reboost 0.4.3__py3-none-any.whl → 0.5.1__py3-none-any.whl

reboost/__init__.py

@@ -1,5 +1,7 @@
 from __future__ import annotations
 
+from lgdo import lh5
+
 from . import build_hit, core, iterator, math, shape
 from ._version import version as __version__
 
@@ -15,3 +17,5 @@ __all__ = [
     "optmap",
     "shape",
 ]
+
+lh5.settings.DEFAULT_HDF5_SETTINGS = {"shuffle": True, "compression": "lzf"}

reboost/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.4.3'
-__version_tuple__ = version_tuple = (0, 4, 3)
+__version__ = version = '0.5.1'
+__version_tuple__ = version_tuple = (0, 5, 1)

reboost/build_evt.py

@@ -102,9 +102,9 @@ def build_evt(
         elif isinstance(info, list):
             channels[group] = info
 
-    for tcm_lh5, _, n_rows_read in LH5Iterator(tcm_file, "tcm", buffer_len=buffer):
+    for tcm_lh5 in LH5Iterator(tcm_file, "tcm", buffer_len=buffer):
         tcm_lh5_sel = tcm_lh5
-        tcm_ak = tcm_lh5_sel.view_as("ak")[:n_rows_read]
+        tcm_ak = tcm_lh5_sel.view_as("ak")
 
         tcm = TCMData(
             id=np.array(ak.flatten(tcm_ak.array_id)),

reboost/build_glm.py

@@ -146,13 +146,14 @@ def get_stp_evtids(
 
     while start_row < n_rows_tot:
         # read the file
-        lh5_obj, n_read = store.read(
+        lh5_obj = store.read(
             f"{lh5_table}/{id_name}",
             stp_file,
             start_row=start_row,
             n_rows=stp_buffer,
         )
         evtids = lh5_obj.view_as("ak")
+        n_read = len(evtids)
 
         # pick the first evtid
         if evtids.ndim > 1:
@@ -247,9 +248,9 @@ def build_glm(
         vfield = f"vtx/{id_name}"
 
         # iterate over the vertex table
-        for vert_obj, vidx, n_evtid in LH5Iterator(stp_file, vfield, buffer_len=evtid_buffer):
+        for vidx, vert_obj in enumerate(LH5Iterator(stp_file, vfield, buffer_len=evtid_buffer)):
             # range of vertices
-            vert_ak = vert_obj.view_as("ak")[:n_evtid]
+            vert_ak = vert_obj.view_as("ak")
 
             msg = f"... read chunk {vidx}"
             log.debug(msg)

reboost/build_hit.py

@@ -41,10 +41,6 @@ A :func:`build_hit` to parse the following configuration file:
              phyvol: OBJECTS.geometry.physical_volume_dict[DETECTOR]
              drift_time_map: lgdo.lh5.read(DETECTOR, ARGS.dtmap_file)
 
-          # this defines "hits", i.e. layout of the output hit table
-          # group steps by time and evtid with 10us window
-          hit_table_layout: reboost.shape.group_by_time(STEPS, window=10)
-
           # finally, the processing chain
           operations:
 
@@ -190,7 +186,6 @@ def build_hit(
     *,
     start_evtid: int = 0,
     n_evtid: int | None = None,
-    in_field: str = "stp",
     out_field: str = "hit",
     buffer: int = int(5e6),
     overwrite: bool = False,
@@ -214,8 +209,6 @@ def build_hit(
         first evtid to read.
     n_evtid
         number of evtid to read, if `None` read all.
-    in_field
-        name of the input field in the remage output.
     out_field
         name of the output field
     buffer
@@ -280,13 +273,17 @@ def build_hit(
                     time_dict=time_dict[proc_name],
                 )
 
+                lh5_group = proc_group.get("lh5_group", "stp")
+                if lh5_group is None:
+                    lh5_group = "/"
+
                 # begin iterating over the glm
                 iterator = GLMIterator(
                     glm_file,
                     stp_file,
                     lh5_group=in_detector,
                     start_row=start_evtid,
-                    stp_field=in_field,
+                    stp_field=lh5_group,
                     n_rows=n_evtid,
                     buffer=buffer,
                     time_dict=time_dict[proc_name],
@@ -396,7 +393,7 @@ def build_hit(
                         lh5.read(obj, stp_file),
                         obj,
                         files.hit[file_idx],
-                        wo_mode="write_safe",
+                        wo_mode="write_safe" if file_idx == 0 else "append",
                     )
                 except LH5EncodeError as e:
                     msg = f"cannot forward object {obj} as it has been already processed by reboost"

