default_normalizing_dataframes = [None, None]

main_and_diff_distributions = [[], []]

already_normalized = [[], []]

settings = []

has_bands = False

# use axis labels as data column headers

x_title = read_configuration_value(config, key='x_title', index=histogram_index_or_none)

y_titles = read_configuration_subplot_values(config, key='y_title', defaults=['__no_default__', '__use_main__'], index=histogram_index_or_none)

# condition will be used as legend title by tsplot

condition = read_configuration_value(config, key='legend_title', default='')

bin_style = read_configuration_value(config, key='bin_style', default='step')

centered = bin_style == 'center'

x_lim = read_configuration_value(config, key='x_lim', default=None, index=histogram_index_or_none)

if read_configuration_value(config, key='err_estimator', default='asymmetric_hessian_error') == 'asymmetric_hessian_error':

err_estimator = asymmetric_hessian_error

else:

err_estimator = standard_error

data_path = read_configuration_value(config, key='data_path', default=".")

config_distributions = read_configuration_value(config, key='distributions')

for distribution in config_distributions:

main_and_diff_normalized_dataframe_or_none, distribution_has_bands, \

main_and_diff_distributions_are_normalized = read_distribution(data_path,

distribution,

centered,

condition,

x_title,

y_titles,

x_lim,

histogram_or_empty=histogram_or_empty,

err_estimator=err_estimator,

settings=settings,

main_and_diff_distributions=main_and_diff_distributions,

config_distributions=config_distributions)[1:]

for i in range(2):

if main_and_diff_normalized_dataframe_or_none[i] is not None:

default_normalizing_dataframes[i] = main_and_diff_normalized_dataframe_or_none[i]

already_normalized[i].append(main_and_diff_distributions_are_normalized[i])

if distribution_has_bands:

has_bands = distribution_has_bands

for i, default_normalizing_dataframe in enumerate(default_normalizing_dataframes):

if default_normalizing_dataframe is not None:

for distribution, is_already_normalized in zip(main_and_diff_distributions[i], already_normalized[i]):

if not is_already_normalized:

for df in distribution:

df[y_titles[i]] /= default_normalizing_dataframe[y_titles[i]]

for i in range(2):

main_and_diff_distributions[i] = [pd.concat(dataframes) if len(dataframes) else None for dataframes in main_and_diff_distributions[i]]