legend-pydataobj 1.12.0a4__py3-none-any.whl → 1.14.0__py3-none-any.whl

{legend_pydataobj-1.12.0a4.dist-info → legend_pydataobj-1.14.0.dist-info}/METADATA

@@ -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.dist-info → legend_pydataobj-1.14.0.dist-info}/RECORD

@@ -1,6 +1,6 @@
-legend_pydataobj-1.12.0a4.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+legend_pydataobj-1.14.0.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
 lgdo/__init__.py,sha256=fkRv79kdtBasw31gPVK9SdLQ2vEEajTV2t3UPDvFg9o,3206
-lgdo/_version.py,sha256=lVyzAOse2pIwNX9sD_s_ucUhU5oPCN_lFpntrC7eKG8,521
+lgdo/_version.py,sha256=zEosD-3Sqrti57GKf-4yC-NurX2Smyv5d6IDkQisUBo,513
 lgdo/cli.py,sha256=s_EWTBWW76l7zWb6gaTSTjiT-0RzzcYEmjeFEQCVxfk,4647
 lgdo/lgdo_utils.py,sha256=6a2YWEwpyEMXlAyTHZMO01aqxy6SxJzPZkGNWKNWuS0,2567
 lgdo/logging.py,sha256=82wIOj7l7xr3WYyeHdpSXbbjzHJsy-uRyKYUYx2vMfQ,1003
@@ -9,47 +9,48 @@ lgdo/utils.py,sha256=WRTmXnaQ-h2hVxwJ27qiOigdsD3DHcaDrdDjvupCuZU,3940
 lgdo/compression/__init__.py,sha256=xHt_8Th0LxxNwj9iYHf5uGNTm3A_4qyW7zEVdAX3NwI,1127
 lgdo/compression/base.py,sha256=82cQJujfvoAOKBFx761dEcx_xM02TBCBBuBo6i78tuI,838
 lgdo/compression/generic.py,sha256=tF3UhLJbUDcovLxpIzgQRxFSjZ5Fz3uDRy9kI4mFntQ,2515
-lgdo/compression/radware.py,sha256=GcNTtjuyL7VBBqziUBmSqNXuhqy1bJJgvcyvyumPtrc,23839
+lgdo/compression/radware.py,sha256=-W7LgvkSVzdVJ6qNn7Ts3O9EcRcl8mUiApTLqR4dtIo,23836
 lgdo/compression/utils.py,sha256=W2RkBrxPpXlat84dnU9Ad7d_tTws0irtGl7O1dNWjnk,1140
-lgdo/compression/varlen.py,sha256=6ZZUItyoOfygDdE0DyoISeFZfqdbH6xl7T0eclfarzg,15127
-lgdo/lh5/__init__.py,sha256=UTzKGmpgFoHwVB_yNULvJsHD_uQQGl-R87l-3QBkh7w,773
+lgdo/compression/varlen.py,sha256=bjyxhHzfpi6PIPy-Uc47W8_LrRbFoJLJ2kVeD5nhyqo,15125
+lgdo/lh5/__init__.py,sha256=smHTawINIiogHNfYJq3aPvtxleTnBMdPADRCdc1wea8,748
 lgdo/lh5/concat.py,sha256=BZCgK7TWPKK8fMmha8K83d3bC31FVO1b5LOW7x-Ru1s,6186
-lgdo/lh5/core.py,sha256=nULH5UoRjUCH0E3Z0-OH_DbFz2PRAQP73Qaf1kfnyPE,13481
-lgdo/lh5/datatype.py,sha256=O_7BqOlX8PFMyG0ppkfUT5aps5HEqX0bpuKcJO3jhu0,1691
+lgdo/lh5/core.py,sha256=tbvitu3Pr-FCF4nOopVxGVOobDhGaVWo4o0HS58TGtY,13806
+lgdo/lh5/datatype.py,sha256=ry3twFaosuBoskiTKqtBYRMk9PQAf403593xKaItfog,1827
 lgdo/lh5/exceptions.py,sha256=3kj8avXl4eBGvebl3LG12gJEmw91W0T8PYR0AfvUAyM,1211
-lgdo/lh5/iterator.py,sha256=1ob9B7Bf3ioGCtZkUZoL6ibTxAwLf4ld8_33ghVVEa4,20498
-lgdo/lh5/store.py,sha256=qkBm3gPbr1R2UlQpUuDR5sGRMzpYJBWFL8fDIry6tmQ,8474
-lgdo/lh5/tools.py,sha256=drtJWHY82wCFuFr6LVVnm2AQgs_wZuFmAvyOB4tcOHs,6431
-lgdo/lh5/utils.py,sha256=f2H7H1D-RfDN3g_YrVDQEPaHevn5yDJFA-uznK9cgx8,6336
+lgdo/lh5/iterator.py,sha256=vuN98pa-xHDWXM2GMxvMxFEJGfHatMX6ajqnaP55PuY,20680
+lgdo/lh5/settings.py,sha256=cmPd6ZvneAF5sFMA1qf-9g_YSSygJcQSRmZDp1_sBEU,1001
+lgdo/lh5/store.py,sha256=HJuDjWQ8ztrKDoyWW3cplhtWDnz3J4a-Fe2WF4fzOY4,8676
+lgdo/lh5/tools.py,sha256=EZTCj3TMMp4Rnocq1F0QeO1yYHzx4yMR7l_Em4G7sC4,6503
+lgdo/lh5/utils.py,sha256=hxPoaG25MOhuu7emrw2xzx3zerl-GzeMWdlfoQmLiYo,6667
 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=UvkZHEhf0V7SFLxzF52eyP68hU0guGOLqosrfmIfeys,11729
