legend-pydataobj 1.12.0a4__tar.gz → 1.14.0__tar.gz

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: legend_pydataobj
-Version: 1.12.0a4
+Version: 1.14.0
 Summary: LEGEND Python Data Objects
 Author: The LEGEND Collaboration
 Maintainer: The LEGEND Collaboration

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/legend_pydataobj.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: legend_pydataobj
-Version: 1.12.0a4
+Version: 1.14.0
 Summary: LEGEND Python Data Objects
 Author: The LEGEND Collaboration
 Maintainer: The LEGEND Collaboration

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/legend_pydataobj.egg-info/SOURCES.txt

@@ -27,6 +27,7 @@ src/lgdo/lh5/core.py
 src/lgdo/lh5/datatype.py
 src/lgdo/lh5/exceptions.py
 src/lgdo/lh5/iterator.py
+src/lgdo/lh5/settings.py
 src/lgdo/lh5/store.py
 src/lgdo/lh5/tools.py
 src/lgdo/lh5/utils.py
@@ -77,6 +78,7 @@ tests/lh5/test_lh5_store.py
 tests/lh5/test_lh5_tools.py
 tests/lh5/test_lh5_utils.py
 tests/lh5/test_lh5_write.py
+tests/lh5/test_pathlib.py
 tests/types/test_array.py
 tests/types/test_arrayofequalsizedarrays.py
 tests/types/test_encoded.py

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '1.12.0a4'
-__version_tuple__ = version_tuple = (1, 12, 0, 'a4')
+__version__ = version = '1.14.0'
+__version_tuple__ = version_tuple = (1, 14, 0)

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/compression/radware.py

@@ -95,13 +95,13 @@ def encode(
     if isinstance(sig_in, np.ndarray):
         s = sig_in.shape
         if len(sig_in) == 0:
-            return np.empty(s[:-1] + (0,), dtype=ubyte), np.empty(0, dtype=uint32)
+            return np.empty((*s[:-1], 0), dtype=ubyte), np.empty(0, dtype=uint32)
 
         if sig_out is None:
             # the encoded signal is an array of bytes
             # -> twice as long as a uint16
             # pre-allocate ubyte (uint8) array, expand last dimension
-            sig_out = np.empty(s[:-1] + (s[-1] * 2,), dtype=ubyte)
+            sig_out = np.empty((*s[:-1], s[-1] * 2), dtype=ubyte)
 
         if sig_out.dtype != ubyte:
             msg = "sig_out must be of type ubyte"
@@ -226,7 +226,7 @@ def decode(
             # allocate output array with lasd dim as large as the longest
             # uncompressed wf
             maxs = np.max(_get_hton_u16(sig_in[0], 0))
-            sig_out = np.empty(s[:-1] + (maxs,), dtype=int32)
+            sig_out = np.empty((*s[:-1], maxs), dtype=int32)
 
         # siglen has one dimension less (the last)
         siglen = np.empty(s[:-1], dtype=uint32)

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/compression/varlen.py

@@ -74,14 +74,14 @@ def encode(
     if isinstance(sig_in, np.ndarray):
         s = sig_in.shape
         if len(sig_in) == 0:
-            return np.empty(s[:-1] + (0,), dtype=ubyte), np.empty(0, dtype=uint32)
+            return np.empty((*s[:-1], 0), dtype=ubyte), np.empty(0, dtype=uint32)
 
         if sig_out is None:
             # the encoded signal is an array of bytes
             # pre-allocate ubyte (uint8) array with a generous (but safe) size
             max_b = int(np.ceil(np.iinfo(sig_in.dtype).bits / 16) * 5)
             # expand last dimension
-            sig_out = np.empty(s[:-1] + (s[-1] * max_b,), dtype=ubyte)
+            sig_out = np.empty((*s[:-1], s[-1] * max_b), dtype=ubyte)
 
         if sig_out.dtype != ubyte:
             msg = "sig_out must be of type ubyte"

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/lh5/__init__.py

@@ -7,7 +7,6 @@ browsed easily in python like any `HDF5 <https://www.hdfgroup.org>`_ file using
 
 from __future__ import annotations
 
-from ._serializers.write.array import DEFAULT_HDF5_SETTINGS
 from .core import read, read_as, write
 from .iterator import LH5Iterator
 from .store import LH5Store
@@ -15,14 +14,15 @@ from .tools import ls, show
 from .utils import read_n_rows
 
