legend-pydataobj 1.11.7__py3-none-any.whl → 1.12.0a2__py3-none-any.whl

{legend_pydataobj-1.11.7.dist-info → legend_pydataobj-1.12.0a2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: legend_pydataobj
-Version: 1.11.7
+Version: 1.12.0a2
 Summary: LEGEND Python Data Objects
 Author: The LEGEND Collaboration
 Maintainer: The LEGEND Collaboration

{legend_pydataobj-1.11.7.dist-info → legend_pydataobj-1.12.0a2.dist-info}/RECORD

@@ -1,9 +1,8 @@
-legend_pydataobj-1.11.7.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-lgdo/__init__.py,sha256=QMYK9HhoMi0pbahPN8mPD18gyTxscFgo7QKfCxVhy-0,3196
-lgdo/_version.py,sha256=WYo6AtimYOvXEEB_DEJYUqS-yeVHGFoR5t7JM_9dSwo,513
+legend_pydataobj-1.12.0a2.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+lgdo/__init__.py,sha256=fkRv79kdtBasw31gPVK9SdLQ2vEEajTV2t3UPDvFg9o,3206
+lgdo/_version.py,sha256=JnfDM_d42edyo74E887XCyb9fjk-f5Vnz-5nNPPYBpo,515
 lgdo/cli.py,sha256=s_EWTBWW76l7zWb6gaTSTjiT-0RzzcYEmjeFEQCVxfk,4647
 lgdo/lgdo_utils.py,sha256=6a2YWEwpyEMXlAyTHZMO01aqxy6SxJzPZkGNWKNWuS0,2567
-lgdo/lh5_store.py,sha256=5BzbJA9sLcqjp8bJDc2olwOiw0VS6rmfg3cfh1kQkRY,8512
 lgdo/logging.py,sha256=82wIOj7l7xr3WYyeHdpSXbbjzHJsy-uRyKYUYx2vMfQ,1003
 lgdo/units.py,sha256=VQYME86_ev9S7Fq8RyCOQNqYr29MphTTYemmEouZafk,161
 lgdo/utils.py,sha256=WRTmXnaQ-h2hVxwJ27qiOigdsD3DHcaDrdDjvupCuZU,3940
@@ -13,19 +12,19 @@ lgdo/compression/generic.py,sha256=tF3UhLJbUDcovLxpIzgQRxFSjZ5Fz3uDRy9kI4mFntQ,2
 lgdo/compression/radware.py,sha256=GcNTtjuyL7VBBqziUBmSqNXuhqy1bJJgvcyvyumPtrc,23839
 lgdo/compression/utils.py,sha256=W2RkBrxPpXlat84dnU9Ad7d_tTws0irtGl7O1dNWjnk,1140
 lgdo/compression/varlen.py,sha256=6ZZUItyoOfygDdE0DyoISeFZfqdbH6xl7T0eclfarzg,15127
-lgdo/lh5/__init__.py,sha256=y1XE_mpFWwamrl7WVjAVSVB25X4PrEfdVXSneSQEmlQ,825
-lgdo/lh5/concat.py,sha256=5nO7dNSb0UEP9rZiWGTKH5Cfwsm5LSm3tBJM4Kd70u0,6336
-lgdo/lh5/core.py,sha256=__-A6Abctzfwfo4-xJi68xs2e4vfzONEQTJVrUCOw-I,13922
+lgdo/lh5/__init__.py,sha256=UTzKGmpgFoHwVB_yNULvJsHD_uQQGl-R87l-3QBkh7w,773
+lgdo/lh5/concat.py,sha256=BZCgK7TWPKK8fMmha8K83d3bC31FVO1b5LOW7x-Ru1s,6186
+lgdo/lh5/core.py,sha256=GjosZGUp4GSO5FtWV9eXUt_6DGU_OwJXODlj5K1j93M,13320