 lgdo/lh5/_serializers/read/encoded.py,sha256=Q98c08d8LkZq2AlY4rThYECVaEqwbv4T2Urn7TGnsyE,4130
-lgdo/lh5/_serializers/read/ndarray.py,sha256=lFCXD6bSzmMOH7cVmvRYXakkfMCI8EoqTPNONRJ1F0s,3690
+lgdo/lh5/_serializers/read/ndarray.py,sha256=cxzZ7esT5BzxyoXfITBG_EDTtCVxSeSu6dVZrohOdOY,3685
 lgdo/lh5/_serializers/read/scalar.py,sha256=kwhWm1T91pXf86CqtUUD8_qheSR92gXZrQVtssV5YCg,922
 lgdo/lh5/_serializers/read/utils.py,sha256=YfSqPO-83A1XvhhuULxQ0Qz2A5ODa3sb7ApNxQVJXd0,7581
 lgdo/lh5/_serializers/read/vector_of_vectors.py,sha256=765P8mElGArAaEPkHTAUXFQ47t1_3-3BQAete0LckBQ,7207
 lgdo/lh5/_serializers/write/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-lgdo/lh5/_serializers/write/array.py,sha256=66DKnW2yqIBlUGNBPWcE-m4W0B2-nTKusDHGX9m6GY0,3223
-lgdo/lh5/_serializers/write/composite.py,sha256=JYoLT9intT_Y4xPeL_l7CSd22O0ZKyEmd0flKkWWPFA,9268
+lgdo/lh5/_serializers/write/array.py,sha256=qzRNPQ4mtvc7HYPE3vUcM6bi7lWYnolNStdJVcDfzPU,3174
+lgdo/lh5/_serializers/write/composite.py,sha256=sZfV8aGZCH0mvMZ2dGDKt-MoepgL4PlR9ZWbT_JNIjQ,12171
 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=vxViJScqKw4zGUrrIOuuU_9Y0oTfOkEEhs0TOyUYjwI,9284
+lgdo/types/array.py,sha256=TpZINHgGIptslwr5mwKYWU_PrYAk8bH1ECJ4XfLkWxg,9338
 lgdo/types/arrayofequalsizedarrays.py,sha256=DOGJiTmc1QCdm7vLbE6uIRXoMPtt8uuCfmwQawgWf5s,4949
-lgdo/types/encoded.py,sha256=_e8u_BPfpjJbLnEdyTo9QG3kbNsGj0BN4gjdj3L1ndw,15640
+lgdo/types/encoded.py,sha256=8DJHb3kxz6RrmjkeLWS6iyjvIJqx86mDInWqqjpMON0,15752
 lgdo/types/fixedsizearray.py,sha256=7RjUwTz1bW0pcrdy27JlfrXPAuOU89Kj7pOuSUCojK8,1527
 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=FkWesoEA9bmGGSW8Ewig1Zs77ffUoR_nggfYSmkWpjU,20079
-lgdo/types/vectorofvectors.py,sha256=GbAKV_ehXN4XdWSwnmKS_ErCiudRetcH_3wo7iDrVjw,26854
+lgdo/types/scalar.py,sha256=nBPiqX4g3GrPavEbG6nCt2Jel7Mj0IchXqwxB6ei_rg,1989
+lgdo/types/struct.py,sha256=m3pYfGfKptV8ti3wb4n1nsPKMvhjdWCFoRdR5YooZBM,6353
+lgdo/types/table.py,sha256=huhgpzdAUx0bRaEaitwnb-Ve7oAu5B6zxPK5EXPUfg0,20233
+lgdo/types/vectorofvectors.py,sha256=k1LwNnX3TcRAhOujj85kNkfZN0MXZYL9aaMUbr82JlE,26910
 lgdo/types/vovutils.py,sha256=LW3ZcwECxVYxxcFadAtY3nnK-9-rk8Xbg_m8hY30lo4,10708
 lgdo/types/waveformtable.py,sha256=9S_NMg894NZTGt2pLuskwH4-zQ5EbLnzWI6FVui6fXE,9827
-legend_pydataobj-1.12.0a4.dist-info/METADATA,sha256=eD1QW8NEKGSWEqxSes1-TFnq1VHoxtdLmLfafsB53nI,44445
-legend_pydataobj-1.12.0a4.dist-info/WHEEL,sha256=ck4Vq1_RXyvS4Jt6SI0Vz6fyVs4GWg7AINwpsaGEgPE,91
-legend_pydataobj-1.12.0a4.dist-info/entry_points.txt,sha256=0KWfnwbuwhNn0vPUqARukjp04Ca6lzfZBSirouRmk7I,76
-legend_pydataobj-1.12.0a4.dist-info/top_level.txt,sha256=KyR-EUloqiXcQ62IWnzBmtInDtvsHl4q2ZJAZgTcLXE,5
-legend_pydataobj-1.12.0a4.dist-info/RECORD,,
+legend_pydataobj-1.14.0.dist-info/METADATA,sha256=JaH2muAaB5Otjd9XhqiFfrgqtf9mR6F4XbIBPlZmB0g,44443
+legend_pydataobj-1.14.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+legend_pydataobj-1.14.0.dist-info/entry_points.txt,sha256=0KWfnwbuwhNn0vPUqARukjp04Ca6lzfZBSirouRmk7I,76
+legend_pydataobj-1.14.0.dist-info/top_level.txt,sha256=KyR-EUloqiXcQ62IWnzBmtInDtvsHl4q2ZJAZgTcLXE,5
+legend_pydataobj-1.14.0.dist-info/RECORD,,