 __all__ = [
-    "DEFAULT_HDF5_SETTINGS",
     "LH5Iterator",
     "LH5Store",
     "concat",
+    "default_hdf5_settings",
     "ls",
     "read",
     "read_as",
     "read_n_rows",
+    "reset_default_hdf5_settings",
     "show",
     "write",
 ]

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/lh5/_serializers/read/ndarray.py

@@ -57,7 +57,7 @@ def _h5_read_ndarray(
             (start_row,) + (0,) * (h5d.rank - 1),
             (1,) * h5d.rank,
             None,
-            (n_rows_to_read,) + fspace.shape[1:],
+            (n_rows_to_read, *fspace.shape[1:]),
         )
     elif use_h5idx:
         # Note that h5s will automatically merge adjacent elements into a range
@@ -67,7 +67,7 @@ def _h5_read_ndarray(
                 (i,) + (0,) * (h5d.rank - 1),
                 (1,) * h5d.rank,
                 None,
-                (1,) + fspace.shape[1:],
+                (1, *fspace.shape[1:]),
                 h5py.h5s.SELECT_OR,
             )
 
@@ -84,7 +84,7 @@ def _h5_read_ndarray(
                 (obj_buf_start,) + (0,) * (h5d.rank - 1),
                 (1,) * h5d.rank,
                 None,
-                (n_rows_to_read,) + fspace.shape[1:],
+                (n_rows_to_read, *fspace.shape[1:]),
             )
             h5d.read(mspace, fspace, obj_buf.nda)
         else:
@@ -93,10 +93,10 @@ def _h5_read_ndarray(
             obj_buf.nda[dest_sel, ...] = tmp[idx, ...]
         nda = obj_buf.nda
     elif n_rows == 0:
-        tmp_shape = (0,) + h5d.shape[1:]
+        tmp_shape = (0, *h5d.shape[1:])
         nda = np.empty(tmp_shape, h5d.dtype)
     else:
-        mspace = h5py.h5s.create_simple((n_rows_to_read,) + fspace.shape[1:])
+        mspace = h5py.h5s.create_simple((n_rows_to_read, *fspace.shape[1:]))
         nda = np.empty(mspace.shape, h5d.dtype)
         if idx is None or use_h5idx:
             h5d.read(mspace, fspace, nda)

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/lh5/_serializers/write/array.py

@@ -6,12 +6,11 @@ import h5py
 import numpy as np
 
 from .... import types
+from ... import settings
 from ...exceptions import LH5EncodeError
 
 log = logging.getLogger(__name__)
 
-DEFAULT_HDF5_SETTINGS: dict[str, ...] = {"shuffle": True, "compression": "gzip"}
-
 
 def _h5_write_array(
     obj,
@@ -41,7 +40,7 @@ def _h5_write_array(
         # this is needed in order to have a resizable (in the first
         # axis) data set, i.e. rows can be appended later
         # NOTE: this automatically turns chunking on!
-        maxshape = (None,) + nda.shape[1:]
+        maxshape = (None, *nda.shape[1:])
         h5py_kwargs.setdefault("maxshape", maxshape)
 
         if wo_mode == "o" and name in group:
@@ -49,7 +48,7 @@ def _h5_write_array(
             del group[name]
 
         # set default compression options
-        for k, v in DEFAULT_HDF5_SETTINGS.items():
+        for k, v in settings.DEFAULT_HDF5_SETTINGS.items():
             h5py_kwargs.setdefault(k, v)
 
         # compress using the 'compression' LGDO attribute, if available

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/lh5/_serializers/write/composite.py

@@ -64,6 +64,59 @@ def _h5_write_lgdo(
 
     group = utils.get_h5_group(group, lh5_file)
 
