fix(example-images): support importing for Other models and pending hashes

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.

py/services/use_cases/example_images/import_example_images_use_case.py

@@ -29,6 +29,7 @@ class ImportExampleImagesUseCase:
 
     async def execute(self, request: web.Request) -> Dict[str, Any]:
         model_hash: str | None = None
+        model_path: str | None = None
         files_to_import: List[str] = []
         temp_files: List[str] = []
 
@@ -40,6 +41,8 @@ class ImportExampleImagesUseCase:
                 first_field = cast(BodyPartReader, first_field_raw) if first_field_raw is not None else None
                 if first_field and first_field.name == "model_hash":
                     model_hash = await first_field.text()
+                elif first_field and first_field.name == "model_path":
+                    model_path = await first_field.text()
                 else:
                     # Support clients that send files first and hash later
                     if first_field is not None:
@@ -49,13 +52,22 @@ class ImportExampleImagesUseCase:
                     field = cast(BodyPartReader, raw_field)
                     if field.name == "model_hash" and not model_hash:
                         model_hash = await field.text()
+                    elif field.name == "model_path" and not model_path:
+                        model_path = await field.text()
                     elif field.name == "files":
                         await self._collect_upload_file(field, files_to_import, temp_files)
             else:
                 data = await request.json()
                 model_hash = data.get("model_hash")
+                model_path = data.get("model_path")
                 files_to_import = list(data.get("file_paths", []))
 
+            # Models with a deferred hash (checkpoints, Other) send an empty
+            # model_hash; locate them by file path and compute the hash on
+            # demand, since example-image folders are keyed by hash.
+            if not model_hash and model_path:
+                model_hash = await self._processor.resolve_hash_for_file_path(model_path)
+
             if not model_hash:
                 raise ImportExampleImagesValidationError("Missing model_hash parameter")
             result = await self._processor.import_images(model_hash, files_to_import)

py/utils/example_images_processor.py

@@ -26,6 +26,42 @@ class ExampleImagesValidationError(ExampleImagesImportError):
 class ExampleImagesProcessor:
     """Processes and manipulates example images"""
 
+    @staticmethod
+    async def _model_scanners() -> list:
+        """Return every scanner whose models can carry example images."""
+        return [
+            await ServiceRegistry.get_lora_scanner(),
+            await ServiceRegistry.get_checkpoint_scanner(),
+            await ServiceRegistry.get_embedding_scanner(),
+            await ServiceRegistry.get_other_scanner(),
+        ]
+
+    @staticmethod
+    async def resolve_hash_for_file_path(file_path: str) -> str:
+        """Return the SHA256 for a cached model file, computing it on demand.
+
+        Checkpoint and Other scanners record ``hash_status="pending"`` with an
+        empty sha256 until something needs the hash; importing example images
+        is such a moment because example folders are keyed by hash.  Returns
+        ``''`` when the file is unknown to every scanner or hashing failed.
+        """
+        if not file_path:
+            return ''
+        normalized = file_path.replace(os.sep, '/')
+        for scanner in await ExampleImagesProcessor._model_scanners():
+            cache = await scanner.get_cached_data()
+            for item in cache.raw_data:
+                if item.get('file_path') != normalized:
+                    continue
+                sha256 = (item.get('sha256') or '').strip()
+                if sha256 and item.get('hash_status', 'completed') == 'completed':
+                    return sha256
+                calculate = getattr(scanner, 'calculate_hash_for_model', None)
+                if calculate is None:
+                    return sha256
+                return (await calculate(normalized)) or ''
+        return ''
+
     @staticmethod
     def generate_short_id(length=8):
         """Generate a short random alphanumeric identifier"""
@@ -450,15 +486,11 @@ class ExampleImagesProcessor:
                 raise ExampleImagesValidationError('No example images path configured')
 
             # Find the model and get current metadata
-            lora_scanner = await ServiceRegistry.get_lora_scanner()
-            checkpoint_scanner = await ServiceRegistry.get_checkpoint_scanner()
-            embedding_scanner = await ServiceRegistry.get_embedding_scanner()
-
             model_data = None
             scanner = None
 
-            # Check both scanners to find the model
+            # Check every scanner to find the model
-            for scan_obj in [lora_scanner, checkpoint_scanner, embedding_scanner]:
+            for scan_obj in await ExampleImagesProcessor._model_scanners():
                 cache = await scan_obj.get_cached_data()
                 for item in cache.raw_data:
                     if item.get('sha256') == model_hash:
@@ -536,6 +568,7 @@ class ExampleImagesProcessor:
                 'errors': errors,
                 'regular_images': regular_images,
                 'custom_images': custom_images,
+                'model_hash': model_hash,
                 "model_file_path": model_data.get('file_path', ''),
             }
 