{legend_pydataobj-1.12.0a4.dist-info → legend_pydataobj-1.14.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.0.0)
+Generator: setuptools (80.9.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

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)

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)

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"

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",
 ]

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)

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

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,

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:

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\{.*\}$"),
         (lgdo.FixedSizeArray, r"^fixedsize_array<\d+>\{.+\}$"),

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(

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
+"""

lgdo/lh5/store.py

@@ -38,7 +38,10 @@ class LH5Store:
     """
 
     def __init__(
-        self, base_path: str = "", keep_open: bool = False, locking: bool = False
+        self,
+        base_path: str | Path = "",
+        keep_open: bool = False,
+        locking: bool = False,
     ) -> None:
         """
         Parameters
@@ -52,6 +55,7 @@ class LH5Store:
         locking
             whether to lock files when reading
         """
+        base_path = str(Path(base_path)) if base_path != "" else ""
         self.base_path = "" if base_path == "" else utils.expand_path(base_path)
         self.keep_open = keep_open
         self.locking = locking
@@ -59,7 +63,7 @@ class LH5Store:
 
     def gimme_file(
         self,
-        lh5_file: str | h5py.File,
+        lh5_file: str | Path | h5py.File,
         mode: str = "r",
         page_buffer: int = 0,
         **file_kwargs,
@@ -83,6 +87,8 @@ class LH5Store:
         if isinstance(lh5_file, h5py.File):
             return lh5_file
 