+    # name already in file
+    if name in group or (
+        ("datatype" in group.attrs or group == "/")
+        and (len(name) <= 2 or "/" not in name[1:-1])
+    ):
+        pass
+    # group is in file but not struct or need to create nesting
+    else:
+        # check if name is nested
+        # if name is nested, iterate up from parent
+        # otherwise we just need to iterate the group
+        if len(name) > 2 and "/" in name[1:-1]:
+            group = utils.get_h5_group(
+                name[:-1].rsplit("/", 1)[0],
+                group,
+            )
+            curr_name = (
+                name.rsplit("/", 1)[1]
+                if name[-1] != "/"
+                else name[:-1].rsplit("/", 1)[1]
+            )
+        else:
+            curr_name = name
+        # initialize the object to be written
+        obj = types.Struct({curr_name.replace("/", ""): obj})
+
+        # if base group already has a child we just append
+        if len(group) >= 1:
+            wo_mode = "ac"
+        else:
+            # iterate up the group hierarchy until we reach the root or a group with more than one child
+            while group.name != "/":
+                if len(group) > 1:
+                    break
+                curr_name = group.name
+                group = group.parent
+                if group.name != "/":
+                    obj = types.Struct({curr_name[len(group.name) + 1 :]: obj})
+                else:
+                    obj = types.Struct({curr_name[1:]: obj})
+            # if the group has more than one child, we need to append else we can overwrite
+            wo_mode = "ac" if len(group) > 1 else "o"
+
+        # set the new name
+        if group.name == "/":
+            name = "/"
+        elif group.parent.name == "/":
+            name = group.name[1:]
+        else:
+            name = group.name[len(group.parent.name) + 1 :]
+        # get the new group
+        group = utils.get_h5_group(group.parent if group.name != "/" else "/", lh5_file)
+
     if wo_mode == "w" and name in group:
         msg = f"can't overwrite '{name}' in wo_mode 'write_safe'"
         raise LH5EncodeError(msg, lh5_file, group, name)
@@ -87,7 +140,7 @@ def _h5_write_lgdo(
             lh5_file,
             group=group,
             start_row=start_row,
-            n_rows=n_rows,
+            n_rows=n_rows,  # if isinstance(obj, types.Table | types.Histogram) else None,
             wo_mode=wo_mode,
             write_start=write_start,
             **h5py_kwargs,
@@ -186,19 +239,31 @@ def _h5_write_struct(
     write_start=0,
     **h5py_kwargs,
 ):
+    # this works for structs and derived (tables)
     assert isinstance(obj, types.Struct)
 
     # In order to append a column, we need to update the
-    # `table{old_fields}` value in `group.attrs['datatype"]` to include
-    # the new fields.  One way to do this is to override
-    # `obj.attrs["datatype"]` to include old and new fields. Then we
-    # can write the fields to the table as normal.
+    # `struct/table{old_fields}` value in `group.attrs['datatype"]` to include
+    # the new fields. One way to do this is to override `obj.attrs["datatype"]`
+    # to include old and new fields. Then we can write the fields to the
+    # struct/table as normal.
     if wo_mode == "ac":
+        if name not in group:
+            msg = "Cannot append column to non-existing struct on disk"
+            raise LH5EncodeError(msg, lh5_file, group, name)
+
         old_group = utils.get_h5_group(name, group)
+        if "datatype" not in old_group.attrs:
+            msg = "Cannot append column to an existing  non-LGDO object on disk"
+            raise LH5EncodeError(msg, lh5_file, group, name)
+
         lgdotype = datatype.datatype(old_group.attrs["datatype"])
         fields = datatype.get_struct_fields(old_group.attrs["datatype"])
-        if not issubclass(lgdotype, types.Struct):
-            msg = f"Trying to append columns to an object of type {lgdotype.__name__}"
+        if lgdotype is not type(obj):
+            msg = (
+                "Trying to append columns to an object of different "
+                f"type {lgdotype.__name__}!={type(obj)}"
+            )
             raise LH5EncodeError(msg, lh5_file, group, name)
 
         # If the mode is `append_column`, make sure we aren't appending
@@ -211,8 +276,14 @@ def _h5_write_struct(
                 "column(s) to a table with the same field(s)"
             )
             raise LH5EncodeError(msg, lh5_file, group, name)
+
         # It doesn't matter what key we access, as all fields in the old table have the same size
-        if old_group[next(iter(old_group.keys()))].size != obj.size:
+        if (
+            isinstance(obj, types.Table)
+            and old_group.attrs["datatype"][:6]
+            != "struct"  # structs dont care about size
+            and old_group[next(iter(old_group.keys()))].size != obj.size
+        ):
             msg = (
                 f"Table sizes don't match. Trying to append column of size {obj.size} "
                 f"to a table of size {old_group[next(iter(old_group.keys()))].size}."
@@ -222,16 +293,27 @@ def _h5_write_struct(
         # Now we can append the obj.keys() to the old fields, and then update obj.attrs.
         fields.extend(list(obj.keys()))
         obj.attrs.pop("datatype")
-        obj.attrs["datatype"] = "table" + "{" + ",".join(fields) + "}"
+
+        obj.attrs["datatype"] = (
+            obj.datatype_name() + "{" + ",".join(sorted(fields)) + "}"
+        )
+
+        # propagating wo_mode="ac" to nested LGDOs does not make any sense
+        wo_mode = "append"
+
+        # overwrite attributes of the existing struct
+        attrs_overwrite = True
+    else:
+        attrs_overwrite = wo_mode == "o"
 
