napari-ome-arrow 0.0.4__py3-none-any.whl → 0.0.5__py3-none-any.whl

napari_ome_arrow/_reader_infer.py

@@ -0,0 +1,233 @@
+"""Inference helpers for determining reader mode and source support."""
+
+from __future__ import annotations
+
+import logging
+from pathlib import Path
+from typing import Any
+
+from ome_arrow.core import OMEArrow
+from ome_arrow.ingest import from_stack_pattern_path
+
+from ._reader_omearrow import _read_vortex_scalar
+
+LOGGER = logging.getLogger(__name__)
+
+
+def _normalize_image_type(value: Any) -> str | None:
+    """Normalize an image type hint to "image" or "labels".
+
+    Args:
+        value: Raw hint value from metadata or user input.
+
+    Returns:
+        "image", "labels", or None if it cannot be inferred.
+    """
+    if value is None:
+        return None
+    # Normalize arbitrary input to a comparable string token.
+    text = str(value).strip().lower()
+    if not text:
+        return None
+    if text in {"image", "intensity", "raw"} or "image" in text:
+        return "image"
+    if (
+        text
+        in {
+            "labels",
+            "label",
+            "mask",
+            "masks",
+            "segmentation",
+            "seg",
+            "outlines",
+            "outline",
+        }
+        or "label" in text
+        or "mask" in text
+        or "seg" in text
+        or "outline" in text
+    ):
+        return "labels"
+    return None
+
+
+def _infer_layer_mode_from_record(record: dict[str, Any]) -> str | None:
+    """Infer the layer mode from a parsed OME-Arrow record.
+
+    Args:
+        record: Parsed metadata record from OME-Arrow.
+
+    Returns:
+        "image", "labels", or None if no hint is present.
+    """
+    # Collect candidate metadata fields that may encode image type.
+    candidates = []
+    if "image_type" in record:
+        candidates.append(record.get("image_type"))
+    pixels_meta = record.get("pixels_meta")
+    if isinstance(pixels_meta, dict) and "image_type" in pixels_meta:
+        candidates.append(pixels_meta.get("image_type"))
+    for value in candidates:
+        inferred = _normalize_image_type(value)
+        if inferred:
+            return inferred
+    return None
+
+
+def _infer_layer_mode_from_ome_arrow(obj: OMEArrow) -> str | None:
+    """Infer the layer mode from an OMEArrow object.
+
+    Args:
+        obj: OMEArrow instance to inspect.
+
+    Returns:
+        "image", "labels", or None if no hint is present.
+    """
+    try:
+        record = obj.data.as_py()
+    except Exception:
+        return None
+    if not isinstance(record, dict):
+        return None
+    return _infer_layer_mode_from_record(record)
+
+
+def _infer_layer_mode_from_source(
+    src: str, *, stack_default_dim: str | None = None
+) -> str | None:
+    """Infer the layer mode by inspecting a data source path.
+
+    Args:
+        src: Source path or stack pattern.
+        stack_default_dim: Default dimension to use for stack patterns.
+
+    Returns:
+        "image", "labels", or None if inference fails.
+    """
+    # Quick path sniffing before attempting to instantiate OMEArrow.
+    s = src.lower()
+    p = Path(src)
+    looks_stack = any(c in src for c in "<>*")
+    looks_zarr = (
+        s.endswith((".ome.zarr", ".zarr"))
+        or ".zarr/" in s
+        or p.exists()
+        and p.is_dir()
+        and p.suffix.lower() == ".zarr"
+    )
+    looks_parquet = s.endswith(
+        (".ome.parquet", ".parquet", ".pq")
+    ) or p.suffix.lower() in {
+        ".parquet",
+        ".pq",
+    }
+    looks_vortex = s.endswith(
+        (".ome.vortex", ".vortex")
+    ) or p.suffix.lower() in {
+        ".vortex",
+    }
+    looks_tiff = s.endswith(
+        (".ome.tif", ".ome.tiff", ".tif", ".tiff")
+    ) or p.suffix.lower() in {
+        ".tif",
+        ".tiff",
+    }
+    if not (
+        looks_stack
+        or looks_zarr
+        or looks_parquet
+        or looks_tiff
+        or looks_vortex
+    ):
+        LOGGER.debug(
+            "Skipping layer-mode inference for non-OME source: %s", src
+        )
+        return None
+
+    scalar = None
+    try:
+        if looks_vortex:
+            scalar = _read_vortex_scalar(src)
+        if looks_stack and stack_default_dim is not None:
+            scalar = from_stack_pattern_path(
+                src,
+                default_dim_for_unspecified=stack_default_dim,
+                map_series_to="T",
+                clamp_to_uint16=True,
+            )
+        obj = OMEArrow(scalar if scalar is not None else src)
+    except Exception:
+        return None
+    return _infer_layer_mode_from_ome_arrow(obj)
+
+
+def _looks_like_ome_source(path_str: str) -> bool:
+    """Check whether a path appears to be an OME-Arrow supported source.
+
+    Args:
+        path_str: Path or stack pattern string.
+
+    Returns:
+        True if the path looks like a supported source, otherwise False.
+    """
+    s = path_str.strip().lower()
+    p = Path(path_str)
+
+    # Bio-Formats-style stack patterns.
+    looks_stack = any(c in path_str for c in "<>*")
+    looks_zarr = (
+        s.endswith((".ome.zarr", ".zarr"))
+        or ".zarr/" in s
+        or p.exists()
+        and p.is_dir()
+        and p.suffix.lower() == ".zarr"
+    )
+    looks_parquet = s.endswith(
+        (".ome.parquet", ".parquet", ".pq")
+    ) or p.suffix.lower() in {
+        ".parquet",
+        ".pq",
+    }
+    looks_vortex = s.endswith(
+        (".ome.vortex", ".vortex")
+    ) or p.suffix.lower() in {
+        ".vortex",
+    }
+    looks_tiff = s.endswith(
+        (".ome.tif", ".ome.tiff", ".tif", ".tiff")
+    ) or p.suffix.lower() in {
+        ".tif",
+        ".tiff",
+    }
+    looks_npy = s.endswith(".npy")
+    looks_dir_stack = False
+    if p.exists() and p.is_dir() and not looks_zarr:
+        try:
+            # Heuristic: directory with multiple image-like files.
+            stack_files = [
+                f
+                for f in p.iterdir()
+                if f.is_file()
+                and f.name.lower().endswith(
+                    (
+                        ".ome.tif",
+                        ".ome.tiff",
+                        ".tif",
+                        ".tiff",
+                        ".npy",
+                    )
+                )
+            ]
+            looks_dir_stack = len(stack_files) > 1
+        except OSError:
+            looks_dir_stack = False
+    return (
+        looks_stack
+        or looks_zarr
+        or looks_parquet
+        or looks_vortex
+        or looks_tiff
+        or looks_npy
+        or looks_dir_stack
+    )