reboost/cli.py

@@ -103,7 +103,6 @@ def cli(args=None) -> None:
     hit_parser.add_argument(
         "--n-evtid", type=int, default=None, help="Number of event id to process."
     )
-    hit_parser.add_argument("--in-field", type=str, default="stp", help="Input field name.")
     hit_parser.add_argument("--out-field", type=str, default="hit", help="Output field name.")
     hit_parser.add_argument("--buffer", type=int, default=int(5e6), help="Buffer size.")
 
@@ -174,7 +173,6 @@ def cli(args=None) -> None:
         msg += f"    hit files:      {hit_files}\n"
         msg += f"    start_evtid:    {args.start_evtid}\n"
         msg += f"    n_evtid:        {args.n_evtid}\n"
-        msg += f"    in_field:       {args.in_field}\n"
         msg += f"    out_field:      {args.out_field}\n"
         msg += f"    buffer:         {args.buffer} \n"
         msg += f"    overwrite:      {args.overwrite} \n"
@@ -189,7 +187,6 @@ def cli(args=None) -> None:
             hit_files=hit_files,
             start_evtid=args.start_evtid,
             n_evtid=args.n_evtid,
-            in_field=args.in_field,
             out_field=args.out_field,
             buffer=args.buffer,
             overwrite=args.overwrite,

reboost/iterator.py

@@ -139,12 +139,14 @@ class GLMIterator:
 
         if self.use_glm:
             if self.glm_file is not None:
-                glm_rows, n_rows_read = self.sto.read(
+                glm_rows = self.sto.read(
                     f"glm/{self.lh5_group}",
                     self.glm_file,
                     start_row=self.start_row_tmp,
                     n_rows=n_rows,
                 )
+                n_rows_read = len(glm_rows.view_as("ak"))
+
             else:
                 # get the maximum row to read
                 max_row = self.start_row_tmp + n_rows
@@ -198,12 +200,14 @@ class GLMIterator:
             time_start = time.time()
 
         try:
-            stp_rows, n_steps = self.sto.read(
+            stp_rows = self.sto.read(
                 f"{self.stp_field}/{self.lh5_group}",
                 self.stp_file,
                 start_row=int(start),
                 n_rows=int(n),
             )
+            n_steps = len(stp_rows.view_as("ak"))
+
         except OverflowError:
             raise StopIteration from None
 

reboost/optmap/convolve.py

@@ -356,9 +356,9 @@ def convolve(
 
     log.info("opening energy deposition hit output %s", edep_file)
     it = LH5Iterator(edep_file, edep_path, buffer_len=buffer_len)
-    for it_count, (edep_lgdo, edep_events, edep_n_rows) in enumerate(it):
-        assert (it_count == 0) == (edep_events == 0)
-        edep_df = edep_lgdo.view_as("pd").iloc[0:edep_n_rows].to_records()
+
+    for it_count, edep_lgdo in enumerate(it):
+        edep_df = edep_lgdo.view_as("pd").iloc.to_records()
 
         log.info("start event processing (%d)", it_count)
         output_map = iterate_stepwise_depositions(

reboost/optmap/create.py

@@ -130,9 +130,8 @@ def _create_optical_maps_chunk(
 
     hits_per_primary = np.zeros(10, dtype=np.int64)
     hits_per_primary_len = 0
-    for it_count, (events_lgdo, events_entry, event_n_rows) in enumerate(optmap_events_it):
-        assert (it_count == 0) == (events_entry == 0)
-        optmap_events = events_lgdo.view_as("pd").iloc[0:event_n_rows]
+    for it_count, events_lgdo in enumerate(optmap_events_it):
+        optmap_events = events_lgdo.view_as("pd").iloc
         hitcounts = optmap_events[all_det_ids].to_numpy()
         loc = optmap_events[["xloc", "yloc", "zloc"]].to_numpy()
 

reboost/optmap/evt.py

@@ -80,8 +80,8 @@ def build_optmap_evt(
             _store_vert_df()
 
             # read the next vertex chunk into memory.
-            (vert_lgdo, vert_entry, vert_n_rows) = next(vert_it)
-            vert_df = vert_lgdo.view_as("pd").iloc[0:vert_n_rows]
+            (vert_lgdo, vert_entry) = next(vert_it)
+            vert_df = vert_lgdo.view_as("pd").iloc
 
             # prepare vertex coordinates.
             vert_df = vert_df.set_index("evtid", drop=True).drop(["n_part", "time"], axis=1)
@@ -94,9 +94,8 @@ def build_optmap_evt(
     log.info("prepare evt table")
     # use smaller integer type uint8 to spare RAM when storing types.
     hit_count_type = np.uint8
-    for opti_it_count, (opti_lgdo, opti_entry, opti_n_rows) in enumerate(opti_it):
-        assert (opti_it_count == 0) == (opti_entry == 0)
-        opti_df = opti_lgdo.view_as("pd").iloc[0:opti_n_rows]
+    for opti_it_count, opti_lgdo in enumerate(opti_it):
+        opti_df = opti_lgdo.view_as("pd").iloc
 
         log.info("build evt table (%d)", opti_it_count)
         for t in opti_df[["evtid", "det_uid"]].itertuples(name=None, index=False):

reboost/utils.py

@@ -30,7 +30,9 @@ def get_wo_mode(
         return "overwrite_file" if overwrite else "write_safe"
 
     # if we have a detector not the first and chunk 0 append column
-    if ((in_det > 0) or (out_det > 0)) & (chunk == 0):
+    is_append = ((in_det > 0) or (out_det > 0)) & (chunk == 0)
+    is_append = is_append or (in_det == 0 and out_det == 0 and chunk == 0 and group > 0)
+    if is_append:
         return "append_column"
     return "append"
 

{reboost-0.4.3.dist-info → reboost-0.5.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: reboost
-Version: 0.4.3
+Version: 0.5.1
 Summary: New LEGEND Monte-Carlo simulation post-processing
 Author-email: Manuel Huber <info@manuelhu.de>, Toby Dixon <toby.dixon.23@ucl.ac.uk>, Luigi Pertoldi <gipert@pm.me>
 Maintainer: The LEGEND Collaboration
@@ -700,7 +700,7 @@ Requires-Dist: colorlog
 Requires-Dist: numpy
 Requires-Dist: scipy
 Requires-Dist: numba
-Requires-Dist: legend-pydataobj>=1.11.10
+Requires-Dist: legend-pydataobj>=1.12.0
 Requires-Dist: legend-pygeom-optics>=0.6.5
 Requires-Dist: hist
 Requires-Dist: dbetto