@@ -577,15 +610,11 @@ class ExampleImagesProcessor:
                 }, status=400)
             
             # Find the model and get current metadata
-            lora_scanner = await ServiceRegistry.get_lora_scanner()
-            checkpoint_scanner = await ServiceRegistry.get_checkpoint_scanner()
-            embedding_scanner = await ServiceRegistry.get_embedding_scanner()
-            
             model_data = None
             scanner = None
 
-            # Check both scanners to find the model
+            # Check every scanner to find the model
-            for scan_obj in [lora_scanner, checkpoint_scanner, embedding_scanner]:
+            for scan_obj in await ExampleImagesProcessor._model_scanners():
                 if scan_obj.has_hash(model_hash):
                     cache = await scan_obj.get_cached_data()
                     for item in cache.raw_data:
@@ -737,14 +766,10 @@ class ExampleImagesProcessor:
                 )
 
         try:
-            lora_scanner = await ServiceRegistry.get_lora_scanner()
-            checkpoint_scanner = await ServiceRegistry.get_checkpoint_scanner()
-            embedding_scanner = await ServiceRegistry.get_embedding_scanner()
-
             model_data = None
             scanner = None
 
-            for scan_obj in [lora_scanner, checkpoint_scanner, embedding_scanner]:
+            for scan_obj in await ExampleImagesProcessor._model_scanners():
                 if scan_obj.has_hash(model_hash):
                     cache = await scan_obj.get_cached_data()
                     for item in cache.raw_data:

static/js/components/shared/showcase/ShowcaseView.js

@@ -1162,10 +1162,20 @@ async function handleImportFiles(files, modelHash, importContainer) {
         let successCount = 0;
         const errors = [];
 
+        // The showcase section carries the freshest hash (an import may have
+        // resolved a deferred hash) plus the file path the backend needs to
+        // locate models whose hash is still pending.
+        const showcaseSection = document.querySelector('.showcase-section');
+        const modelPath = showcaseSection?.dataset.filepath || '';
+        const currentHash = showcaseSection?.dataset.modelHash || modelHash;
+
         for (const file of validFiles) {
             try {
                 const formData = new FormData();
-                formData.append('model_hash', modelHash);
+                formData.append('model_hash', currentHash);
+                if (modelPath) {
+                    formData.append('model_path', modelPath);
+                }
                 formData.append('files', file);
 
                 const response = await fetch('/api/lm/import-example-images', {
@@ -1192,8 +1202,17 @@ async function handleImportFiles(files, modelHash, importContainer) {
 
         const result = lastSuccessResult;
 
+        // A model with a deferred hash (checkpoint / Other) is imported via
+        // model_path; the backend resolves and returns the real hash. Persist
+        // it so every follow-up (file list, NSFW toggle, delete, re-render)
+        // targets the hash-keyed example folder.
+        const effectiveHash = result.model_hash || currentHash;
+        if (showcaseSection && effectiveHash) {
+            showcaseSection.dataset.modelHash = effectiveHash;
+        }
+
         // Get updated local files
-        const updatedFilesResponse = await fetch(`/api/lm/example-image-files?model_hash=${modelHash}`);
+        const updatedFilesResponse = await fetch(`/api/lm/example-image-files?model_hash=${effectiveHash}`);
         const updatedFilesResult = await updatedFilesResponse.json();
 
         if (!updatedFilesResult.success) {
@@ -1221,7 +1240,7 @@ async function handleImportFiles(files, modelHash, importContainer) {
             }
 
             // Initialize the import UI for the new content
-            initExampleImport(modelHash, showcaseTab);
+            initExampleImport(effectiveHash, showcaseTab);
 
             if (errors.length > 0) {
                 showToast('toast.import.imagesPartial', { success: successCount, failed: errors.length }, 'warning');

tests/services/test_use_cases.py

@@ -146,6 +146,8 @@ class StubExampleImagesProcessor(ExampleImagesProcessor):
         self.calls: List[Dict[str, Any]] = []
         self.error: Optional[str] = None
         self.response: Dict[str, Any] = {"success": True}
+        self.resolved_hash: Optional[str] = None
+        self.resolve_calls: List[str] = []
 
     async def import_images(self, model_hash: str, files: List[str]) -> Dict[str, Any]:  # pyright: ignore[reportIncompatibleMethodOverride]
         self.calls.append({"model_hash": model_hash, "files": files})
@@ -155,6 +157,10 @@ class StubExampleImagesProcessor(ExampleImagesProcessor):
             raise ExampleImagesImportError("boom")
         return self.response
 
