fitscube 2.2.0__tar.gz → 2.3.0__tar.gz

{fitscube-2.2.0 → fitscube-2.3.0}/.github/workflows/cd.yml

@@ -25,7 +25,7 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           fetch-depth: 0
 
@@ -43,7 +43,7 @@ jobs:
     if: github.event_name == 'release' && github.event.action == 'published'
 
     steps:
-      - uses: actions/download-artifact@v4
+      - uses: actions/download-artifact@v5
         with:
           name: Packages
           path: dist

{fitscube-2.2.0 → fitscube-2.3.0}/.github/workflows/ci.yml

@@ -21,7 +21,7 @@ jobs:
     name: Format
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           fetch-depth: 0
       - uses: actions/setup-python@v5
@@ -42,7 +42,7 @@ jobs:
         runs-on: [ubuntu-latest]
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           fetch-depth: 0
 
@@ -62,6 +62,6 @@ jobs:
           --durations=20
 
       - name: Upload coverage report
-        uses: codecov/codecov-action@v5.4.3
+        uses: codecov/codecov-action@v5.5.0
         with:
           token: ${{ secrets.CODECOV_TOKEN }}

{fitscube-2.2.0 → fitscube-2.3.0}/.pre-commit-config.yaml

@@ -10,7 +10,7 @@ repos:
   #       additional_dependencies: [black==24.*]
 
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: "v5.0.0"
+    rev: "v6.0.0"
     hooks:
       - id: check-added-large-files
       - id: check-case-conflict
@@ -33,7 +33,7 @@ repos:
       - id: rst-inline-touching-normal
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: "v0.12.7"
+    rev: "v0.12.11"
     hooks:
       - id: ruff
         args: ["--fix", "--show-fixes"]
@@ -56,7 +56,7 @@ repos:
           - tomli
 
   - repo: https://github.com/shellcheck-py/shellcheck-py
-    rev: "v0.10.0.1"
+    rev: "v0.11.0.1"
     hooks:
       - id: shellcheck
 
@@ -75,7 +75,7 @@ repos:
         additional_dependencies: ["validate-pyproject-schema-store[all]"]
 
   - repo: https://github.com/python-jsonschema/check-jsonschema
-    rev: "0.33.2"
+    rev: "0.33.3"
     hooks:
       - id: check-dependabot
       - id: check-github-workflows

{fitscube-2.2.0 → fitscube-2.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: fitscube
-Version: 2.2.0
+Version: 2.3.0
 Summary: A package to produce produce FITS cubes.
 Project-URL: Homepage, https://github.com/AlecThomson/fitscube
 Project-URL: Bug Tracker, https://github.com/AlecThomson/fitscube/issues

{fitscube-2.2.0 → fitscube-2.3.0}/fitscube/_version.py

@@ -1,7 +1,14 @@
 # file generated by setuptools-scm
 # don't change, don't track in version control
 
-__all__ = ["__version__", "__version_tuple__", "version", "version_tuple"]
+__all__ = [
+    "__version__",
+    "__version_tuple__",
+    "version",
+    "version_tuple",
+    "__commit_id__",
+    "commit_id",
+]
 
 TYPE_CHECKING = False
 if TYPE_CHECKING:
@@ -9,13 +16,19 @@ if TYPE_CHECKING:
     from typing import Union
 
     VERSION_TUPLE = Tuple[Union[int, str], ...]
+    COMMIT_ID = Union[str, None]
 else:
     VERSION_TUPLE = object
+    COMMIT_ID = object
 
 version: str
 __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
+commit_id: COMMIT_ID
+__commit_id__: COMMIT_ID
 
-__version__ = version = '2.2.0'
-__version_tuple__ = version_tuple = (2, 2, 0)
+__version__ = version = '2.3.0'
+__version_tuple__ = version_tuple = (2, 3, 0)
+
+__commit_id__ = commit_id = None