+        lh5_file = str(Path(lh5_file))
+
         if mode == "r":
             lh5_file = utils.expand_path(lh5_file, base_path=self.base_path)
             file_kwargs["locking"] = self.locking
@@ -147,7 +153,7 @@ class LH5Store:
     def get_buffer(
         self,
         name: str,
-        lh5_file: str | h5py.File | Sequence[str | h5py.File],
+        lh5_file: str | Path | h5py.File | Sequence[str | Path | h5py.File],
         size: int | None = None,
         field_mask: Mapping[str, bool] | Sequence[str] | None = None,
     ) -> types.LGDO:
@@ -162,7 +168,7 @@ class LH5Store:
     def read(
         self,
         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,
@@ -180,7 +186,7 @@ class LH5Store:
         .lh5.core.read
         """
         # grab files from store
-        if isinstance(lh5_file, (str, h5py.File)):
+        if isinstance(lh5_file, (str, Path, h5py.File)):
             h5f = self.gimme_file(lh5_file, "r", **file_kwargs)
         else:
             h5f = [self.gimme_file(f, "r", **file_kwargs) for f in lh5_file]
@@ -201,7 +207,7 @@ class LH5Store:
         self,
         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,
@@ -256,14 +262,14 @@ class LH5Store:
             **h5py_kwargs,
         )
 
-    def read_n_rows(self, name: str, lh5_file: str | h5py.File) -> int | None:
+    def read_n_rows(self, name: str, lh5_file: str | Path | h5py.File) -> int | None:
         """Look up the number of rows in an Array-like object called `name` in `lh5_file`.
 
         Return ``None`` if it is a :class:`.Scalar` or a :class:`.Struct`.
         """
         return utils.read_n_rows(name, self.gimme_file(lh5_file, "r"))
 
-    def read_size_in_bytes(self, name: str, lh5_file: str | h5py.File) -> int:
+    def read_size_in_bytes(self, name: str, lh5_file: str | Path | h5py.File) -> int:
         """Look up the size (in B) of the object in memory. Will recursively
         crawl through all objects in a Struct or Table
         """

lgdo/lh5/tools.py

@@ -3,6 +3,7 @@ from __future__ import annotations
 import fnmatch
 import logging
 from copy import copy
+from pathlib import Path
 
 import h5py
 
@@ -13,7 +14,7 @@ log = logging.getLogger(__name__)
 
 
 def ls(
-    lh5_file: str | h5py.Group,
+    lh5_file: str | Path | h5py.Group,
     lh5_group: str = "",
     recursive: bool = False,
 ) -> list[str]:
@@ -39,8 +40,8 @@ def ls(
 
     lh5_st = LH5Store()
     # To use recursively, make lh5_file a h5group instead of a string
-    if isinstance(lh5_file, str):
-        lh5_file = lh5_st.gimme_file(lh5_file, "r")
+    if isinstance(lh5_file, (str, Path)):
+        lh5_file = lh5_st.gimme_file(str(Path(lh5_file)), "r")
         if lh5_group.startswith("/"):
             lh5_group = lh5_group[1:]
 
@@ -75,7 +76,7 @@ def ls(
 
 
 def show(
-    lh5_file: str | h5py.Group,
+    lh5_file: str | Path | h5py.Group,
     lh5_group: str = "/",
     attrs: bool = False,
     indent: str = "",
@@ -121,8 +122,8 @@ def show(
         return
 
     # open file
-    if isinstance(lh5_file, str):
-        lh5_file = h5py.File(utils.expand_path(lh5_file), "r", locking=False)
+    if isinstance(lh5_file, (str, Path)):
+        lh5_file = h5py.File(utils.expand_path(Path(lh5_file)), "r", locking=False)
 
     # go to group
     if lh5_group != "/":

lgdo/lh5/utils.py

@@ -21,7 +21,7 @@ log = logging.getLogger(__name__)
 
 def get_buffer(
     name: str,
-    lh5_file: str | h5py.File | Sequence[str | h5py.File],
+    lh5_file: str | Path | h5py.File | Sequence[str | Path | h5py.File],
     size: int | None = None,
     field_mask: Mapping[str, bool] | Sequence[str] | None = None,
 ) -> types.LGDO:
@@ -39,7 +39,7 @@ def get_buffer(
     return obj
 
 
-def read_n_rows(name: str, h5f: str | h5py.File) -> int | None:
+def read_n_rows(name: str, h5f: str | Path | h5py.File) -> int | None:
     """Look up the number of rows in an Array-like LGDO object on disk.
 