 lgdo/lh5/datatype.py,sha256=O_7BqOlX8PFMyG0ppkfUT5aps5HEqX0bpuKcJO3jhu0,1691
 lgdo/lh5/exceptions.py,sha256=3kj8avXl4eBGvebl3LG12gJEmw91W0T8PYR0AfvUAyM,1211
-lgdo/lh5/iterator.py,sha256=ZaBBnmuNIjinwO0JUY55wLxX8Om9rVRRzXBC5uHmSKM,19772
-lgdo/lh5/store.py,sha256=3wAaQDd1Zmo0_bQ9DbB-FbKS4Uy_Tb642qKHXtZpSw4,10643
-lgdo/lh5/tools.py,sha256=T9CgHA8A3_tVBMtiNJ6hATQKhdqI61m3cX4p2wGKc6c,9937
+lgdo/lh5/iterator.py,sha256=1ob9B7Bf3ioGCtZkUZoL6ibTxAwLf4ld8_33ghVVEa4,20498
+lgdo/lh5/store.py,sha256=MYbMt-Mc7izELxuyLlSrrYrylCIzxc2CLzZYIVbZ33w,8455
+lgdo/lh5/tools.py,sha256=drtJWHY82wCFuFr6LVVnm2AQgs_wZuFmAvyOB4tcOHs,6431
 lgdo/lh5/utils.py,sha256=ioz8DlyXZsejwnU2qYdIccdHcF12H62jgLkZsiDOLSM,6243
 lgdo/lh5/_serializers/__init__.py,sha256=eZzxMp1SeZWG0PkEXUiCz3XyprQ8EmelHUmJogC8xYE,1263
 lgdo/lh5/_serializers/read/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 lgdo/lh5/_serializers/read/array.py,sha256=uWfMCihfAmW2DE2ewip2qCK_kvQC_mb2zvOv26uzijc,1000
-lgdo/lh5/_serializers/read/composite.py,sha256=yTm5dfTgkIL7eG9iZXxhdiRhG04cQLd_hybP4wmxCJE,11809
+lgdo/lh5/_serializers/read/composite.py,sha256=UvkZHEhf0V7SFLxzF52eyP68hU0guGOLqosrfmIfeys,11729
 lgdo/lh5/_serializers/read/encoded.py,sha256=Q98c08d8LkZq2AlY4rThYECVaEqwbv4T2Urn7TGnsyE,4130
 lgdo/lh5/_serializers/read/ndarray.py,sha256=lFCXD6bSzmMOH7cVmvRYXakkfMCI8EoqTPNONRJ1F0s,3690
 lgdo/lh5/_serializers/read/scalar.py,sha256=kwhWm1T91pXf86CqtUUD8_qheSR92gXZrQVtssV5YCg,922
@@ -37,20 +36,20 @@ lgdo/lh5/_serializers/write/composite.py,sha256=I6lH0nWFIpAfZyG4-0rLxzg3mfazZ_FE
 lgdo/lh5/_serializers/write/scalar.py,sha256=JPt_fcdTKOSFp5hfJdcKIfK4hxhcD8vhOlvDF-7btQ8,763
 lgdo/lh5/_serializers/write/vector_of_vectors.py,sha256=puGQX9XF5P_5DVbm_Cc6TvPrsDywgBLSYtkqFNltbB4,3493
 lgdo/types/__init__.py,sha256=DNfOErPiAZg-7Gygkp6ZKAi20Yrm1mfderZHvKo1Y4s,821
-lgdo/types/array.py,sha256=sUxh1CNCaefrnybt5qdjmmMpVQa_RqFxUv1tJ_pyBbc,6537
+lgdo/types/array.py,sha256=vxViJScqKw4zGUrrIOuuU_9Y0oTfOkEEhs0TOyUYjwI,9284
 lgdo/types/arrayofequalsizedarrays.py,sha256=DOGJiTmc1QCdm7vLbE6uIRXoMPtt8uuCfmwQawgWf5s,4949
-lgdo/types/encoded.py,sha256=JW4U5ow7KLMzhKnmhdnxbC3SZJAs4bOEDZWKG4KY1uU,15293
