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

Defined in `awkward.operations.convert <https://github.com/scikit-hep/awkward-1.0/blob/80bbef0738a6b7928333d7c705ee1b359991de5b/src/awkward/operations/convert.py>`__ on `line 3849 <https://github.com/scikit-hep/awkward-1.0/blob/80bbef0738a6b7928333d7c705ee1b359991de5b/src/awkward/operations/convert.py#L3849>`__.

.. py:function:: ak.from_parquet(source, columns=None, row_groups=None, use_threads=True, include_partition_columns=True, lazy=False, lazy_cache='new', lazy_cache_key=None, highlevel=True, behavior=None)


    :param source: Where to
               get the Parquet file. If ``source`` is the name of a local directory
               (str or Path), then it is interpreted as a partitioned Parquet dataset.
    :type source: str, Path, file-like object, pyarrow.NativeFile
    :param columns: If None, read all columns; otherwise,
                read a specified set of columns.
    :type columns: None or list of str
    :param row_groups: If None, read all row groups;
                   otherwise, read a single or list of row groups.
    :type row_groups: None, int, or list of int
    :param use_threads: Passed to the pyarrow.parquet.ParquetFile.read
                    functions; if True, do multithreaded reading.
    :type use_threads: bool
    :param include_partition_columns: If True and ``source`` is a partitioned
                                  Parquet dataset with subdirectory names defining partition names
                                  and values, include those special columns in the output.
    :type include_partition_columns: bool
    :param lazy: If True, read columns in row groups on demand (as
             :py:obj:`ak.layout.VirtualArray`, possibly in :py:obj:`ak.partition.PartitionedArray`
             if the file has more than one row group); if False, read all
             requested data immediately.
    :type lazy: bool
    :param lazy_cache: If lazy, pass this
                   cache to the VirtualArrays. If "new", a new dict (keep-forever cache)
                   is created. If None, no cache is used.
    :type lazy_cache: None, "new", or MutableMapping
    :param lazy_cache_key: If lazy, pass this cache_key to the
                       VirtualArrays. If None, a process-unique string is constructed.
    :type lazy_cache_key: None or str
    :param highlevel: If True, return an :py:obj:`ak.Array`; otherwise, return
                  a low-level :py:obj:`ak.layout.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 options: All other options are passed to pyarrow.parquet.ParquetFile.

Reads a Parquet file into an Awkward Array (through pyarrow).

.. code-block:: python


    >>> ak.from_parquet("array1.parquet")
    <Array [[1, 2, 3], [], ... [], [6, 7, 8, 9]] type='6 * var * ?int64'>

See also :py:obj:`ak.from_arrow`, which is used as an intermediate step.
See also :py:obj:`ak.to_parquet`.

