legend-pydataobj 1.11.3__tar.gz → 1.11.5__tar.gz

{legend_pydataobj-1.11.3 → legend_pydataobj-1.11.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: legend_pydataobj
-Version: 1.11.3
+Version: 1.11.5
 Summary: LEGEND Python Data Objects
 Author: The LEGEND Collaboration
 Maintainer: The LEGEND Collaboration

{legend_pydataobj-1.11.3 → legend_pydataobj-1.11.5}/src/legend_pydataobj.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: legend_pydataobj
-Version: 1.11.3
+Version: 1.11.5
 Summary: LEGEND Python Data Objects
 Author: The LEGEND Collaboration
 Maintainer: The LEGEND Collaboration

{legend_pydataobj-1.11.3 → legend_pydataobj-1.11.5}/src/lgdo/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '1.11.3'
-__version_tuple__ = version_tuple = (1, 11, 3)
+__version__ = version = '1.11.5'
+__version_tuple__ = version_tuple = (1, 11, 5)

{legend_pydataobj-1.11.3 → legend_pydataobj-1.11.5}/src/lgdo/lh5/_serializers/write/array.py

@@ -64,6 +64,11 @@ def _h5_write_array(
         if "hdf5_settings" in obj.attrs:
             h5py_kwargs |= obj.attrs["hdf5_settings"]
 
+        # HACK: a tuple is strictly requested for the "chunks" setting, but
+        # we'd like to pass a list too in some situations
+        if "chunks" in h5py_kwargs and isinstance(h5py_kwargs["chunks"], list):
+            h5py_kwargs["chunks"] = tuple(h5py_kwargs["chunks"])
+
         # create HDF5 dataset
         ds = group.create_dataset(name, data=nda, **h5py_kwargs)
 

{legend_pydataobj-1.11.3 → legend_pydataobj-1.11.5}/src/lgdo/types/vectorofvectors.py

@@ -13,6 +13,7 @@ import awkward as ak
 import awkward_pandas as akpd
 import numpy as np
 import pandas as pd
+from numba import jit
 from numpy.typing import ArrayLike, DTypeLike, NDArray
 
 from .. import utils
@@ -560,18 +561,16 @@ class VectorOfVectors(LGDO):
             compatible one.
         """
         if self.ndim == 2:
-            ak_arr = self.view_as("ak")
-
             if max_len is None:
-                max_len = int(ak.max(ak.count(ak_arr, axis=-1)))
-
-            nda = ak.fill_none(
-                ak.pad_none(ak_arr, max_len, clip=True), fill_val
-            ).to_numpy(allow_missing=False)
-
+                lens = np.copy(self.cumulative_length)
+                lens[1:] = lens[1:] - lens[:-1]
+                max_len = int(np.max(lens))
+            nda = np.full((len(self), max_len), fill_val)
             if preserve_dtype:
                 nda = nda.astype(self.flattened_data.dtype, copy=False)
 
+            _to_aoesa(self.flattened_data.nda, self.cumulative_length.nda, nda)
+
             return aoesa.ArrayOfEqualSizedArrays(nda=nda, attrs=self.getattrs())
 
         raise NotImplementedError
@@ -664,3 +663,11 @@ class VectorOfVectors(LGDO):
 
         msg = f"{library} is not a supported third-party format."
         raise ValueError(msg)
+
+
+@jit
+def _to_aoesa(flattened_array, cumulative_length, nda):
+    prev_cl = 0
+    for i, cl in enumerate(cumulative_length):
+        nda[i, : (cl - prev_cl)] = flattened_array[prev_cl:cl]
+        prev_cl = cl

{legend_pydataobj-1.11.3 → legend_pydataobj-1.11.5}/tests/lh5/test_lh5_write.py

@@ -44,6 +44,19 @@ def test_write_with_hdf5_compression(lgnd_file, tmptestdir):
         assert h5f["/geds/raw/waveform/values"].compression is None
         assert h5f["/geds/raw/waveform/values"].shuffle is False
 
+    store.write(
+        wft.values,
+        "/geds/raw/waveform/values",
+        f"{tmptestdir}/tmp-pygama-hdf5-compressed-wfs.lh5",
+        wo_mode="overwrite_file",
+        chunks=[1, 10],
+        compression=None,
+        shuffle=False,
+    )
+    with h5py.File(f"{tmptestdir}/tmp-pygama-hdf5-compressed-wfs.lh5") as h5f:
+        assert h5f["/geds/raw/waveform/values"].compression is None
+        assert h5f["/geds/raw/waveform/values"].shuffle is False
+
 
 def test_write_empty_vov(tmptestdir):
     vov = types.VectorOfVectors(flattened_data=[], cumulative_length=[])