def configure(self):
# use this to count through the single events
self.event_idx = 0
# get the original file to compare to
filename = data_path(
"offline/mcv6.0.gsg_muon_highE-CC_50-500GeV.km3sim.jterbr00008357.jorcarec.aanet.905.root"
)
self.f = km3io.OfflineReader(filename)
def test_apply_to_hits_from_km3io_iterator(self):
calib = Calibration(filename=data_path("detx/km3net_offline.detx"))
f = km3io.OfflineReader(data_path("offline/km3net_offline.root"))
for event in f:
chits = calib.apply(event.hits)
assert 176 == len(chits.t0)
assert np.allclose(
[207747.825, 207745.656, 207743.836], chits.t0.tolist()[:3]
)
break
def test_apply_to_hits_from_km3io(self):
calib = Calibration(filename=data_path("detx/km3net_offline.detx"))
hits = km3io.OfflineReader(data_path("offline/km3net_offline.root"))[0].hits
chits = calib.apply(hits)
assert 176 == len(chits.t0)
assert np.allclose([207747.825, 207745.656, 207743.836], chits.t0.tolist()[:3])
chits = calib.apply(hits[:3])
assert 3 == len(chits.t0)
assert np.allclose([207747.825, 207745.656, 207743.836], chits.t0.tolist()[:3])
def configure(self):
self._filename = self.get("filename")
step_size = self.get("step_size", default=2000)
self._reader = km3io.OfflineReader(self._filename, step_size=step_size)
self.header = self._reader.header
self.blobs = self._blob_generator()
Provenance().record_input(self._filename,
uuid=str(self._reader.uuid),
comment="OfflinePump input")
self.expose(self.header, "offline_header")
def test_apply_to_hits_with_pmt_id_aka_mc_hits_from_km3io(self):
calib = Calibration(filename=data_path("detx/KM3NeT_-00000001_20171212.detx"))
f = km3io.OfflineReader(
data_path(
"offline/mcv6.gsg_nue-CCHEDIS_1e4-1e6GeV.sirene.jte.jchain.aanet.1.root"
)
)
for event in f:
chits = calib.apply(event.mc_hits)
assert 840 == len(chits.t0)
assert np.allclose([3, 26, 24, 4, 23, 25], chits.channel_id[:6])
assert np.allclose([3401, 3401, 3406, 3411, 5501, 5501], chits.dom_id[:6])
assert np.allclose([1, 1, 6, 11, 1, 1], chits.floor[:6])
assert np.allclose([34, 34, 34, 34, 55, 55], chits.du[:6])
assert np.allclose(
[
1679.18706571,
1827.14262054,
1926.71722628,
2433.83097585,
1408.35942832,
1296.51397496,
],
chits.time[:6],
)
assert np.allclose(
[2.034, 1.847, 1.938, 2.082, -54.96, -55.034], chits.pos_x[:6]
)
assert np.allclose(
[-233.415, -233.303, -233.355, -233.333, -341.346, -341.303],
chits.pos_y[:6],
)
assert np.allclose(
[65.059, 64.83, 244.83, 425.111, 64.941, 64.83], chits.pos_z[:6]
)
assert np.allclose([4, 4, 4, 26, 4, 4], f.mc_hits.origin[0][:6].tolist())
assert np.allclose(
[36835, 36881, 37187, 37457, 60311, 60315],
f.mc_hits.pmt_id[0][:6].tolist(),
)
break
def test_conversion_of_km3io_header(self):
header = km3io.OfflineReader(data_path("offline/numucc.root")).header
tab = header2table(header)
print(tab)
for p in [
b"DAQ",
b"PDF",
b"can",
b"can_user",
b"coord_origin",
b"cut_in",
b"cut_nu",
b"cut_primary",
b"cut_seamuon",
b"decay",
b"detector",
b"drawing",
b"genhencut",
b"genvol",
b"kcut",
b"livetime",
b"model",
b"ngen",
b"norma",
b"nuflux",
b"physics",
b"seed",
b"simul",
b"sourcemode",
b"spectrum",
b"start_run",
b"target",
b"usedetfile",
b"xlat_user",
b"xparam",
b"zed_user",
]:
assert p in tab.parameter
h5header = HDF5Header.from_table(tab)
assert h5header.can.zmin == header.can.zmin
def h5extractf(root_file,
outfile=None,
without_full_reco=False,
without_calibration=False):
if without_calibration:
calibration_fields = []
else:
calibration_fields = [
"pos_x",
"pos_y",
"pos_z",
"dir_x",
"dir_y",
"dir_z",
"tdc",
]
fields = {
"event_info": [
("id", "event_id"), # id gets renamed to event_id
"run_id",
("t_sec", "timestamp"),
("t_ns", "nanoseconds"),
("mc_t", "mc_time"),
"trigger_mask",
"trigger_counter",
"overlays",
"det_id",
"frame_index",
"mc_run_id",
],
# weights get put into event_info as well
"event_info_weights": [
"weight_w1",
"weight_w2",
"weight_w3",
"weight_w4",
],
"hits": [
"channel_id",
"dom_id",
("t", "time"),
"tot",
("trig", "triggered"),
*calibration_fields,
],
"mc_hits": [
"a",
"origin",
"pmt_id",
("t", "time"),
],
"tracks": [
"pos_x",
"pos_y",
"pos_z",
"dir_x",
"dir_y",
"dir_z",
"E",
"t",
("len", "length"),
"rec_type",
("lik", "likelihood"),
"id",
],
"mc_tracks": [
"pos_x",
"pos_y",
"pos_z",
"dir_x",
"dir_y",
"dir_z",
("E", "energy"),
("t", "time"),
("len", "length"),
"pdgid",
"id",
],
}
if outfile is None:
outfile = root_file + ".h5"
start_time = time.time()
with h5py.File(outfile, "w") as f:
_uuid = str(uuid4())
Provenance().record_output(outfile,
uuid=_uuid,
comment="Converted HDF5 file")
with km3io.OfflineReader(root_file) as r:
Provenance().record_input(root_file,
uuid=str(r.uuid),
comment="Input ROOT file")
if r.header is not None:
print("Processing header")
f.create_dataset(
"raw_header",
data=kp.io.hdf5.header2table(r.header),
)
print("Processing event_info")
np_event_info = _branch_to_numpy(r, fields["event_info"])
np_weights = _ak_to_numpy(r.w, fields["event_info_weights"])
np_event_info[0].update(np_weights[0])
np_w2 = _parse_w2list(r)
if np_w2 is not None:
np_event_info[0].update(np_w2[0])
_to_hdf(f, "event_info", np_event_info)
# TODO remove group_info once km3pipe does not require it anymore
group_info = np.core.records.fromarrays(
[np.arange(len(np_event_info[1]))], names=["group_id"])
f.create_dataset("group_info", data=group_info)
print("Processing tracks")
reco = f.create_group("reco")
for branch_data in _yield_tracks(
r.tracks,
fields["tracks"],
without_full_reco=without_full_reco):
_to_hdf(reco, *branch_data)
for field_name in ("hits", "mc_hits", "mc_tracks"):
if r[field_name] is None:
continue
print("Processing", field_name)
np_branch = _branch_to_numpy(r[field_name], fields[field_name])
if np_branch[1].sum() == 0:
# empty branch, e.g. mc_hits for data files
continue
_to_hdf(f, field_name, np_branch)
f.attrs.create("format_version", FORMAT_VERSION)
f.attrs.create("km3pipe", kp.__version__)
f.attrs.create("origin", root_file)
f.attrs.create("kid", _uuid)
print("Completed in {:.1f} s".format(time.time() - start_time))
