def make_dirs(dir_to_make):
"""
Helper function to checking and making a directory
"""
dir_to_make = expanduser(dir_to_make)
if not exists(dir_to_make):
get_logger().info("Make directory %s", dir_to_make)
makedirs(dir_to_make)
def parse_yaml(filepath):
"""
Parse a YAML file and return dictionary
Args:
filepath: Path to the YAML file to be parsed.
"""
if not os.path.isfile(filepath):
get_logger().critical("YAML file %s does not exist.", filepath)
with open(filepath) as f:
return yaml.safe_load(f)
def parallelizer(function, argument_list, maxperchunk, max_n_procs=2):
"""
A centralized version for quickly parallelizing a function.
"""
get_logger().info("Parallelizing function %s", function.__name__)
chunks = [argument_list[x:x+maxperchunk] \
for x in range(0, len(argument_list), maxperchunk)]
for chunk in chunks:
get_logger().info("Processing new chunck size = %i", maxperchunk)
pool = mp.Pool(max_n_procs)
_ = [
pool.apply_async(function, args=chunk[i])
for i in range(len(chunk))
]
pool.close()
pool.join()
def __init__(self, config):
self.logger = get_logger()
self.config = config
self.data = {}
self.read()
def __init__(self, dimensions):
self.logger = get_logger()
self.coord_columns = range(dimensions + 1)
# TODO Add errors
#self.error_columns = range(dimensions + 1, 3 * (dimensions + 1))
self.dimensions = dimensions
self.df = pd.DataFrame(columns=self.coord_columns)
def __init__(self, config):
self.data_manager = DataManager(config)
self.config = config
self.logger = get_logger()
# TODO Put defaults in a dictionary
self.default_font_size = 50
self.default_figure_width = 40
self.default_figure_height = 30
self.default_dpi = 100
self.default_x_axis_label = "x_axis"
self.default_y_axis_label = "y_axis"
def __init__(self, config: dict):
self.logger = get_logger()
# Static config variables
self.delim = "::"
self.this_flag = "THIS"
self.var_flag = "VAR"
self.sources_flag = "SOURCES"
# The raw config dictionary
self.config_raw = parse_yaml(config)
# The digested configuration
self.variables = None
self.sources = None
self.plots = None
self.check_config()
self.digest()
def digest_root_source(**kwargs):
# NOTE TODO Can for now only extract TH1 and TH2
logger = get_logger()
def digest_1d(histo):
to_be_added = []
axis = histo.GetXaxis()
get_coord = None
if axis.IsAlphanumeric():
get_coord = axis.GetBinLabel
else:
get_coord = axis.getBinCenter
for ibin in range(1, axis.GetNbins() + 1):
coord = get_coord(ibin)
if coord == "":
continue
to_be_added.append(((coord,), histo.GetBinContent(ibin)))
return to_be_added
def digest_2d(histo):
to_be_added = []
x_axis = histo.GetXaxis()
y_axis = histo.GetYaxis()
get_coord_x = None
get_coord_y = None
if x_axis.IsAlphanumeric():
get_coord_x = x_axis.GetBinLabel
else:
get_coord_x = x_axis.getBinCenter
if y_axis.IsAlphanumeric():
get_coord_y = y_axis.GetBinLabel
else:
get_coord_y = y_axis.getBinCenter
for ibin in range(1, x_axis.GetNbins() + 1):
for jbin in range(1, y_axis.GetNbins() + 1):
coord_x = get_coord_x(ibin)
coord_y = get_coord_y(ibin)
if coord_x == "" or coord_y == "":
continue
to_be_added.append(((coord_x, coord_y,), histo.GetBinContent(ibin, jbin)))
return to_be_added
files = kwargs.pop("files", None)
in_root_dir = kwargs.pop("dir", None)
name = kwargs.pop("name", None)
if None in [files, in_root_dir, name]:
logger.fatal("Not all fields provided")
if kwargs:
logger.warning("There are unknown arguments when digesting ROOT file which will be ignored")
print_dict(kwargs)
# TODO For now treating everything as 1d histogram
data = None
for f in files:
root_file = TFile.Open(f, "READ")
root_dir = root_file.Get(in_root_dir)
histo = root_dir.Get(name)
if isinstance(histo, TH2):
if data is None:
data = DataSource(2)
to_be_added = digest_2d(histo)
else:
if data is None:
data = DataSource(1)
to_be_added = digest_1d(histo)
data.extend(to_be_added)
return data