napari_ome_arrow/_reader_napari.py

@@ -0,0 +1,107 @@
+"""Napari viewer helpers for the OME-Arrow reader."""
+
+from __future__ import annotations
+
+import math
+from typing import Any
+
+import numpy as np
+
+
+def _maybe_set_viewer_3d(arr: np.ndarray) -> None:
+    """Switch the active napari viewer to 3D if the array has a Z-stack.
+
+    Assumes OME-Arrow's TCZYX convention or a subset, i.e. Z is always the
+    third-from-last axis. No-op if there's no active viewer.
+
+    Args:
+        arr: Array to inspect for a Z dimension.
+    """
+    # Need at least (Z, Y, X)
+    if arr.ndim < 3:
+        return
+
+    z_size = arr.shape[-3]
+    if z_size <= 1:
+        return
+
+    try:
+        import napari
+
+        viewer = napari.current_viewer()
+    except Exception:
+        # no viewer / not in GUI context → silently skip
+        return
+
+    if viewer is not None:
+        viewer.dims.ndisplay = 3
+
+
+def _as_labels(arr: np.ndarray) -> np.ndarray:
+    """Convert an array into an integer label array.
+
+    Args:
+        arr: Input array.
+
+    Returns:
+        Array converted to an integer dtype suitable for labels.
+    """
+    if arr.dtype.kind == "f":
+        arr = np.nan_to_num(arr, nan=0.0, posinf=0.0, neginf=0.0)
+        arr = np.round(arr).astype(np.int32, copy=False)
+    elif arr.dtype.kind not in ("i", "u"):
+        arr = arr.astype(np.int32, copy=False)
+    return arr
+
+
+def _enable_grid(n_layers: int) -> None:
+    """Switch the current viewer into grid view when possible.
+
+    Args:
+        n_layers: Number of layers to arrange in the grid.
+    """
+    if n_layers <= 1:
+        return
+    try:
+        import napari
+
+        viewer = napari.current_viewer()
+    except Exception:
+        return
+    if viewer is None:
+        return
+
+    cols = math.ceil(math.sqrt(n_layers))
+    rows = math.ceil(n_layers / cols)
+    try:
+        viewer.grid.enabled = True
+        viewer.grid.shape = (rows, cols)
+    except Exception:
+        # grid is best-effort; ignore if unavailable
+        return
+
+
+def _strip_channel_axis(
+    arr: np.ndarray, add_kwargs: dict[str, Any]
+) -> tuple[np.ndarray, dict[str, Any]]:
+    """Strip the channel axis from an array if configured.
+
+    Args:
+        arr: Input array.
+        add_kwargs: Layer kwargs that may include "channel_axis".
+
+    Returns:
+        Tuple of (array without channel axis, updated kwargs).
+    """
+    # Only strip when channel_axis is present and valid.
+    channel_axis = add_kwargs.pop("channel_axis", None)
+    if channel_axis is None:
+        return arr, add_kwargs
+    try:
+        axis = int(channel_axis)
+    except Exception:
+        return arr, add_kwargs
+    if arr.ndim <= axis:
+        return arr, add_kwargs
+    arr = np.take(arr, 0, axis=axis)
+    return arr, add_kwargs