET /api/lm/model-sources` lets the UI render the supported-site
  list from the server.
- Frontend: a `modelSourceHelpers` mirror of the registry drives the link
  dialog, the card/modal globe (branded "View on ModelScope/TensorArt"), the
  version-group key and the enrichment gate; the versions tab no longer sends
  ms:/ta: keys to the CivitAI API.

TensorArt stays in the list because provenance is worth keeping even when the
card is unreadable - the dialog says so plainly ("Sites that don't expose one
(currently TensorArt) can only be linked") and the context menu disables
enrichment with a matching tooltip, instead of the user getting
"Unsupported URL".

Verified against the real ModelScope API: jj3550945163/Krea-2-LORA returns a
1882-byte card whose frontmatter carries base_model/tags/trigger_words, and
relative images resolve to .../resolve/master/....

Tests: backend 2815 passed; frontend 1130 JS + 91 Vue passed; pytest
tests/i18n and a Jinja compile pass over templates/. The nine locales carry
[TODO: Translate] for the new strings, completed in the next commit.
This commit is contained in:
Will Miao
2026-09-14 07:24:08 +08:00
parent 84146b62fd
commit 5ab0e88abc
51 changed files with 2518 additions and 279 deletions
+17
View File
@@ -7,6 +7,7 @@ from dataclasses import dataclass, field
from typing import Any, Dict, List, Mapping, Optional, Sequence, Tuple
from ..utils.cache_paths import CacheType, resolve_cache_path_with_migration
from .model_sources import normalize_metadata_source
logger = logging.getLogger(__name__)
@@ -62,6 +63,8 @@ class PersistentModelCache:
"db_checked",
"last_checked_at",
"hash_status",
"source_platform",
"source_url",
"hf_url",
)
_MODEL_UPDATE_COLUMNS: Tuple[str, ...] = _MODEL_COLUMNS[2:]
@@ -206,8 +209,13 @@ class PersistentModelCache:
"skip_metadata_refresh": bool(row["skip_metadata_refresh"]),
"license_flags": int(license_value),
"hash_status": row["hash_status"] or "completed",
"source_platform": row["source_platform"] or "",
"source_url": row["source_url"] or "",
"hf_url": row["hf_url"] or "",
}
# Legacy rows only carry `hf_url`; derive the canonical pair so
# every consumer sees the same shape.
normalize_metadata_source(item)
if row["autov3"] is not None:
item["autov3"] = (row["autov3"] or "").lower()
raw_data.append(item)
@@ -562,6 +570,8 @@ class PersistentModelCache:
db_checked INTEGER,
last_checked_at REAL,
hash_status TEXT,
source_platform TEXT DEFAULT '',
source_url TEXT DEFAULT '',
hf_url TEXT DEFAULT '',
PRIMARY KEY (model_type, file_path)
);
@@ -629,6 +639,8 @@ class PersistentModelCache:
# Persisting without explicit flags should assume CivitAI's documented defaults (0b111001 == 57).
"license_flags": f"INTEGER DEFAULT {DEFAULT_LICENSE_FLAGS}",
"hash_status": "TEXT DEFAULT 'completed'",
"source_platform": "TEXT DEFAULT ''",
"source_url": "TEXT DEFAULT ''",
"hf_url": "TEXT DEFAULT ''",
"autov3": "TEXT",
}
@@ -650,6 +662,9 @@ class PersistentModelCache:
return conn
def _prepare_model_row(self, model_type: str, item: Dict[str, Any]) -> Tuple[Any, ...]:
# Keep `source_*` and the legacy `hf_url` alias consistent no matter
# which caller populated the item.
normalize_metadata_source(item)
civitai = item.get("civitai") or {}
trained_words = civitai.get("trainedWords")
if isinstance(trained_words, str):
@@ -713,6 +728,8 @@ class PersistentModelCache:
1 if item.get("db_checked") else 0,
float(item.get("last_checked_at") or 0.0),
item.get("hash_status", "completed"),
item.get("source_platform") or "",
item.get("source_url") or "",
item.get("hf_url") or "",
)