fitscube-2.3.0/fitscube/bounding_box.py

@@ -0,0 +1,149 @@
+"""Some basic utilities to help with the creating of
+bounding boxes to use in fitscube"""
+
+from __future__ import annotations
+
+import asyncio
+from dataclasses import dataclass
+from pathlib import Path
+
+import numpy as np
+from astropy.io import fits
+
+from fitscube.logging import logger
+
+
+@dataclass(frozen=True)
+class BoundingBox:
+    """Simple container to represent a bounding box"""
+
+    xmin: int
+    """Minimum x pixel"""
+    xmax: int
+    """Maximum x pixel"""
+    ymin: int
+    """Minimum y pixel"""
+    ymax: int
+    """Maximum y pixel"""
+    original_shape: tuple[int, int]
+    """The original shape of the image. If constructed against a cube this is the shape of a single plane."""
+    y_span: int
+    """The span between ymax and ymin"""
+    x_span: int
+    """The span between xmax and xmin"""
+
+
+def create_bound_box_plane(image_data: np.ndarray) -> BoundingBox | None:
+    """Create a bounding box around pixels in a 2D image. If all
+    pixels are not valid, then ``None`` is returned.
+
+    Args:
+        image_data (np.ndarray): The 2D image to construct a bounding box around
+
+    Returns:
+        Optional[BoundingBox]: None if no valid pixels, a bounding box with the (xmin,xmax,ymin,ymax) of valid pixels
+    """
+    assert len(image_data.shape) == 2, (
+        f"Only two-dimensional arrays supported, received {image_data.shape}"
+    )
+
+    # First convert to a boolean array
+    image_valid = np.isfinite(image_data)
+
+    if not any(image_valid.reshape(-1)):
+        logger.info("No pixels to creating bounding box for")
+        return None
+
+    # Then make them 1D arrays
+    x_valid = np.any(image_valid, axis=1)
+    y_valid = np.any(image_valid, axis=0)
+
+    # Now get the first and last index
+    xmin, xmax = np.where(x_valid)[0][[0, -1]]
+    ymin, ymax = np.where(y_valid)[0][[0, -1]]
+
+    y_span = ymax - ymin
+    x_span = xmax - xmin
+
+    return BoundingBox(
+        xmin=xmin,
+        xmax=xmax,
+        ymin=ymin,
+        ymax=ymax,
+        y_span=y_span,
+        x_span=x_span,
+        original_shape=image_data.shape[-2:],
+    )
+
+
+def extract_common_bounding_box(
+    bounding_boxes: list[BoundingBox | None],
+) -> BoundingBox:
+    """Get the smallest bounding box that encompasses all bounding boxes
+
+    Args:
+        bounding_boxes (list[BoundingBox | None]): A list of bounding boxes. If None (returned for invalid images) skip it.
+
+    Raises:
+        ValueError: If all input bounding boxes are invalid
+        ValueError: If there is an `original_shape` mismatch
+
+    Returns:
+        BoundingBox: The smallest bounding box
+    """
+
+    # Step 1: filter out all Nones
+    valid_boxes: list[BoundingBox] = [bb for bb in bounding_boxes if bb is not None]
+
+    if len(valid_boxes) == 0:
+        msg = "No valid input boxes to consider"
+        raise ValueError(msg)
+
+    if not all(
+        valid_boxes[0].original_shape == bb.original_shape for bb in valid_boxes
+    ):
+        msg = "Different shapes, and not sure this is really supported or meaningful"
+        raise ValueError(msg)
+
+    xmin = int(np.min([bb.xmin for bb in valid_boxes]))
+    xmax = int(np.max([bb.xmax for bb in valid_boxes]))
+    ymin = int(np.min([bb.ymin for bb in valid_boxes]))
+    ymax = int(np.max([bb.ymax for bb in valid_boxes]))
+
+    y_span = ymax - ymin
+    x_span = xmax - xmin
+
+    return BoundingBox(
+        xmin=xmin,
+        xmax=xmax,
+        ymin=ymin,
+        ymax=ymax,
+        y_span=y_span,
+        x_span=x_span,
+        original_shape=valid_boxes[0].original_shape,
+    )
+
+
+async def get_bounding_box_for_fits_coro(
+    fits_path: Path, invalidate_zeros: bool = False
+) -> BoundingBox | None:
+    """Create a bounding box for an image contained in a FITS file.
+
+    The assumption is that the FITS file contains an image, not a cube.
+    If the cube can bot be reshapped to an image without losing data
+    the underlying bounding box creation will fail.
+
+    Args:
+        fits_path (Path): The fits image to call
+        invalidate_zeros (bool, optional): Mark pixels that are exactly 0.0 as invalid (NaN them). Defaults to False.
+
+    Returns:
+        BoundingBox | None: The bounding box that describes the bounds of valid data. If all data are invalid (and not bounding box possible) None is returned.
+    """
+    data = await asyncio.to_thread(fits.getdata, fits_path, memmap=False)
+    data = np.squeeze(data)
+
+    if invalidate_zeros:
+        data[data == 0.0] = np.nan
+
+    return await asyncio.to_thread(create_bound_box_plane, image_data=data)

{fitscube-2.2.0 → fitscube-2.3.0}/fitscube/combine_fits.py

@@ -34,6 +34,11 @@ from radio_beam.beam import NoBeamException
 from tqdm.asyncio import tqdm
 
 from fitscube.asyncio import gather_with_limit, sync_wrapper
+from fitscube.bounding_box import (
+    BoundingBox,
+    extract_common_bounding_box,
+    get_bounding_box_for_fits_coro,
+)
 from fitscube.logging import TQDM_OUT, logger, set_verbosity
 
 T = TypeVar("T")
@@ -170,6 +175,7 @@ async def create_cube_from_scratch_coro(
     output_shape = output_wcs.array_shape
     msg = f"Creating a new FITS file with shape {output_shape}"
     logger.info(msg)
+
     # If the output shape is less than 1801, we can create a blank array
     # in memory and write it to disk
     if np.prod(output_shape) < 1801:
@@ -240,6 +246,7 @@ async def create_output_cube_coro(
     single_beam: bool = False,
     overwrite: bool = False,
     time_domain_mode: bool = False,
+    bounding_box: BoundingBox | None = None,
 ) -> InitResult:
     """Initialize the data cube.
 