{reboost-0.4.3.dist-info → reboost-0.5.1.dist-info}/RECORD

@@ -1,16 +1,15 @@
-reboost/__init__.py,sha256=2Lc9Idexf1S3IQvHQS1LPD-SCud7Mnmpr2dJ6U-ThvU,303
-reboost/_version.py,sha256=zJwW9_MgFPmVYNh3YnSsdJ4y2EqGvu1bzYeID1Rrd0A,511
-reboost/build_evt.py,sha256=NuI5HRaA4vQApZWPTzfNeUxUhzLfaubLKr6ZT7KWCEI,4809
-reboost/build_glm.py,sha256=-MbPTaUdXlmBml1yzvDB1ddfCh2Vjtvblo3GWo_KzrY,9537
-reboost/build_hit.py,sha256=e0aSNIlbYIsO43jVXTEIJIMHAOGEnkHieWS1XGgpOkc,15247
-reboost/build_tcm.py,sha256=jeuyyVpXYMmvJp2lUKewhbU2diV-_d5fFlkA3Zv2JFU,2956
-reboost/cli.py,sha256=YZvCaxkU0VA_xMAeyw1PN_x17RUkl8uzv19nkNkz3kE,6509
+reboost/__init__.py,sha256=0xtTxM_T8Sp6U3-EJRgFZ6WuqPk31QGQms7jhwouhvs,403
+reboost/_version.py,sha256=HAyT1n_bawwWJ8KZ9BPQ7noEsDs3TEVVzFKchoJw0NQ,511
+reboost/build_evt.py,sha256=TnJRiZ2O6eChorZeh-_SGgWa9jZQYAl_SsmJsM-wYrY,4779
+reboost/build_glm.py,sha256=uYoKl8U5gHy4oW1Neik5F_1jy5yhnFKS1ezVBA-vQaQ,9550
+reboost/build_hit.py,sha256=z3ZpQ18ihOPlurdgOeR471woOgCs6wtZrAN8AWDL9Io,15120
+reboost/cli.py,sha256=HZgqUZK0tSmnlGqoXjrbmLitW_i001TzibxvDrRxLLg,6324
 reboost/core.py,sha256=B577b5KzAYpGI7c4eCxIRKgt9tq6hwVr2-DtYTml1e8,12826
-reboost/iterator.py,sha256=YFsNjC2Onsyhvkogbs79eD8iqstCdCAfFEaEOMr_ujg,6810
+reboost/iterator.py,sha256=J--4f7ANW15Nb4kL_OQ0kbIw7U0ks73asefLd-uGoV4,6898
 reboost/log_utils.py,sha256=VqS_9OC5NeNU3jcowVOBB0NJ6ssYvNWnirEY-JVduEA,766
 reboost/profile.py,sha256=EOTmjmS8Rm_nYgBWNh6Rntl2XDsxdyed7yEdWtsZEeg,2598
 reboost/units.py,sha256=3EH8XlpbsObdu5vLgxhm1600L6UNYD5jng4SjJT_1QE,2202
-reboost/utils.py,sha256=MjnZNXP80gD3DGFfNEdT-WyiTsA1GRMElYsoW5hX4v8,9705
+reboost/utils.py,sha256=0xU68p0YSnDKTyLdso8zX4JYrGTVCjIiTBx7cuSCz98,9820
 reboost/hpge/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 reboost/hpge/psd.py,sha256=jAUAoQ_PMz76wyA1NXYHNKtOwoCnRT3My8_LCFrKi-U,13860
 reboost/hpge/surface.py,sha256=lbWcFnFFWKxtFKs755GyM9US_IfyxaoM6MpOIZgIMM0,7478
@@ -20,9 +19,9 @@ reboost/math/functions.py,sha256=OymiYTcA0NXxxm-MBDw5kqyNwHoLCmuv4J48AwnSrbU,563
 reboost/math/stats.py,sha256=iiOEi87x93kqPWeSmlRiA5Oe-R8XR-plm6Z532PhC9M,1401
 reboost/optmap/__init__.py,sha256=imvuyld-GLw8qdwqW-lXCg2feptcTyQo3wIzPvDHwmY,93
 reboost/optmap/cli.py,sha256=SzbPRgsbR5Llm3aSJubH02Ym8FQyTH7kvuLjK7faLiY,9572
-reboost/optmap/convolve.py,sha256=jCH_d04yioB8hsJEPunm0zynA0ne4lx0ldSC1GJG_eY,14129
-reboost/optmap/create.py,sha256=Nm5-xEe8M9q2GFQnUv8oN8qpAz9nZArIrQcPboqRmCQ,17153
-reboost/optmap/evt.py,sha256=9rfAdN9MqL6UuUxUcMDGVwpcuqRVc2RwmEmd87jgYww,4698
+reboost/optmap/convolve.py,sha256=vj5Wl2M5BeyvLbcnLo8xzaOAIKXUHUu023sXIP7xTCI,14034
+reboost/optmap/create.py,sha256=gd9VWceU0AOX_ABuLvmnAi2GkUdiugNF6CMUT2CbDbE,17053
+reboost/optmap/evt.py,sha256=CRhUXpgnV3nTA6gY-e6V2W2C4K6nJ8AEoW2ffa2ww9s,4571
 reboost/optmap/mapview.py,sha256=73kpe0_SKDj9bIhEx1ybX1sBP8TyvufiLfps84A_ijA,6798
 reboost/optmap/numba_pdg.py,sha256=y8cXR5PWE2Liprp4ou7vl9do76dl84vXU52ZJD9_I7A,731
 reboost/optmap/optmap.py,sha256=j4rfbQ84PYSpE-BvP4Rdt96ZjPdwy8P4e4eZz1mATys,12817
@@ -30,9 +29,9 @@ reboost/shape/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 reboost/shape/cluster.py,sha256=RIvBlhHzp88aaUZGofp5SD9bimnoiqIOddhQ84jiwoM,8135
 reboost/shape/group.py,sha256=_z2qCOret3E-kj-nrp1-J5j2lEwQpgfYdQp2pgpDHR8,4449
 reboost/shape/reduction.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-reboost-0.4.3.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-reboost-0.4.3.dist-info/METADATA,sha256=KeW1VoCUjCxRuz8nRD5gnkYF-2Z4H53x87V586TmUwM,44251
-reboost-0.4.3.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-reboost-0.4.3.dist-info/entry_points.txt,sha256=DxhD6BidSWNot9BrejHJjQ7RRLmrMaBIl52T75oWTwM,93
-reboost-0.4.3.dist-info/top_level.txt,sha256=q-IBsDepaY_AbzbRmQoW8EZrITXRVawVnNrB-_zyXZs,8
-reboost-0.4.3.dist-info/RECORD,,
+reboost-0.5.1.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+reboost-0.5.1.dist-info/METADATA,sha256=P9dBK9bEPeiEEzYWtrzjiDf5q14NKG9GRCYzLYT0piU,44250
+reboost-0.5.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+reboost-0.5.1.dist-info/entry_points.txt,sha256=DxhD6BidSWNot9BrejHJjQ7RRLmrMaBIl52T75oWTwM,93
+reboost-0.5.1.dist-info/top_level.txt,sha256=q-IBsDepaY_AbzbRmQoW8EZrITXRVawVnNrB-_zyXZs,8
+reboost-0.5.1.dist-info/RECORD,,

reboost/build_tcm.py

@@ -1,111 +0,0 @@
-from __future__ import annotations
-
-import logging
-import re
-
-import awkward as ak
-from lgdo import Table, lh5
-
-from .shape import group
-
-log = logging.getLogger(__name__)
-
-
-def build_tcm(
-    hit_file: str,
-    out_file: str,
-    channels: list[str],
-    time_name: str = "t0",
-    idx_name: str = "global_evtid",
-    time_window_in_us: float = 10,
-) -> None:
-    """Build the (Time Coincidence Map) TCM from the hit tier.
-
-    Parameters
-    ----------
-    hit_file
-        path to hit tier file.
-    out_file
-        output path for tcm.
-    channels
-        list of channel names to include.
-    time_name
-        name of the hit tier field used for time grouping.
-    idx_name
-        name of the hit tier field used for index grouping.
-    time_window_in_us
-        time window used to define the grouping.
-    """
-    hash_func = r"\d+"
-
-    msg = "start building time-coincidence map"
-    log.info(msg)
-
-    chan_ids = [re.search(hash_func, channel).group() for channel in channels]
-
-    hit_data = []
-    for channel in channels:
-        hit_data.append(
-            lh5.read(f"{channel}/hit", hit_file, field_mask=[idx_name, time_name]).view_as("ak")
-        )
-    tcm = get_tcm_from_ak(
-        hit_data, chan_ids, window=time_window_in_us, time_name=time_name, idx_name=idx_name
-    )
-
-    if tcm is not None:
-        lh5.write(tcm, "tcm", out_file, wo_mode="overwrite_file")
-
-
-def get_tcm_from_ak(
-    hit_data: list[ak.Array],
-    channels: list[int],
-    *,
-    window: float = 10,
-    time_name: str = "t0",
-    idx_name: str = "global_evtid",
-) -> Table:
-    """Builds a time-coincidence map from a hit of hit data Tables.
-
-    - build an ak.Array of the data merging channels with fields base on "time_name", and "idx_name" and adding a field `rawid` from the channel idx, also add the row (`hit_idx`)
-    - sorts this array by "idx_name" then "time_name" fields
-    - group by "idx_name" and "time_name" based on the window parameter
-
-    Parameters
-    ----------
-    hit_data
-        list of hit tier data for each channel
-    channels
-        list of channel indices
-    window
-        time window for selecting coincidences (in us)
-    time_name
-        name of the field for time information
-    idx_name
-        name of the decay index field
-
-    Returns
-    -------
-    an LGDO.VectorOfVectors containing the time-coincidence map
-    """
-    # build ak_obj for sorting
-    sort_objs = []
-
-    for ch_idx, data_tmp in zip(channels, hit_data):
-        obj_tmp = ak.copy(data_tmp)
-        obj_tmp = obj_tmp[[time_name, idx_name]]
-        hit_idx = ak.local_index(obj_tmp)
-
-        obj_tmp = ak.with_field(obj_tmp, hit_idx, "array_idx")
-
-        obj_tmp["array_id"] = int(ch_idx)
-        sort_objs.append(obj_tmp)
-
-    obj_tot = ak.concatenate(sort_objs)
-
-    return group.group_by_time(
-        obj_tot,
-        time_name=time_name,
-        evtid_name=idx_name,
-        window=window,
-        fields=["array_id", "array_idx"],
-    )