     Return ``None`` if `name` is a :class:`.Scalar` or a :class:`.Struct`.
@@ -56,7 +56,7 @@ def read_n_rows(name: str, h5f: str | h5py.File) -> int | None:
     return _serializers.read.utils.read_n_rows(h5o, h5f.name, name)
 
 
-def read_size_in_bytes(name: str, h5f: str | h5py.File) -> int | None:
+def read_size_in_bytes(name: str, h5f: str | Path | h5py.File) -> int | None:
     """Look up the size (in B) in an LGDO object in memory. Will crawl
     recursively through members of a Struct or Table
     """
@@ -111,7 +111,7 @@ def get_h5_group(
         grp_attrs is not None
         and len(set(grp_attrs.items()) ^ set(group.attrs.items())) > 0
     ):
-        if not overwrite:
+        if not overwrite and len(group.attrs) != 0:
             msg = (
                 f"Provided {grp_attrs=} are different from "
                 f"existing ones {dict(group.attrs)=} but overwrite flag is not set"
@@ -158,10 +158,10 @@ def expand_vars(expr: str, substitute: dict[str, str] | None = None) -> str:
 
 
 def expand_path(
-    path: str,
+    path: str | Path,
     substitute: dict[str, str] | None = None,
     list: bool = False,
-    base_path: str | None = None,
+    base_path: str | Path | None = None,
 ) -> str | list:
     """Expand (environment) variables and wildcards to return absolute paths.
 
@@ -184,18 +184,26 @@ def expand_path(
         Unique absolute path, or list of all absolute paths
     """
     if base_path is not None and base_path != "":
-        base_path = Path(os.path.expandvars(base_path)).expanduser()
-        path = base_path / path
+        base_path = Path(expand_vars(str(base_path))).expanduser()
+        if not Path(path).expanduser().is_absolute():
+            path = base_path / path
 
     # first expand variables
-    _path = expand_vars(path, substitute)
+    _path = expand_vars(str(path), substitute)
 
     # then expand wildcards
     # pathlib glob works differently so use glob for now
     paths = sorted(glob.glob(str(Path(_path).expanduser())))  # noqa: PTH207
 
     if base_path is not None and base_path != "":
-        paths = [os.path.relpath(p, base_path) for p in paths]
+        rel_paths = []
+        for p in paths:
+            p_path = Path(p)
+            try:
+                rel_paths.append(str(p_path.relative_to(base_path)))
+            except ValueError:
+                rel_paths.append(str(p_path))
+        paths = rel_paths
 
     if not list:
         if len(paths) == 0:

lgdo/types/array.py

@@ -109,14 +109,14 @@ class Array(LGDOCollection):
 
     @property
     def shape(self):
-        return (len(self),) + self._nda.shape[1:]
+        return (len(self), *self._nda.shape[1:])
 
     def reserve_capacity(self, capacity: int) -> None:
         "Set size (number of rows) of internal memory buffer"
         if capacity < len(self):
             msg = "Cannot reduce capacity below Array length"
             raise ValueError(msg)
-        self._nda.resize((capacity,) + self._nda.shape[1:], refcheck=False)
+        self._nda.resize((capacity, *self._nda.shape[1:]), refcheck=False)
 
     def get_capacity(self) -> int:
         "Get capacity (i.e. max size before memory must be re-allocated)"
@@ -190,6 +190,9 @@ class Array(LGDOCollection):
 
         return False
 
+    def __hash__(self):
+        return hash(self.name)
+
     def __iter__(self) -> Iterator:
         yield from self.nda
 

lgdo/types/encoded.py

@@ -92,6 +92,9 @@ class VectorOfEncodedVectors(LGDOCollection):
 
         return False
 
+    def __hash__(self):
+        return hash(self.name)
+
     def reserve_capacity(self, *capacity: int) -> None:
         self.encoded_data.reserve_capacity(*capacity)
         self.decoded_size.reserve_capacity(capacity[0])
@@ -345,6 +348,9 @@ class ArrayOfEncodedEqualSizedArrays(LGDOCollection):
 
         return False
 
+    def __hash__(self):
+        return hash(self.name)
+
     def reserve_capacity(self, *capacity: int) -> None:
         self.encoded_data.reserve_capacity(capacity)
 

lgdo/types/scalar.py

@@ -63,6 +63,9 @@ class Scalar(LGDO):
 
         return False
 
+    def __hash__(self):
+        return hash(self.name)
+
     def __str__(self) -> str:
         attrs = self.getattrs()
         return f"{self.value!s} with attrs={attrs!r}"

lgdo/types/struct.py

@@ -5,7 +5,9 @@ utilities.
 
 from __future__ import annotations
 
+import copy
 import logging
+import re
 from collections.abc import Mapping
 from typing import Any
 
@@ -56,7 +58,21 @@ class Struct(LGDO, dict):
                     # assign
                     super().update({k: v})
 
