ak.from_parquet
---------------

.. py:module: ak.from_parquet

Defined in `awkward.operations.ak_from_parquet <https://github.com/scikit-hep/awkward/blob/36da52cfa8846355c390beb6555eac1d31c27c26/src/awkward/operations/ak_from_parquet.py>`__ on `line 17 <https://github.com/scikit-hep/awkward/blob/36da52cfa8846355c390beb6555eac1d31c27c26/src/awkward/operations/ak_from_parquet.py#L17>`__.

.. py:function:: ak.from_parquet(path, *, columns=None, row_groups=None, storage_options=None, max_gap=64000, max_block=256000000, footer_sample_size=1000000, generate_bitmasks=False, highlevel=True, behavior=None, attrs=None)


    :param path: Local filename or remote URL, passed to fsspec for resolution.
             May contain glob patterns.
    :type path: str
    :param columns: Glob pattern(s) including bash-like curly
                brackets for matching column names. Nested records are separated by dots.
                If a list of patterns, the logical-or is matched. If None, all columns
                are read. A list of lists can be provided to select columns with literal dots
                in their names -- The inner list provides column names or patterns.
    :type columns: None, str, or iterable of (str or iterable of str)
    :param row_groups: Row groups to read; must be non-negative.
                   Order is ignored: the output array is presented in the order specified by
                   Parquet metadata. If None, all row groups/all rows are read.
    :type row_groups: None or set of int
    :param storage_options: Passed to ``fsspec.parquet.open_parquet_file``.
    :param max_gap: Passed to ``fsspec.parquet.open_parquet_file``.
    :type max_gap: int
    :param max_block: Passed to ``fsspec.parquet.open_parquet_file``.
    :type max_block: int
    :param footer_sample_size: Passed to ``fsspec.parquet.open_parquet_file``.
    :type footer_sample_size: int
    :param generate_bitmasks: If enabled and Arrow/Parquet does not have Awkward
                          metadata, ``generate_bitmasks=True`` creates empty bitmasks for nullable
                          types that don't have bitmasks in the Arrow/Parquet data, so that the
                          Form (BitMaskedForm vs UnmaskedForm) is predictable.
    :type generate_bitmasks: bool
    :param highlevel: If True, return an :py:obj:`ak.Array`; otherwise, return
                  a low-level :py:obj:`ak.contents.Content` subclass.
    :type highlevel: bool
    :param behavior: Custom :py:obj:`ak.behavior` for the output array, if
                 high-level.
    :type behavior: None or dict
    :param attrs: Custom attributes for the output array, if
              high-level.
    :type attrs: None or dict

Reads data from a local or remote Parquet file or collection of files.

The data are eagerly (not lazily) read and must fit into memory. Use ``columns``
and/or ``row_groups`` to select and filter manageable subsets of the data, and
use :py:obj:`ak.metadata_from_parquet` to find column names and the range of row groups
that a dataset has.

See also :py:obj:`ak.to_parquet`, :py:obj:`ak.metadata_from_parquet`.