def save_run_info(h5out: tb.file.File, run_number: int) -> None:
"""
Saves the run number used for the detsim
format job in the format expected by the
IC cities.
h5out : pytables file
The open output file
run_number : int
The run number set in the config
"""
try:
run_table = getattr(h5out.root.Run, 'runInfo')
except tb.NoSuchNodeError:
try:
run_group = getattr(h5out.root, 'Run')
run_table = h5out.create_table(run_group, "runInfo", RunInfo,
"Run number used in detsim")
except tb.NoSuchNodeError:
run_group = h5out.create_group(h5out.root, 'Run')
run_table = h5out.create_table(run_group, "runInfo", RunInfo,
"Run number used in detsim")
row = run_table.row
row["run_number"] = run_number
row.append()
def copy_sensor_table(h5in_name: str, h5out: tb.file.File):
with tb.open_file(h5in_name) as dIn:
try:
sensor_info = dIn.root.Sensors
h5out.copy_node(sensor_info, newparent=h5out.root, recursive=True)
except tb.exceptions.NoSuchNodeError:
sensor_info = None
def rwf_writer(h5out: tb.file.File,
*,
group_name: str,
table_name: str,
compression: str = 'ZLIB4',
n_sensors: int,
waveform_length: int) -> Callable:
"""
Defines group and table where raw waveforms
will be written.
h5out : pytables file
Output file where waveforms to be saved
group_name : str
Name of the group in h5in.root
Known options: RD, BLR
Setting to None will save directly in root
table_name : str
Name of the table
Known options: pmtrwf, pmtcwf, sipmrwf
compression : str
file compression
n_sensors : int
number of sensors in the table (shape[0])
waveform_length : int
Number of samples per sensor
"""
if group_name is None:
rwf_group = h5out.root
elif group_name in h5out.root:
rwf_group = getattr(h5out.root, group_name)
else:
rwf_group = h5out.create_group(h5out.root, group_name)
rwf_table = h5out.create_earray(rwf_group,
table_name,
atom=tb.Int16Atom(),
shape=(0, n_sensors, waveform_length),
filters=tbl.filters(compression))
def write_rwf(waveform: np.ndarray) -> None:
"""
Writes raw waveform arrays to file.
waveform : np.ndarray
shape = (n_sensors, waveform_length) array
of sensor charge.
"""
rwf_table.append(waveform.reshape(1, n_sensors, waveform_length))
return write_rwf
def save_run_info(h5out : tb.file.File,
run_number : int) -> None:
try:
run_table = getattr(h5out.root.Run, 'runInfo')
except tb.NoSuchNodeError:
try:
run_group = getattr(h5out.root, 'Run')
run_table = h5out.create_table(run_group, "runInfo", RunInfo,
"Run number used in detsim")
except tb.NoSuchNodeError:
run_group = h5out.create_group(h5out.root, 'Run')
run_table = h5out.create_table(run_group, "runInfo", RunInfo,
"Run number used in detsim")
row = run_table.row
row["run_number"] = run_number
row.append()
def _store_pandas_as_tables(h5out: tb.file.File,
df: pd.DataFrame,
group_name: str,
table_name: str,
compression: str = 'ZLIB4',
descriptive_string: [str] = "",
str_col_length: int = 32) -> None:
if len(df) == 0:
warnings.warn(f'dataframe is empty', UserWarning)
if group_name not in h5out.root:
group = h5out.create_group(h5out.root, group_name)
group = getattr(h5out.root, group_name)
if table_name not in group:
tabledef = _make_tabledef(df.dtypes)
table = make_table(h5out,
group=group_name,
name=table_name,
fformat=tabledef,
description=descriptive_string,
compression=compression)
table = getattr(group, table_name)
if not np.array_equal(df.columns, table.colnames):
raise TableMismatch(
f'dataframe differs from already existing table structure')
for indx in df.index:
tablerow = table.row
for colname in table.colnames:
tablerow[colname] = df.at[indx, colname]
tablerow.append()
table.flush()
def df_writer(h5out: tb.file.File,
df: pd.DataFrame,
group_name: str,
table_name: str,
compression: str = 'ZLIB4',
descriptive_string: str = "",
str_col_length: int = 32,
columns_to_index: Optional[Sequence[str]] = None) -> None:
""" The function writes a dataframe to open pytables file.
Parameters:
h5out : open pytable file for writing
df : DataFrame to be written
group_name : group name where table is to be saved)
(group is created if doesnt exist)
table_name : table name
(table is created if doesnt exist)
compression : compression type
descriptive_string : table description
str_col_length : maximum length in characters of strings
columns_to_index : list of columns to be flagged for indexing
"""
if group_name not in h5out.root:
group = h5out.create_group(h5out.root, group_name)
group = getattr(h5out.root, group_name)
arr = df.to_records(index=False)
if table_name not in group:
tabledef = _make_tabledef(arr.dtype, str_col_length=str_col_length)
table = make_table(h5out,
group=group_name,
name=table_name,
fformat=tabledef,
description=descriptive_string,
compression=compression)
else:
table = getattr(group, table_name)
data_types = table.dtype
if len(arr) == 0:
warnings.warn(f'dataframe is empty', UserWarning)
else:
_check_castability(arr, data_types)
columns = list(data_types.names)
arr = arr[columns].astype(data_types)
table.append(arr)
table.flush()
if columns_to_index is not None:
if set(columns_to_index).issubset(set(df.columns)):
table.set_attr('columns_to_index', columns_to_index)
else:
not_found = list(set(columns_to_index).difference(set(df.columns)))
raise KeyError(f'columns {not_found} not present in the dataframe')
def store_events_reco_counters(oFile: tb.file.File, group_name: str,
simulated_events: int, stored_events: int,
smE_filter_events: int,
fid_filter_events: int) -> None:
"""
Stores the event counters as attributes of oFile / group_name
"""
oFile.set_node_attr(group_name, 'simulated_events', simulated_events)
oFile.set_node_attr(group_name, 'stored_events', stored_events)
oFile.set_node_attr(group_name, 'smE_filter_events', smE_filter_events)
oFile.set_node_attr(group_name, 'fid_filter_events', fid_filter_events)
def store_events_ana_counters(oFile: tb.file.File, group_name: str,
events_df: pd.DataFrame) -> Tuple[int, int, int]:
"""
Stores the event counters as attributes of oFile / group_name
"""
tracks_filter_events = events_df.tracks_filter.sum()
blobs_filter_events = events_df.blobs_filter.sum()
roi_filter_events = events_df.roi_filter.sum()
oFile.set_node_attr(group_name, 'tracks_filter_events',
tracks_filter_events)
oFile.set_node_attr(group_name, 'blobs_filter_events', blobs_filter_events)
oFile.set_node_attr(group_name, 'roi_filter_events', roi_filter_events)
return tracks_filter_events, blobs_filter_events, roi_filter_events