-        # call LGDO constructor to setup attributes
+        # check the datatype attribute passed by the user and sort the fields
+        # to ensure consistent behavior
+        if attrs is not None and "datatype" in attrs:
+            _attrs = copy.copy(dict(attrs))
+
+            if not _is_struct_datatype(self.datatype_name(), _attrs["datatype"]):
+                msg = (
+                    f"datatype attribute ({self.attrs['datatype']}) is not "
+                    f"compatible with class datatype!"
+                )
+                raise ValueError(msg)
+
+            _attrs["datatype"] = _sort_datatype_fields(_attrs["datatype"])
+            attrs = _attrs
+
         super().__init__(attrs)
 
     def datatype_name(self) -> str:
@@ -64,7 +80,10 @@ class Struct(LGDO, dict):
 
     def form_datatype(self) -> str:
         return (
-            self.datatype_name() + "{" + ",".join([str(k) for k in self.keys()]) + "}"
+            self.datatype_name()
+            + "{"
+            + ",".join(sorted([str(k) for k in self.keys()]))
+            + "}"
         )
 
     def update_datatype(self) -> None:
@@ -157,3 +176,34 @@ class Struct(LGDO, dict):
             "not possible. Call view_as() on the fields instead."
         )
         raise NotImplementedError(msg)
+
+
+def _is_struct_datatype(dt_name, expr):
+    return re.search("^" + dt_name + r"\{(.*)\}$", expr) is not None
+
+
+def _get_struct_fields(expr: str) -> list[str]:
+    assert _is_struct_datatype(".*", expr)
+
+    arr = re.search(r"\{(.*)\}$", expr).group(1).split(",")
+    if arr == [""]:
+        arr = []
+
+    return sorted(arr)
+
+
+def _struct_datatype_equal(dt_name, dt1, dt2):
+    if any(not _is_struct_datatype(dt_name, dt) for dt in (dt1, dt2)):
+        return False
+
+    return _get_struct_fields(dt1) == _get_struct_fields(dt2)
+
+
+def _sort_datatype_fields(expr):
+    assert _is_struct_datatype(".*", expr)
+
+    match = re.search(r"^(.*)\{.*\}$", expr)
+    struct_type = match.group(1)
+    fields = _get_struct_fields(expr)
+
+    return struct_type + "{" + ",".join(sorted([str(k) for k in fields])) + "}"

lgdo/types/table.py

@@ -81,8 +81,9 @@ class Table(Struct, LGDOCollection):
             col_dict = _ak_to_lgdo_or_col_dict(col_dict)
 
         # call Struct constructor
-        Struct.__init__(self, obj_dict=col_dict)
-        LGDOCollection.__init__(self, attrs=attrs)
+        Struct.__init__(self, obj_dict=col_dict, attrs=attrs)
+        # no need to call the LGDOCollection constructor, as we are calling the
+        # Struct constructor already
 
         # if col_dict is not empty, set size according to it
         # if size is also supplied, resize all fields to match it
@@ -329,9 +330,10 @@ class Table(Struct, LGDOCollection):
             :func:`numexpr.evaluate`` as `local_dict` argument or to
             :func:`eval` as `locals` argument.
         modules
-            a dictionary of additional modules used by the expression. If this is not `None`
-            then :func:`eval`is used and the expression can depend on any modules from this dictionary in
-            addition to awkward and numpy. These are passed to :func:`eval` as `globals` argument.
+            a dictionary of additional modules used by the expression. If this
+            is not `None` then :func:`eval`is used and the expression can
+            depend on any modules from this dictionary in addition to awkward
+            and numpy. These are passed to :func:`eval` as `globals` argument.
 
         Examples
         --------
@@ -402,7 +404,10 @@ class Table(Struct, LGDOCollection):
                 return _make_lgdo(out_data)
 
             except Exception:
-                msg = f"Warning {expr} could not be evaluated with numexpr probably due to some not allowed characters, trying with eval()."
+                msg = (
+                    f"Warning {expr} could not be evaluated with numexpr probably "
+                    "due to some not allowed characters, trying with eval()."
+                )
                 log.debug(msg)
 
         # resort to good ol' eval()

lgdo/types/vectorofvectors.py

@@ -284,6 +284,9 @@ class VectorOfVectors(LGDOCollection):
 
         return False
 
+    def __hash__(self):
+        return hash(self.name)
+
     def __getitem__(self, i: int) -> NDArray:
         """Return a view of the vector at index `i` along the first axis."""
         if self.ndim == 2: