_HdfArray

class _HdfArray(*args, cache='lazy', **kwargs)

Bases: ABC

Abstract base class for a single HDF dataset with optional caching.

Provides the common interface for both data arrays (_HdfData) and coordinate scale arrays (_HdfScale). Concrete subclasses supply the HDF-version-specific implementations of the abstract properties (_shape, dtype, size, nbytes, ndim, attrs) and the _dataset() factory.

Subclasses must not be instantiated directly. Use PsiData() or the concrete classes H4Data, H5Data, H4Scale, and H5Scale.

Attributes:

_vcachenp.ndarray | None

In-memory copy of the full dataset array, or None when not cached.

_cacheCacheType

Active cache mode: 'lazy', 'eager', or None.

See also

_HdfData

Subclass that adds interpolation and scale management.

_HdfScale

Subclass for one-dimensional coordinate arrays.

PsiData

Public factory function.

Methods

clear(**kwargs)	Release the in-memory data cache.
load(**kwargs)	Load the full dataset into the in-memory cache.
read(*args[, unit, mesh, order, scales])	Read data by index with optional unit conversion.
select(id_)	Return the HDF dataset or sub-dataset identified by id_.
slice(*args, **kwargs)	Alias for read().
validate_metadata()	Validate resolved metadata and emit MetaDataWarning for any issues.

Properties

attrs	HDF dataset-level attributes as a plain Python dictionary.
cache	Active cache mode.
cached	Alias for data_cached.
data_cached	Whether the full data array is currently held in memory.
dataset	The primary HDF dataset object for this reader.
desc	Human-readable description of the physical quantity.
dtype	Element data type of the HDF dataset.
mesh	Yee-grid stagger code for this dataset.
name	Canonical lower-case quantity identifier.
nbytes	Dataset size in bytes.
ndim	Number of spatial dimensions in the dataset.
order	Memory layout of the stored array.
shape	Array shape in physical (r, t, p) order.
size	Total number of elements in the dataset.
unit	Astropy unit for converting from code units to physical units.