+lgdo/types/encoded.py,sha256=_e8u_BPfpjJbLnEdyTo9QG3kbNsGj0BN4gjdj3L1ndw,15640
 lgdo/types/fixedsizearray.py,sha256=7RjUwTz1bW0pcrdy27JlfrXPAuOU89Kj7pOuSUCojK8,1527
-lgdo/types/histogram.py,sha256=y6j2VDuGYYnLy7WI4J90ApS0PAwic4kCpouZPX09Nus,19974
-lgdo/types/lgdo.py,sha256=RQ2P70N7IWMBDnLLuJI3sm6zQTIKyOMSsKZtBNzmE90,2928
+lgdo/types/histogram.py,sha256=Jz1lLH56BfYnmcUhxUHK1h2wLDQ0Abgyd-6LznU-3-k,19979
+lgdo/types/lgdo.py,sha256=21YNtJCHnSO3M60rjsAdbMO5crDjL_0BtuFpudZ2xvU,4500
 lgdo/types/scalar.py,sha256=c5Es2vyDqyWTPV6mujzfIzMpC1jNWkEIcvYyWQUxH3Q,1933
 lgdo/types/struct.py,sha256=Q0OWLVd4B0ciLb8t6VsxU3MPbmGLZ7WfQNno1lSQS0Q,4918
-lgdo/types/table.py,sha256=VIHQOPXJHJgiCjMMb_p7EdbcCqLFSObHMdHSxC1Dm5Y,19212
-lgdo/types/vectorofvectors.py,sha256=K8w7CZou857I9YGkeOe2uYB20gbHl4OV9xhnnJPNOjc,24665
-lgdo/types/vovutils.py,sha256=7BWPP0BSj-92ifbCIUBcfqxG5-TS8uxujTyJJuDFI04,10302
-lgdo/types/waveformtable.py,sha256=f2tS4f1OEoYaTM5ldCX9zmw8iSISCT3t3wS1SrPdu_o,9901
-legend_pydataobj-1.11.7.dist-info/METADATA,sha256=Z0-UFMzWILag78U1HkNpbYwKDb_JZkZ8kZLtW4T8gw0,44443
-legend_pydataobj-1.11.7.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
-legend_pydataobj-1.11.7.dist-info/entry_points.txt,sha256=0KWfnwbuwhNn0vPUqARukjp04Ca6lzfZBSirouRmk7I,76
-legend_pydataobj-1.11.7.dist-info/top_level.txt,sha256=KyR-EUloqiXcQ62IWnzBmtInDtvsHl4q2ZJAZgTcLXE,5
-legend_pydataobj-1.11.7.dist-info/RECORD,,
+lgdo/types/table.py,sha256=FkWesoEA9bmGGSW8Ewig1Zs77ffUoR_nggfYSmkWpjU,20079
+lgdo/types/vectorofvectors.py,sha256=-5m3g5w03nqs__Uv4cO36A_7_h_4mJhFpIhzJh3Y5D0,24855
+lgdo/types/vovutils.py,sha256=LW3ZcwECxVYxxcFadAtY3nnK-9-rk8Xbg_m8hY30lo4,10708
+lgdo/types/waveformtable.py,sha256=9S_NMg894NZTGt2pLuskwH4-zQ5EbLnzWI6FVui6fXE,9827
+legend_pydataobj-1.12.0a2.dist-info/METADATA,sha256=71-hhjEgQZ9NqNS7FQkFYPALw6VZJf6vDMtCGMrdohE,44445
+legend_pydataobj-1.12.0a2.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
+legend_pydataobj-1.12.0a2.dist-info/entry_points.txt,sha256=0KWfnwbuwhNn0vPUqARukjp04Ca6lzfZBSirouRmk7I,76
+legend_pydataobj-1.12.0a2.dist-info/top_level.txt,sha256=KyR-EUloqiXcQ62IWnzBmtInDtvsHl4q2ZJAZgTcLXE,5
+legend_pydataobj-1.12.0a2.dist-info/RECORD,,

lgdo/__init__.py

@@ -45,7 +45,7 @@ browsed easily in python like any `HDF5 <https://www.hdfgroup.org>`_ file using
 from __future__ import annotations
 
 from ._version import version as __version__
-from .lh5_store import LH5Iterator, LH5Store, load_dfs, load_nda, ls, show
+from .lh5 import LH5Iterator, ls, read, read_as, read_n_rows, show, write
 from .types import (
     LGDO,
     Array,
@@ -69,7 +69,6 @@ __all__ = [
     "FixedSizeArray",
     "Histogram",
     "LH5Iterator",
-    "LH5Store",
     "Scalar",
     "Struct",
     "Table",
@@ -77,8 +76,10 @@ __all__ = [
     "VectorOfVectors",
     "WaveformTable",
     "__version__",
-    "load_dfs",
-    "load_nda",
     "ls",
+    "read",
+    "read_as",
+    "read_n_rows",
     "show",
+    "write",
 ]

lgdo/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '1.11.7'
-__version_tuple__ = version_tuple = (1, 11, 7)
+__version__ = version = '1.12.0a2'
+__version_tuple__ = version_tuple = (1, 12, 0)

lgdo/lh5/__init__.py

@@ -11,7 +11,7 @@ from ._serializers.write.array import DEFAULT_HDF5_SETTINGS
 from .core import read, read_as, write
 from .iterator import LH5Iterator
 from .store import LH5Store
-from .tools import load_dfs, load_nda, ls, show
+from .tools import ls, show
 from .utils import read_n_rows
 
 __all__ = [
@@ -19,8 +19,6 @@ __all__ = [
     "LH5Iterator",
     "LH5Store",
     "concat",
-    "load_dfs",
-    "load_nda",
     "ls",
     "read",
     "read_as",

lgdo/lh5/_serializers/read/composite.py

@@ -353,15 +353,13 @@ def _h5_read_table(
             table = Table(col_dict=col_dict, attrs=attrs)
 
         # set (write) loc to end of tree
-        table.loc = n_rows_read
+        table.resize(do_warn=True)
         return table, n_rows_read
 
     # We have read all fields into the object buffer. Run
     # checks: All columns should be the same size. So update
     # table's size as necessary, warn if any mismatches are found
     obj_buf.resize(do_warn=True)
-    # set (write) loc to end of tree
-    obj_buf.loc = obj_buf_start + n_rows_read
 
     # check attributes
     utils.check_obj_buf_attrs(obj_buf.attrs, attrs, fname, oname)

lgdo/lh5/concat.py

@@ -76,7 +76,7 @@ def _get_lgdos(file, obj_list):
                 continue
 
             # read as little as possible
-            obj, _ = store.read(current, h5f0, n_rows=1)
+            obj = store.read(current, h5f0, n_rows=1)
             if isinstance(obj, (Table, Array, VectorOfVectors)):
                 lgdos.append(current)
 
@@ -139,12 +139,6 @@ def _remove_nested_fields(lgdos: dict, obj_list: list):
         _inplace_table_filter(key, val, obj_list)
 
 
-def _slice(obj, n_rows):
-    ak_obj = obj.view_as("ak")[:n_rows]
-    obj_type = type(obj)
-    return obj_type(ak_obj)
-
-
 def lh5concat(
     lh5_files: list,
     output: str,
@@ -186,8 +180,8 @@ def lh5concat(
     # loop over lgdo objects
     for lgdo in lgdos:
         # iterate over the files
-        for lh5_obj, _, n_rows in LH5Iterator(lh5_files, lgdo):
-            data = {lgdo: _slice(lh5_obj, n_rows)}
+        for lh5_obj in LH5Iterator(lh5_files, lgdo):
+            data = {lgdo: lh5_obj}
 
