API reference

Everything documented here is importable from the top-level h5col package, which is the intended way to use the library. The pages group the public API by topic; each entry’s full documentation, including the exceptions it can raise, is generated from the docstrings in the source.

• Tables and columns
• Specifications
• Column datatypes
• Filters
• Missing values
• Search indexes
• The query layer
• Exceptions
• Low-level modules

At a glance

Tables and columns:

h5col.Table(group)	A H5Col column-oriented table backed by an HDF5 group.
h5col.Column(dataset, table)	One column of a H5Col table, wrapping its rank-1 HDF5 dataset.
h5col.ListColumn(group, table)	One list column of a H5Col table, wrapping its CLASS=LIST_COLUMN group.

Write-side specifications:

h5col.TableSpec(*, columns, title, ...)	Specification of a whole table: its columns and table-level attributes.
h5col.ColumnSpec(*, name[, dtype, chunks, ...])	Specification of one column dataset.
h5col.ListColumnSpec(*, name, values[, ...])	Specification of a list column (a CLASS=LIST_COLUMN group).
h5col.LeafValuesSpec(*, dtype[, chunks, ...])	A leaf VALUES member of a list column: a rank-1 element dataset.
h5col.StringValuesSpec(*[, nullable, ...])	A STRING_VALUES member: variable-length UTF-8 via OFFSETS + CHARS.
h5col.NestedListSpec(*, values[, nullable, ...])	A nested LIST_COLUMN level: its VALUES member plus this level's mask.

Column datatypes:

h5col.FixedString(nbytes[, encoding])	A fixed-length HDF5 string datatype of nbytes bytes.
h5col.ascii_token_dtype(value)	Return a fixed-length ASCII string dtype sized to hold value plus a NUL.
h5col.bool_dtype()	Return the H5Col boolean dtype (enum over little-endian signed int8).
h5col.is_bool_dtype(dtype)	Return True if dtype is an acceptable H5Col boolean datatype.
h5col.encode_bool(values)	Encode a boolean/0-1 array-like to an int8 array of codes.
h5col.decode_bool(values)	Decode stored integer codes to a NumPy boolean array.

Filters:

h5col.FilterPipeline([filters])	An ordered, immutable sequence of Filter entries.
h5col.Filter(plugin_id[, cd_values, ...])	One entry of a filter pipeline.
h5col.Deflate([level])	The built-in deflate compression filter (HDF5's H5Z_FILTER_DEFLATE).
h5col.Shuffle()	The built-in byte-shuffle filter.
h5col.Fletcher32()	The built-in Fletcher-32 checksum filter.
h5col.from_hdf5plugin(obj)	Adapt an hdf5plugin filter instance to a Filter.

Missing values:

h5col.recommended_fill(dtype)	Return H5Col's recommended fill value for dtype.
h5col.is_missing(values, fill_value)	Apply the canonical missing-value test element-wise.
h5col.validate_fill_outside_range(fill[, ...])	Check that fill lies strictly outside [valid_min, valid_max].

Search indexes:

h5col.SearchIndex(dataset, table[, column_ds])	One search-index dataset of a table, wrapping its HDF5 dataset.
h5col.ChunkMinMaxIndex(dataset, table[, ...])	A CHUNK_MINMAX zone map: per-chunk min/max plus missing counts.
h5col.SortedRowsIndex(dataset, table[, ...])	A SORTED_ROWS permutation: row positions in sorted-value order.
h5col.BitmapIndex(dataset, table[, column_ds])	A BITMAP: per-value packed row bitmaps for equality predicates.

The query layer:

h5col.field(name)	Reference a column by name for building query Expression objects.
h5col.Expression(node)	A boolean combination of predicates, built by field() + & | ~.
h5col.Selection(table, expr)	A lazy, composable row selection over a table.
h5col.QueryPlan(nrows, terms, matched)	Machine-readable explanation of how a Selection was evaluated.

The exception family is on its own page, and the low-level submodules (h5col.indexes, h5col.lists, h5col.missing, h5col.ordering, h5col.references, h5col.reserved) on theirs.