     group = utils.get_h5_group(
         name,
         group,
         grp_attrs=obj.attrs,
-        overwrite=(wo_mode in ["o", "ac"]),
+        overwrite=attrs_overwrite,
     )
     # If the mode is overwrite, then we need to peek into the file's
-    # table's existing fields.  If we are writing a new table to the
+    # table's existing fields. If we are writing a new table to the
     # group that does not contain an old field, we should delete that
     # old field from the file
     if wo_mode == "o":
@@ -260,11 +342,9 @@ def _h5_write_struct(
         else:
             obj_fld = obj[field]
 
-        # Convert keys to string for dataset names
-        f = str(field)
         _h5_write_lgdo(
             obj_fld,
-            f,
+            str(field),
             lh5_file,
             group=group,
             start_row=start_row,

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/lh5/core.py

@@ -5,6 +5,7 @@ import inspect
 import sys
 from collections.abc import Mapping, Sequence
 from contextlib import suppress
+from pathlib import Path
 from typing import Any
 
 import h5py
@@ -18,7 +19,7 @@ from .utils import read_n_rows
 
 def read(
     name: str,
-    lh5_file: str | h5py.File | Sequence[str | h5py.File],
+    lh5_file: str | Path | h5py.File | Sequence[str | Path | h5py.File],
     start_row: int = 0,
     n_rows: int = sys.maxsize,
     idx: ArrayLike = None,
@@ -111,8 +112,8 @@ def read(
     """
     if isinstance(lh5_file, h5py.File):
         lh5_obj = lh5_file[name]
-    elif isinstance(lh5_file, str):
-        lh5_file = h5py.File(lh5_file, mode="r", locking=locking)
+    elif isinstance(lh5_file, (str, Path)):
+        lh5_file = h5py.File(str(Path(lh5_file)), mode="r", locking=locking)
         try:
             lh5_obj = lh5_file[name]
         except KeyError as ke:
@@ -194,7 +195,7 @@ def read(
 def write(
     obj: types.LGDO,
     name: str,
-    lh5_file: str | h5py.File,
+    lh5_file: str | Path | h5py.File,
     group: str | h5py.Group = "/",
     start_row: int = 0,
     n_rows: int | None = None,
@@ -268,11 +269,13 @@ def write(
           end of array is the same as ``append``.
         - ``overwrite_file`` or ``of``: delete file if present prior to
           writing to it. `write_start` should be 0 (its ignored).
-        - ``append_column`` or ``ac``: append columns from an
-          :class:`~.lgdo.table.Table` `obj` only if there is an existing
-          :class:`~.lgdo.table.Table` in the `lh5_file` with the same
-          `name` and :class:`~.lgdo.table.Table.size`. If the sizes don't
-          match, or if there are matching fields, it errors out.
+        - ``append_column`` or ``ac``: append fields/columns from an
+          :class:`~.lgdo.struct.Struct` `obj` (and derived types such as
+          :class:`~.lgdo.table.Table`) only if there is an existing
+          :class:`~.lgdo.struct.Struct` in the `lh5_file` with the same `name`.
+          If there are matching fields, it errors out. If appending to a
+          ``Table`` and the size of the new column is different from the size
+          of the existing table, it errors out.
     write_start
         row in the output file (if already existing) to start overwriting
         from.
@@ -288,7 +291,12 @@ def write(
         datasets. **Note: `compression` Ignored if compression is specified
         as an `obj` attribute.**
     """
-    if wo_mode in ("w", "write", "of", "overwrite_file"):
+
+    if (
+        isinstance(lh5_file, str)
+        and not Path(lh5_file).is_file()
+        and wo_mode in ("w", "write_safe", "of", "overwrite_file")
+    ):
         h5py_kwargs.update(
             {
                 "fs_strategy": "page",
@@ -310,7 +318,7 @@ def write(
 
 def read_as(
     name: str,
-    lh5_file: str | h5py.File | Sequence[str | h5py.File],
+    lh5_file: str | Path | h5py.File | Sequence[str | Path | h5py.File],
     library: str,
     **kwargs,
 ) -> Any:

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/lh5/datatype.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import re
 from collections import OrderedDict
+from itertools import permutations as perm
 
 from .. import types as lgdo
 
@@ -14,7 +15,10 @@ _lgdo_datatype_map: dict[str, lgdo.LGDO] = OrderedDict(
             lgdo.ArrayOfEncodedEqualSizedArrays,
             r"^array_of_encoded_equalsized_arrays<1,1>\{.+\}$",
         ),
-        (lgdo.Histogram, r"^struct\{binning,weights,isdensity\}$"),
+        (
+            lgdo.Histogram,
+            rf"^struct\{{(?:{'|'.join([','.join(p) for p in perm(['binning', 'weights', 'isdensity'])])})\}}$",
+        ),
         (lgdo.Struct, r"^struct\{.*\}$"),
         (lgdo.Table, r"^table\{.*\}$"),
         (lgdo.FixedSizeArray, r"^fixedsize_array<\d+>\{.+\}$"),

{legend_pydataobj-1.12.0a4 → legend_pydataobj-1.14.0}/src/lgdo/lh5/iterator.py

@@ -17,54 +17,50 @@ LGDO = typing.Union[Array, Scalar, Struct, VectorOfVectors]
 
 
 class LH5Iterator(typing.Iterator):
-    """
-    A class for iterating through one or more LH5 files, one block of entries
-    at a time. This also accepts an entry list/mask to enable event selection,
-    and a field mask.
-
-    This can be used as an iterator:
-
-
-    >>> for lh5_obj in LH5Iterator(...):
-    >>>    # do the thing!
-
-    This is intended for if you are reading a large quantity of data. This
-    will ensure that you traverse files efficiently to minimize caching time
-    and will limit your memory usage (particularly when reading in waveforms!).
-    The ``lh5_obj`` that is read by this class is reused in order to avoid
-    reallocation of memory; this means that if you want to hold on to data
-    between reads, you will have to copy it somewhere!
-
-    When defining an LH5Iterator, you must give it a list of files and the
-    hdf5 groups containing the data tables you are reading. You may also
-    provide a field mask, and an entry list or mask, specifying which entries
-    to read from the files. You may also pair it with a friend iterator, which
-    contains a parallel group of files which will be simultaneously read.
-    In addition to accessing requested data via ``lh5_obj``, several
-    properties exist to tell you where that data came from:
-
-    - lh5_it.current_i_entry: get the index within the entry list of the
-      first entry that is currently read
-    - lh5_it.current_local_entries: get the entry numbers relative to the
-      file the data came from
-    - lh5_it.current_global_entries: get the entry number relative to the
-      full dataset
-    - lh5_it.current_files: get the file name corresponding to each entry
-    - lh5_it.current_groups: get the group name corresponding to each entry
-
-    This class can also be used for random access:
-
-    >>> lh5_obj = lh5_it.read(i_entry)
-
-    to read the block of entries starting at i_entry. In case of multiple files
-    or the use of an event selection, i_entry refers to a global event index
-    across files and does not count events that are excluded by the selection.
+    """Iterate over chunks of entries from LH5 files.
+
+    The iterator reads ``buffer_len`` entries at a time from one or more
+    files.  The LGDO instance returned at each iteration is reused to avoid
+    reallocations, so copy the data if it should be preserved.
+
+    Examples
+    --------
+    Iterate through a table one chunk at a time::
+
+        from lgdo.lh5 import LH5Iterator
+
+        for table in LH5Iterator("data.lh5", "geds/raw/energy", buffer_len=100):
+            process(table)
+
+    ``LH5Iterator`` can also be used for random access::
+
+        it = LH5Iterator(files, groups)
+        table = it.read(i_entry)
+
+    In case of multiple files or an entry selection, ``i_entry`` refers to the
+    global event index across all files.
+
+    When instantiating an iterator you must provide a list of files and the
+    HDF5 groups to read.  Optional parameters allow field masking, event
+    selection and pairing the iterator with a "friend" iterator that is read in
+    parallel.  Several properties are available to obtain the provenance of the
+    data currently loaded:
+
+    - ``current_i_entry`` -- index within the entry list of the first entry in
+      the buffer
+    - ``current_local_entries`` -- entry numbers relative to the file the data
+      came from
+    - ``current_global_entries`` -- entry number relative to the full dataset
+    - ``current_files`` -- file name corresponding to each entry in the buffer
+    - ``current_groups`` -- group name corresponding to each entry in the
+      buffer
     """
 