             # remove the nested fields
             _remove_nested_fields(data, obj_list)

lgdo/lh5/core.py

@@ -4,6 +4,7 @@ import bisect
 import inspect
 import sys
 from collections.abc import Mapping, Sequence
+from contextlib import suppress
 from typing import Any
 
 import h5py
@@ -92,8 +93,7 @@ def read(
         will be set to ``True``, while the rest will default to ``False``.
     obj_buf
         Read directly into memory provided in `obj_buf`. Note: the buffer
-        will be expanded to accommodate the data requested. To maintain the
-        buffer length, send in ``n_rows = len(obj_buf)``.
+        will be resized to accommodate the data retrieved.
     obj_buf_start
         Start location in ``obj_buf`` for read. For concatenating data to
         array-like objects.
@@ -106,12 +106,8 @@ def read(
 
     Returns
     -------
-    (object, n_rows_read)
-        `object` is the read-out object `n_rows_read` is the number of rows
-        successfully read out. Essential for arrays when the amount of data
-        is smaller than the object buffer.  For scalars and structs
-        `n_rows_read` will be``1``. For tables it is redundant with
-        ``table.loc``. If `obj_buf` is ``None``, only `object` is returned.
+    object
+        the read-out object
     """
     if isinstance(lh5_file, h5py.File):
         lh5_obj = lh5_file[name]
@@ -119,12 +115,12 @@ def read(
         lh5_file = h5py.File(lh5_file, mode="r", locking=locking)
         lh5_obj = lh5_file[name]
     else:
-        lh5_files = list(lh5_file)
-
-        n_rows_read = 0
-        obj_buf_is_new = False
+        if obj_buf is not None:
+            obj_buf.resize(obj_buf_start)
+        else:
+            obj_buf_start = 0
 
-        for i, h5f in enumerate(lh5_files):
+        for i, h5f in enumerate(lh5_file):
             if (
                 isinstance(idx, (list, tuple))
                 and len(idx) > 0
@@ -146,33 +142,26 @@ def read(
                 idx = np.array(idx[0])[n_rows_to_read_i:] - n_rows_i
             else:
                 idx_i = None
-            n_rows_i = n_rows - n_rows_read
 
-            obj_ret = read(
+            obj_buf_start_i = len(obj_buf) if obj_buf else 0
+            n_rows_i = n_rows - (obj_buf_start_i - obj_buf_start)
+
+            obj_buf = read(
                 name,
                 h5f,
-                start_row,
+                start_row if i == 0 else 0,
                 n_rows_i,
                 idx_i,
                 use_h5idx,
                 field_mask,
                 obj_buf,
-                obj_buf_start,
+                obj_buf_start_i,
                 decompress,
             )
-            if isinstance(obj_ret, tuple):
-                obj_buf, n_rows_read_i = obj_ret
-                obj_buf_is_new = True
-            else:
-                obj_buf = obj_ret
-                n_rows_read_i = len(obj_buf)
 
-            n_rows_read += n_rows_read_i
-            if n_rows_read >= n_rows or obj_buf is None:
-                return obj_buf, n_rows_read
-            start_row = 0
-            obj_buf_start += n_rows_read_i
-        return obj_buf if obj_buf_is_new else (obj_buf, n_rows_read)
+            if obj_buf is None or (len(obj_buf) - obj_buf_start) >= n_rows:
+                return obj_buf
+        return obj_buf
 
     if isinstance(idx, (list, tuple)) and len(idx) > 0 and not np.isscalar(idx[0]):
         idx = idx[0]
@@ -192,8 +181,10 @@ def read(
         obj_buf_start=obj_buf_start,
         decompress=decompress,
     )
+    with suppress(AttributeError):
+        obj.resize(obj_buf_start + n_rows_read)
 
-    return obj if obj_buf is None else (obj, n_rows_read)
+    return obj
 
 
 def write(

lgdo/lh5/iterator.py

@@ -24,7 +24,8 @@ class LH5Iterator(typing.Iterator):
 
     This can be used as an iterator:
 
-    >>> for lh5_obj, i_entry, n_rows in LH5Iterator(...):
+
+    >>> for lh5_obj in LH5Iterator(...):
     >>>    # do the thing!
 
     This is intended for if you are reading a large quantity of data. This
@@ -42,6 +43,8 @@ class LH5Iterator(typing.Iterator):
     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
@@ -49,9 +52,9 @@ class LH5Iterator(typing.Iterator):
     - 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 either for random access:
+    This class can also be used for random access:
 
-    >>> lh5_obj, n_rows = lh5_it.read(i_entry)
+    >>> 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
@@ -65,6 +68,8 @@ class LH5Iterator(typing.Iterator):
         base_path: str = "",
         entry_list: list[int] | list[list[int]] | None = None,
         entry_mask: list[bool] | list[list[bool]] | None = None,
+        i_start: int = 0,
+        n_entries: int | None = None,
         field_mask: dict[str, bool] | list[str] | tuple[str] | None = None,
         buffer_len: int = "100*MB",
         file_cache: int = 10,
@@ -89,6 +94,10 @@ class LH5Iterator(typing.Iterator):
         entry_mask
             mask of entries to read. If a list of arrays is provided, expect
             one for each file. Ignore if a selection list is provided.
+        i_start
+            index of first entry to start at when iterating
+        n_entries
+            number of entries to read before terminating iteration
         field_mask
             mask of which fields to read. See :meth:`LH5Store.read` for
             more details.
@@ -183,7 +192,8 @@ class LH5Iterator(typing.Iterator):
             msg = f"can't open any files from {lh5_files}"
             raise RuntimeError(msg)
 
-        self.n_rows = 0
+        self.i_start = i_start
+        self.n_entries = n_entries
         self.current_i_entry = 0
         self.next_i_entry = 0
 
@@ -317,14 +327,21 @@ class LH5Iterator(typing.Iterator):
                 )
         return self.global_entry_list
 
-    def read(self, i_entry: int) -> tuple[LGDO, int]:
-        """Read the nextlocal chunk of events, starting at i_entry. Return the
-        LH5 buffer and number of rows read."""
-        self.n_rows = 0
-        i_file = np.searchsorted(self.entry_map, i_entry, "right")
+    def read(self, i_entry: int, n_entries: int | None = None) -> LGDO:
+        "Read the nextlocal chunk of events, starting at entry."
+        self.lh5_buffer.resize(0)
+
+        if n_entries is None:
+            n_entries = self.buffer_len
+        elif n_entries == 0:
+            return self.lh5_buffer
+        elif n_entries > self.buffer_len:
+            msg = "n_entries cannot be larger than buffer_len"
+            raise ValueError(msg)
 
         # if file hasn't been opened yet, search through files
         # sequentially until we find the right one
+        i_file = np.searchsorted(self.entry_map, i_entry, "right")
         if i_file < len(self.lh5_files) and self.entry_map[i_file] == np.iinfo("q").max:
             while i_file < len(self.lh5_files) and i_entry >= self._get_file_cumentries(
                 i_file
@@ -332,10 +349,10 @@ class LH5Iterator(typing.Iterator):
                 i_file += 1
 
         if i_file == len(self.lh5_files):
-            return (self.lh5_buffer, self.n_rows)
+            return self.lh5_buffer
         local_i_entry = i_entry - self._get_file_cumentries(i_file - 1)
 
-        while self.n_rows < self.buffer_len and i_file < len(self.file_map):
+        while len(self.lh5_buffer) < n_entries and i_file < len(self.file_map):
             # Loop through files
             local_idx = self.get_file_entrylist(i_file)
             if local_idx is not None and len(local_idx) == 0:
@@ -344,18 +361,17 @@ class LH5Iterator(typing.Iterator):
                 continue
 
             i_local = local_i_entry if local_idx is None else local_idx[local_i_entry]
-            self.lh5_buffer, n_rows = self.lh5_st.read(
+            self.lh5_buffer = self.lh5_st.read(
                 self.groups[i_file],
                 self.lh5_files[i_file],
                 start_row=i_local,
-                n_rows=self.buffer_len - self.n_rows,
+                n_rows=n_entries - len(self.lh5_buffer),
                 idx=local_idx,
                 field_mask=self.field_mask,
                 obj_buf=self.lh5_buffer,
-                obj_buf_start=self.n_rows,
+                obj_buf_start=len(self.lh5_buffer),
             )
 
-            self.n_rows += n_rows
             i_file += 1
             local_i_entry = 0
 
@@ -364,7 +380,7 @@ class LH5Iterator(typing.Iterator):
         if self.friend is not None:
             self.friend.read(i_entry)
 
-        return (self.lh5_buffer, self.n_rows)
+        return self.lh5_buffer
 
     def reset_field_mask(self, mask):
         """Replaces the field mask of this iterator and any friends with mask"""
@@ -375,7 +391,7 @@ class LH5Iterator(typing.Iterator):
     @property
     def current_local_entries(self) -> NDArray[int]:
         """Return list of local file entries in buffer"""
-        cur_entries = np.zeros(self.n_rows, dtype="int32")
+        cur_entries = np.zeros(len(self.lh5_buffer), dtype="int32")
         i_file = np.searchsorted(self.entry_map, self.current_i_entry, "right")
         file_start = self._get_file_cumentries(i_file - 1)
         i_local = self.current_i_entry - file_start
@@ -402,7 +418,7 @@ class LH5Iterator(typing.Iterator):
     @property
     def current_global_entries(self) -> NDArray[int]:
         """Return list of local file entries in buffer"""
-        cur_entries = np.zeros(self.n_rows, dtype="int32")
+        cur_entries = np.zeros(len(self.lh5_buffer), dtype="int32")
         i_file = np.searchsorted(self.entry_map, self.current_i_entry, "right")
         file_start = self._get_file_cumentries(i_file - 1)
         i_local = self.current_i_entry - file_start
@@ -433,7 +449,7 @@ class LH5Iterator(typing.Iterator):
     @property
     def current_files(self) -> NDArray[str]:
         """Return list of file names for entries in buffer"""
-        cur_files = np.zeros(self.n_rows, dtype=object)
+        cur_files = np.zeros(len(self.lh5_buffer), dtype=object)
         i_file = np.searchsorted(self.entry_map, self.current_i_entry, "right")
         file_start = self._get_file_cumentries(i_file - 1)
         i_local = self.current_i_entry - file_start
@@ -455,7 +471,7 @@ class LH5Iterator(typing.Iterator):
     @property
     def current_groups(self) -> NDArray[str]:
         """Return list of group names for entries in buffer"""
-        cur_groups = np.zeros(self.n_rows, dtype=object)
+        cur_groups = np.zeros(len(self.lh5_buffer), dtype=object)
         i_file = np.searchsorted(self.entry_map, self.current_i_entry, "right")
         file_start = self._get_file_cumentries(i_file - 1)
         i_local = self.current_i_entry - file_start
@@ -485,14 +501,19 @@ class LH5Iterator(typing.Iterator):
     def __iter__(self) -> typing.Iterator:
         """Loop through entries in blocks of size buffer_len."""
         self.current_i_entry = 0
-        self.next_i_entry = 0
+        self.next_i_entry = self.i_start
         return self
 
     def __next__(self) -> tuple[LGDO, int, int]:
-        """Read next buffer_len entries and return lh5_table, iterator entry
-        and n_rows read."""
-        buf, n_rows = self.read(self.next_i_entry)
-        self.next_i_entry = self.current_i_entry + n_rows
-        if n_rows == 0:
+        """Read next buffer_len entries and return lh5_table and iterator entry."""
+        n_entries = self.n_entries
+        if n_entries is not None:
+            n_entries = min(
+                self.buffer_len, n_entries + self.i_start - self.next_i_entry
+            )
+
+        buf = self.read(self.next_i_entry, n_entries)
+        if len(buf) == 0:
             raise StopIteration
-        return (buf, self.current_i_entry, n_rows)
+        self.next_i_entry = self.current_i_entry + len(buf)
+        return buf

lgdo/lh5/store.py

@@ -5,7 +5,6 @@ HDF5 files.
 
 from __future__ import annotations
 
-import bisect
 import logging
 import os
 import sys
@@ -15,11 +14,11 @@ from inspect import signature
 from typing import Any
 
 import h5py
-import numpy as np
 from numpy.typing import ArrayLike
 
 from .. import types
 from . import _serializers, utils
+from .core import read
 
 log = logging.getLogger(__name__)
 
@@ -155,7 +154,7 @@ class LH5Store:
         """Returns an LH5 object appropriate for use as a pre-allocated buffer
         in a read loop. Sets size to `size` if object has a size.
         """
-        obj, n_rows = self.read(name, lh5_file, n_rows=0, field_mask=field_mask)
+        obj = self.read(name, lh5_file, n_rows=0, field_mask=field_mask)
         if hasattr(obj, "resize") and size is not None:
             obj.resize(new_size=size)
         return obj
@@ -182,72 +181,20 @@ class LH5Store:
         """
         # grab files from store
         if isinstance(lh5_file, (str, h5py.File)):
-            lh5_obj = self.gimme_file(lh5_file, "r", **file_kwargs)[name]
+            h5f = self.gimme_file(lh5_file, "r", **file_kwargs)
         else:
-            lh5_files = list(lh5_file)
-            n_rows_read = 0
-
-            for i, h5f in enumerate(lh5_files):
-                if (
-                    isinstance(idx, (list, tuple))
-                    and len(idx) > 0
-                    and not np.isscalar(idx[0])
-                ):
-                    # a list of lists: must be one per file
-                    idx_i = idx[i]
-                elif idx is not None:
-                    # make idx a proper tuple if it's not one already
-                    if not (isinstance(idx, tuple) and len(idx) == 1):
-                        idx = (idx,)
-                    # idx is a long continuous array
-                    n_rows_i = utils.read_n_rows(name, h5f)
-                    # find the length of the subset of idx that contains indices
-                    # that are less than n_rows_i
-                    n_rows_to_read_i = bisect.bisect_left(idx[0], n_rows_i)
-                    # now split idx into idx_i and the remainder
-                    idx_i = np.array(idx[0])[:n_rows_to_read_i]
-                    idx = np.array(idx[0])[n_rows_to_read_i:] - n_rows_i
-                else:
-                    idx_i = None
-                n_rows_i = n_rows - n_rows_read
-
-                obj_buf, n_rows_read_i = self.read(
-                    name,
-                    h5f,
-                    start_row,
-                    n_rows_i,
-                    idx_i,
-                    use_h5idx,
-                    field_mask,
-                    obj_buf,
-                    obj_buf_start,
-                    decompress,
-                )
-
-                n_rows_read += n_rows_read_i
-                if n_rows_read >= n_rows or obj_buf is None:
-                    return obj_buf, n_rows_read
-                start_row = 0
-                obj_buf_start += n_rows_read_i
-            return obj_buf, n_rows_read
-
-        if isinstance(idx, (list, tuple)) and len(idx) > 0 and not np.isscalar(idx[0]):
-            idx = idx[0]
-        if isinstance(idx, np.ndarray) and idx.dtype == np.dtype("?"):
-            idx = np.where(idx)[0]
-
-        return _serializers._h5_read_lgdo(
-            lh5_obj.id,
-            lh5_obj.file.filename,
-            lh5_obj.name,
-            start_row=start_row,
-            n_rows=n_rows,
-            idx=idx,
-            use_h5idx=use_h5idx,
-            field_mask=field_mask,
-            obj_buf=obj_buf,
-            obj_buf_start=obj_buf_start,
-            decompress=decompress,
+            h5f = [self.gimme_file(f, "r", **file_kwargs) for f in lh5_file]
+        return read(
+            name,
+            h5f,
+            start_row,
+            n_rows,
+            idx,
+            use_h5idx,
+            field_mask,
+            obj_buf,
+            obj_buf_start,
+            decompress,
         )
 
     def write(