+    async def resolve_hash_for_file_path(self, file_path: str) -> str:
+        self.resolve_calls.append(file_path)
+        return self.resolved_hash or ""
+
 
 async def test_auto_organize_use_case_executes_with_lock() -> None:
     file_service = StubFileService()
@@ -506,6 +512,34 @@ async def test_import_example_images_use_case_propagates_generic_error() -> None
         await use_case.execute(request)  # pyright: ignore[reportArgumentType]
 
 
+async def test_import_example_images_use_case_resolves_hash_from_model_path() -> None:
+    """Models with a deferred hash (checkpoints, Other) send model_path
+    instead of model_hash; the use case must resolve the hash on demand."""
+    processor = StubExampleImagesProcessor()
+    processor.resolved_hash = "f" * 64
+    use_case = ImportExampleImagesUseCase(processor=processor)
+
+    request = DummyJsonRequest(
+        {"model_path": "/models/vae/x.safetensors", "file_paths": ["/tmp/file"]}
+    )
+    result = await use_case.execute(request)  # pyright: ignore[reportArgumentType]
+
+    assert processor.resolve_calls == ["/models/vae/x.safetensors"]
+    assert processor.calls == [{"model_hash": "f" * 64, "files": ["/tmp/file"]}]
+    assert result == {"success": True}
+
+
+async def test_import_example_images_use_case_rejects_unresolvable_model_path() -> None:
+    processor = StubExampleImagesProcessor()
+    use_case = ImportExampleImagesUseCase(processor=processor)
+    request = DummyJsonRequest(
+        {"model_path": "/models/unknown.safetensors", "file_paths": []}
+    )
+
+    with pytest.raises(ImportExampleImagesValidationError):
+        await use_case.execute(request)  # pyright: ignore[reportArgumentType]
+
+
 class StubLifecycleService:
     def __init__(self, scanner: Optional[MockScanner] = None) -> None:
         self.renames: List[Dict[str, str]] = []

tests/utils/test_example_images_processor_unit.py

@@ -183,6 +183,7 @@ def stub_scanners(monkeypatch: pytest.MonkeyPatch, tmp_path) -> StubScanner:
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_lora_scanner", classmethod(_return_scanner))
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_checkpoint_scanner", classmethod(_return_scanner))
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_embedding_scanner", classmethod(_return_scanner))
+    monkeypatch.setattr(processor_module.ServiceRegistry, "get_other_scanner", classmethod(_return_scanner))
 
     return scanner
 
@@ -240,6 +241,7 @@ async def test_import_images_raises_when_model_not_found(monkeypatch: pytest.Mon
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_lora_scanner", classmethod(_empty_scanner))
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_checkpoint_scanner", classmethod(_empty_scanner))
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_embedding_scanner", classmethod(_empty_scanner))
+    monkeypatch.setattr(processor_module.ServiceRegistry, "get_other_scanner", classmethod(_empty_scanner))
 
     with pytest.raises(processor_module.ExampleImagesImportError):
         await processor_module.ExampleImagesProcessor.import_images("a" * 64, [str(tmp_path / "missing.png")])
@@ -288,6 +290,7 @@ async def test_delete_custom_image_preserves_existing_metadata(monkeypatch: pyte
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_lora_scanner", classmethod(_return_scanner))
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_checkpoint_scanner", classmethod(_return_scanner))
     monkeypatch.setattr(processor_module.ServiceRegistry, "get_embedding_scanner", classmethod(_return_scanner))
+    monkeypatch.setattr(processor_module.ServiceRegistry, "get_other_scanner", classmethod(_return_scanner))
 