     def __init__(
         self,
         lh5_files: str | list[str],
         groups: str | list[str] | list[list[str]],
+        *,
         base_path: str = "",
         entry_list: list[int] | list[list[int]] | None = None,
         entry_mask: list[bool] | list[list[bool]] | None = None,
@@ -75,6 +71,7 @@ class LH5Iterator(typing.Iterator):
         file_cache: int = 10,
         file_map: NDArray[int] = None,
         friend: typing.Iterator | None = None,
+        h5py_open_mode: str = "r",
     ) -> None:
         """
         Parameters
@@ -115,9 +112,21 @@ class LH5Iterator(typing.Iterator):
             The friend should have the same length and entry list. A single
             LH5 table containing columns from both iterators will be returned.
             Note that buffer_len will be set to the minimum of the two.
+        h5py_open_mode
+            file open mode used when acquiring file handles. ``r`` (default)
+            opens files read-only while ``a`` allow opening files for
+            write-appending as well.
         """
         self.lh5_st = LH5Store(base_path=base_path, keep_open=file_cache)
 
+        if h5py_open_mode == "read":
+            h5py_open_mode = "r"
+        if h5py_open_mode == "append":
+            h5py_open_mode = "a"
+        if h5py_open_mode not in ["r", "a"]:
+            msg = f"unknown h5py_open_mode '{h5py_open_mode}'"
+            raise ValueError(msg)
+
         # List of files, with wildcards and env vars expanded
         if isinstance(lh5_files, str):
             lh5_files = [lh5_files]
@@ -152,6 +161,10 @@ class LH5Iterator(typing.Iterator):
                 self.lh5_files += [f_exp] * len(g)
                 self.groups += list(g)
 
+        # open files in the requested mode so they are writable if needed
+        for f in set(self.lh5_files):
+            self.lh5_st.gimme_file(f, mode=h5py_open_mode)
+
         if entry_list is not None and entry_mask is not None:
             msg = "entry_list and entry_mask arguments are mutually exclusive"
             raise ValueError(msg)
@@ -505,7 +518,7 @@ class LH5Iterator(typing.Iterator):
         return self
 
     def __next__(self) -> tuple[LGDO, int, int]:
-        """Read next buffer_len entries and return lh5_table and iterator entry."""
+        """Read the next chunk of entries and return the buffer."""
         n_entries = self.n_entries
         if n_entries is not None:
             n_entries = min(

legend_pydataobj-1.14.0/src/lgdo/lh5/settings.py

@@ -0,0 +1,34 @@
+from __future__ import annotations
+
+from typing import Any
+
+
+def default_hdf5_settings() -> dict[str, Any]:
+    """Returns the HDF5 settings for writing data to disk to the pydataobj defaults.
+
+    Examples
+    --------
+    >>> from lgdo import lh5
+    >>> lh5.DEFAULT_HDF5_SETTINGS["compression"] = "lzf"
+    >>> lh5.write(data, "data", "file.lh5")  # compressed with LZF
+    >>> lh5.DEFAULT_HDF5_SETTINGS = lh5.default_hdf5_settings()
+    >>> lh5.write(data, "data", "file.lh5", "of")  # compressed with default settings (GZIP)
+    """
+
+    return {
+        "shuffle": True,
+        "compression": "gzip",
+    }
+
+
+DEFAULT_HDF5_SETTINGS: dict[str, ...] = default_hdf5_settings()
+"""Global dictionary storing the default HDF5 settings for writing data to disk.
+
+Modify this global variable before writing data to disk with this package.
+
+Examples
+--------
+>>> from lgdo import lh5
+>>> lh5.DEFAULT_HDF5_SETTINGS["compression"] = "lzf"
+>>> lh5.write(data, "data", "file.lh5")  # compressed with LZF
+"""