Support suffixed mask pairing

backend/app/modules/dataset/service.py

@@ -22,6 +22,7 @@ ARCHIVE_ALIASES = {
     "masks": {"mask", "masks", "label", "labels", "gt", "annotation", "annotations"},
     "labels": {"label", "labels", "txt", "annotation", "annotations"},
 }
+MASK_STEM_SUFFIXES = ("-mask", "_mask", "-label", "_label", "-gt", "_gt", "-seg", "_seg")
 
 
 def uploads_root() -> Path:
@@ -44,6 +45,16 @@ def safe_filename(value: str | None) -> str:
     return stem
 
 
+def upload_exts_for_kind(kind: str) -> set[str]:
+    if kind == "images":
+        return IMAGE_EXTS
+    if kind == "masks":
+        return IMAGE_EXTS | LABEL_EXTS
+    if kind == "labels":
+        return LABEL_EXTS
+    return set()
+
+
 def is_archive_name(value: str | None) -> bool:
     name = (value or "").lower()
     return any(name.endswith(suffix) for suffix in ARCHIVE_SUFFIXES)
@@ -88,9 +99,19 @@ def safe_archive_member(raw_name: str, kind: str) -> Path | None:
         return None
     safe_parts = [slugify(part) for part in raw_parts[:-1]]
     filename = safe_filename(raw_parts[-1])
+    if Path(filename).suffix.lower() not in upload_exts_for_kind(kind):
+        return None
     return Path(*safe_parts, filename) if safe_parts else Path(filename)
 
 
+def normalize_mask_stem(stem: str) -> str:
+    lower = stem.lower()
+    for suffix in MASK_STEM_SUFFIXES:
+        if lower.endswith(suffix) and len(stem) > len(suffix):
+            return stem[: -len(suffix)]
+    return stem
+
+
 def save_archive_member(target: Path, member_name: str, kind: str, source) -> dict | None:
     relative = safe_archive_member(member_name, kind)
     if relative is None:
@@ -147,12 +168,13 @@ def _iter_files(root: Path) -> Iterable[Path]:
     return sorted(path for path in root.rglob("*") if path.is_file())
 
 
-def _stem_map(paths: Iterable[Path], exts: set[str] | None = None) -> dict[str, Path]:
+def _stem_map(paths: Iterable[Path], exts: set[str] | None = None, normalize=None) -> dict[str, Path]:
     result: dict[str, Path] = {}
     for path in paths:
         if exts and path.suffix.lower() not in exts:
             continue
-        result.setdefault(path.stem, path)
+        stem = normalize(path.stem) if normalize else path.stem
+        result.setdefault(stem, path)
     return result
 
 
@@ -234,7 +256,7 @@ def validate_dataset(name: str) -> dict:
     mask_files = list(_iter_files(root / "masks"))
     images = _stem_map(image_files, IMAGE_EXTS)
     labels = _stem_map(label_files, LABEL_EXTS)
-    masks = _stem_map(mask_files, IMAGE_EXTS)
+    masks = _stem_map(mask_files, IMAGE_EXTS, normalize_mask_stem)
 
     image_stems = set(images)
     label_stems = set(labels)

backend/tests/test_dataset_service.py

@@ -31,7 +31,7 @@ def test_validate_dataset_and_generate_yolo_yaml(tmp_path, monkeypatch):
     image = np.zeros((32, 32, 3), dtype=np.uint8)
     mask = np.zeros((32, 32), dtype=np.uint8)
     cv2.imwrite(str(root / "images" / "sample.png"), image)
-    cv2.imwrite(str(root / "masks" / "sample.png"), mask)
+    cv2.imwrite(str(root / "masks" / "sample-mask.png"), mask)
     (root / "labels" / "sample.txt").write_text("0 0.25 0.25 0.75 0.25 0.75 0.75 0.25 0.75\n", encoding="utf-8")
 
     validation = validate_dataset("case_yolo")
@@ -57,6 +57,7 @@ def test_upload_zip_extracts_matching_dataset_kind(tmp_path, monkeypatch):
     image_zip = io.BytesIO()
     with zipfile.ZipFile(image_zip, "w") as archive:
         archive.writestr("bundle/images/sample.png", b"image")
+        archive.writestr("bundle/images/manifest.csv", b"name\nsample.png\n")
         archive.writestr("bundle/masks/sample.png", b"mask")
     image_zip.seek(0)
     image_upload = UploadFile(filename="images.zip", file=image_zip)
@@ -65,13 +66,13 @@ def test_upload_zip_extracts_matching_dataset_kind(tmp_path, monkeypatch):
     mask_zip = io.BytesIO()
     with zipfile.ZipFile(mask_zip, "w") as archive:
         archive.writestr("bundle/images/sample.png", b"image")
-        archive.writestr("bundle/masks/sample.png", b"mask")
+        archive.writestr("bundle/masks/sample-mask.png", b"mask")
     mask_zip.seek(0)
     mask_upload = UploadFile(filename="masks.zip", file=mask_zip)
     mask_result = asyncio.run(save_upload("case_zip", "masks", [mask_upload]))
 
     assert [item["relative_path"] for item in image_result["saved"]] == ["var/uploads/datasets/case_zip/images/sample.png"]
-    assert [item["relative_path"] for item in mask_result["saved"]] == ["var/uploads/datasets/case_zip/masks/sample.png"]
+    assert [item["relative_path"] for item in mask_result["saved"]] == ["var/uploads/datasets/case_zip/masks/sample-mask.png"]
     described = describe_dataset("case_zip")
     assert described["counts"]["images"] == 1
     assert described["counts"]["masks"] == 1