     model_folder = get_model_folder(model_hash)
     os.makedirs(model_folder, exist_ok=True)
@@ -329,3 +332,135 @@ async def test_delete_custom_image_preserves_existing_metadata(monkeypatch: pyte
     _, _, updated_metadata = scanner.updated[-1]
     assert updated_metadata["civitai"]["images"] == existing_metadata["civitai"]["images"]
     assert updated_metadata["civitai"]["customImages"] == []
+
+
+def _patch_scanner_getters(monkeypatch: pytest.MonkeyPatch, **scanners) -> None:
+    """Point every model-scanner getter at the given stub (empty by default)."""
+
+    async def _empty(cls=None):
+        return StubScanner([])
+
+    for name in (
+        "get_lora_scanner",
+        "get_checkpoint_scanner",
+        "get_embedding_scanner",
+        "get_other_scanner",
+    ):
+        stub = scanners.get(name)
+        if stub is None:
+            getter = _empty
+        else:
+            async def getter(cls=None, _stub=stub):  # noqa: B023 - bound per iteration
+                return _stub
+        monkeypatch.setattr(
+            processor_module.ServiceRegistry, name, classmethod(getter)
+        )
+
+
+@pytest.mark.asyncio
+async def test_import_images_finds_model_in_other_scanner(
+    monkeypatch: pytest.MonkeyPatch, tmp_path
+) -> None:
+    """Other-category models (VAEs, text encoders) live in the Other scanner;
+    importing example images must search it too."""
+    settings_manager = get_settings_manager()
+    settings_manager.settings["example_images_path"] = str(tmp_path / "examples")
+    settings_manager.settings["libraries"] = {"default": {}}
+    settings_manager.settings["active_library"] = "default"
+
+    model_hash = "b" * 64
+    model_data = {
+        "sha256": model_hash,
+        "model_name": "VAE",
+        "file_path": str(tmp_path / "vae.safetensors"),
+        "civitai": {},
+    }
+    other_scanner = StubScanner([model_data])
+    _patch_scanner_getters(monkeypatch, get_other_scanner=other_scanner)
+
+    source_file = tmp_path / "upload.png"
+    source_file.write_bytes(b"PNG data")
+    monkeypatch.setattr(
+        processor_module.ExampleImagesProcessor,
+        "generate_short_id",
+        staticmethod(lambda: "short"),
+    )
+
+    recorded: Dict[str, Any] = {}
+
+    async def fake_update_metadata(model_hash, model_data, scanner, paths):
+        recorded["scanner"] = scanner
+        return [], []
+
+    monkeypatch.setattr(
+        processor_module.MetadataUpdater,
+        "update_metadata_after_import",
+        staticmethod(fake_update_metadata),
+    )
+
+    result = await processor_module.ExampleImagesProcessor.import_images(
+        model_hash, [str(source_file)]
+    )
+
+    assert result["success"] is True
+    assert result["model_hash"] == model_hash
+    assert recorded["scanner"] is other_scanner
+
+
+@pytest.mark.asyncio
+async def test_resolve_hash_for_file_path_computes_pending_hash(
+    monkeypatch: pytest.MonkeyPatch, tmp_path
+) -> None:
+    """A model whose hash is still pending gets it computed on demand."""
+    model_path = str(tmp_path / "encoder.safetensors").replace(os.sep, "/")
+    item = {"file_path": model_path, "sha256": "", "hash_status": "pending"}
+
+    calculated: list[str] = []
+
+    class PendingScanner(StubScanner):
+        async def calculate_hash_for_model(self, file_path: str):
+            calculated.append(file_path)
+            return "f" * 64
+
+    _patch_scanner_getters(monkeypatch, get_other_scanner=PendingScanner([item]))
+
+    result = await processor_module.ExampleImagesProcessor.resolve_hash_for_file_path(
+        model_path
+    )
+
+    assert result == "f" * 64
+    assert calculated == [model_path]
+
+
+@pytest.mark.asyncio
+async def test_resolve_hash_for_file_path_returns_completed_hash_without_recompute(
+    monkeypatch: pytest.MonkeyPatch, tmp_path
+) -> None:
+    model_path = str(tmp_path / "model.safetensors").replace(os.sep, "/")
+    item = {"file_path": model_path, "sha256": "c" * 64, "hash_status": "completed"}
+
+    class EagerScanner(StubScanner):
+        async def calculate_hash_for_model(self, file_path: str):  # pragma: no cover
+            raise AssertionError("must not recompute a completed hash")
+
+    _patch_scanner_getters(monkeypatch, get_lora_scanner=EagerScanner([item]))
+
+    result = await processor_module.ExampleImagesProcessor.resolve_hash_for_file_path(
+        model_path
+    )
+
+    assert result == "c" * 64
+
+
+@pytest.mark.asyncio
+async def test_resolve_hash_for_file_path_unknown_file_returns_empty(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    _patch_scanner_getters(monkeypatch)
+
+    assert (
+        await processor_module.ExampleImagesProcessor.resolve_hash_for_file_path(
+            "/models/nope.safetensors"
+        )
+        == ""
+    )