@@ -358,6 +365,14 @@ async def create_output_cube_coro(
         )
         del new_header["BMAJ"], new_header["BMIN"], new_header["BPA"]
 
+    if bounding_box:
+        logger.info("Updating CRPIX1 and CRPIX2 header values to reflect bounding box")
+        new_header["CRPIX1"] -= bounding_box.ymin
+        new_header["CRPIX2"] -= bounding_box.xmin
+        logger.info("Updating NAXIS1 and NAXIS2 ro reflect bounding box")
+        new_header["NAXIS1"] = bounding_box.y_span
+        new_header["NAXIS2"] = bounding_box.x_span
+
     plane_shape = list(old_data.shape)
     cube_shape = plane_shape.copy()
     if is_2d:
@@ -619,6 +634,8 @@ async def process_channel(
     old_channel: int,
     is_missing: bool,
     file_list: list[Path],
+    bounding_box: BoundingBox | None = None,
+    invalidate_zeros: bool = False,
 ) -> None:
     msg = f"Processing channel {new_channel}"
     logger.info(msg)
@@ -631,6 +648,15 @@ async def process_channel(
             fits.getdata, file_list[old_channel], memmap=False
         )
 
+    if bounding_box is not None:
+        plane = plane[
+            ...,
+            bounding_box.xmin : bounding_box.xmax,
+            bounding_box.ymin : bounding_box.ymax,
+        ]
+    if invalidate_zeros:
+        plane[plane == 0.0] = np.nan
+
     await write_channel_to_cube_coro(
         file_handle=file_handle,
         plane=plane,
@@ -650,6 +676,8 @@ async def combine_fits_coro(
     overwrite: bool = False,
     max_workers: int | None = None,
     time_domain_mode: bool = False,
+    bounding_box: bool = False,
+    invalidate_zeros: bool = False,
 ) -> u.Quantity:
     """Combine FITS files into a cube.
     Can handle either frequency or time dimensions agnostically
@@ -660,6 +688,8 @@ async def combine_fits_coro(
         ignore_spec (bool, optional): Ignore frequency/time information. Defaults to False.
         create_blanks (bool, optional): Attempt to create even frequency spacing. Defaults to False.
         time_domain_mode (bool, optional): Work in time domain mode - make a time-cube. Default = False.
+        bounding_box (bool, optional): Clip invalid/padded pixels when crafting the fits cube. When True an extra read of the input daata is needed, but output cube is smaller. Defaults to False.
+        invalidate_zeros (bool, optionals): Set pixels whose values are exactly zero to NaNs. Defaults to False.
 
     Returns:
         tuple[fits.HDUList, u.Quantity]: The combined FITS cube and frequencies
@@ -704,6 +734,21 @@ async def combine_fits_coro(
     specs = specs[new_sort_idx]
     missing_chan_idx = missing_chan_idx[new_sort_idx]
 
+    # Get the bounding box, if requested
+    final_bounding_box = None
+    if bounding_box:
+        boxes_futures = [
+            get_bounding_box_for_fits_coro(
+                fits_path=fits_path, invalidate_zeros=invalidate_zeros
+            )
+            for fits_path in file_list
+        ]
+        boxes = await gather_with_limit(
+            max_workers, *boxes_futures, desc="Bounding boxes"
+        )
+        final_bounding_box = extract_common_bounding_box(bounding_boxes=boxes)
+        logger.info(f"The final bounding box is: {final_bounding_box=}")
+
     # Initialize the data cube
     new_header, _, _, _ = await create_output_cube_coro(
         old_name=file_list[0],
@@ -714,6 +759,7 @@ async def combine_fits_coro(
         single_beam=single_beam,
         overwrite=overwrite,
         time_domain_mode=time_domain_mode,
+        bounding_box=final_bounding_box,
     )
 
     new_channels = np.arange(len(specs))
@@ -741,6 +787,8 @@ async def combine_fits_coro(
                 old_channel=old_channel,
                 is_missing=is_missing,
                 file_list=file_list,
+                bounding_box=final_bounding_box,
+                invalidate_zeros=invalidate_zeros,
             )
             coros.append(coro)
 
@@ -823,6 +871,16 @@ def get_parser(
         default=None,
         help="Maximum number of workers to use for concurrent processing",
     )
+    parser.add_argument(
+        "--bounding-box",
+        action="store_true",
+        help="Attempt to consider padded images when creating the cube. Requires an extract read of the input data.",
+    )
+    parser.add_argument(
+        "--invalidate-zeros",
+        action="store_true",
+        help="Set pixels whose values are exactly zero to NaNs",
+    )
 
     return parser
 
@@ -863,6 +921,8 @@ def cli(args: argparse.Namespace | None = None) -> None:
         overwrite=overwrite,
         max_workers=args.max_workers,
         time_domain_mode=time_domain_mode,
+        bounding_box=args.bounding_box,
+        invalidate_zeros=args.invalidate_zeros,
     )
 
     spequency = "times" if time_domain_mode else "frequencies"