def fit(args):
config_fp = args.config_fp
config = get_config_parser()
print('reading configuration file path: {}'.format(config_fp))
with open(config_fp) as config_file:
config.read_file(config_file)
level = config.get('log', 'level', fallback='INFO')
logging.basicConfig(level=logging.getLevelName(level))
log = logging.getLogger(name=__name__)
# log the configuration file
with io.StringIO() as string_buffer:
config.write(string_buffer)
log.info('configuration file contents:\n{}'.format(
string_buffer.getvalue()))
try:
fitter = build_fit_task(config)
plots_pdf_file_dir = os.path.expanduser(
config.get('output', 'plots_pdf_dir', fallback=None))
if not os.path.exists(plots_pdf_file_dir):
log.info('output directory "%s" will be created',
plots_pdf_file_dir)
os.makedirs(plots_pdf_file_dir)
fitter.fit_all(plots_pdf_dp=plots_pdf_file_dir)
except ConfigurationFileError as e:
log.error(e)
log.exception('configuration file error in "%s"', config_fp)
quit()
table_file_path = os.path.expanduser(
config.get('output', 'table_fp', fallback=None))
if table_file_path:
log.info(
'writing table output to file path {}'.format(table_file_path))
fitter.write_table(table_file_path)
else:
log.warning('No file path specified for table output')
#plots_pdf_file_path = os.path.expanduser(config.get('output', 'plots_pdf_fp', fallback=None))
#if plots_pdf_file_path:
# log.info('writing plots to PDF {}'.format(plots_pdf_file_path))
# fitter.draw_plots_matplotlib(plots_pdf_file_path)
# plots_html_file_path = os.path.splitext(plots_pdf_file_path)[0] + '.html'
# log.info('writing plots to HTML {}'.format(plots_html_file_path))
# fitter.draw_plots_bokeh(plots_html_file_path)
#else:
# log.warning('No file path specified for plot output')
best_fit_files_dir_path = os.path.expanduser(
config.get('output', 'best_fit_files_dir', fallback=None))
if best_fit_files_dir_path:
if not os.path.expanduser(best_fit_files_dir_path):
log.info('creating directory %s', best_fit_files_dir_path)
os.makedirs(best_fit_files_dir_path)
fitter.write_best_fit_arrays(best_fit_files_dir_path)
else:
log.warning('No directory specified for best fit files')
def test_get_fit_parameters_from_config_file():
fit_config = fit_task_builder.get_config_parser()
fit_config.read_string("""\
[fit]
max_component_count = 3
min_component_count = 1
""")
def test_build_reference_spectrum_list_from_config_file(fs):
reference_config = fit_task_builder.get_config_parser()
reference_config.read_string("""\
[references]
references/*.e
""")
fs.CreateDirectory(directory_path='references')
fs.CreateFile(file_path='references/arsenate_aqueous_avg_als_cal.e', contents=_spectrum_file_content)
fs.CreateFile(file_path='references/arsenate_sorbed_anth_avg_als_cal.e', contents=_spectrum_file_content)
reference_list = fit_task_builder.build_reference_spectrum_list_from_config_file(reference_config)
assert len(reference_list) == 2
def test_build_unknown_spectrum_list_from_config_file(fs):
data_config = fit_task_builder.get_config_parser()
data_config.read_string("""\
[data]
data/*.e
""")
fs.CreateDirectory(directory_path='data')
fs.CreateFile(file_path='data/data_0.e', contents=_spectrum_file_content)
fs.CreateFile(file_path='data/data_1.e', contents=_spectrum_file_content)
data_list = fit_task_builder.build_unknown_spectrum_list_from_config_file(data_config)
assert len(data_list) == 2
def test_build_reference_spectrum_list_from_prm_section__bad_component_counts(fs):
reference_config = fit_task_builder.get_config_parser()
reference_config.read_string("""\
[prm]
NBCompoMax = 1
NBCompoMin = 4
arsenate_aqueous_avg_als_cal.e
arsenate_sorbed_anth_avg_als_cal.e
""")
fs.CreateFile(file_path='arsenate_aqueous_avg_als_cal.e', contents=_spectrum_file_content)
fs.CreateFile(file_path='arsenate_sorbed_anth_avg_als_cal.e', contents=_spectrum_file_content)
with pytest.raises(fit_task_builder.ConfigurationFileError):
fit_task_builder.build_reference_spectrum_list_from_config_prm_section(reference_config)
def test__get_required_config_value():
config = fit_task_builder.get_config_parser()
test_section = 'blah'
test_option = 'bleh'
test_value = 'blih'
config.add_section(section=test_section)
config.set(section=test_section, option=test_option, value=test_value)
assert test_value == fit_task_builder._get_required_config_value(
config=config, section=test_section, option=test_option)
with pytest.raises(fit_task_builder.ConfigurationFileError):
fit_task_builder._get_required_config_value(config=config, section='missing', option='missing')
def test_build_reference_spectrum_list_from_prm_section(fs):
reference_config = fit_task_builder.get_config_parser()
reference_config.read_string("""\
[prm]
NBCompoMax = 4
NBCompoMin = 1
arsenate_aqueous_avg_als_cal.e
arsenate_sorbed_anth_avg_als_cal.e
""")
fs.CreateFile(file_path='arsenate_aqueous_avg_als_cal.e', contents=_spectrum_file_content)
fs.CreateFile(file_path='arsenate_sorbed_anth_avg_als_cal.e', contents=_spectrum_file_content)
max_count, min_count, reference_list = \
fit_task_builder.build_reference_spectrum_list_from_config_prm_section(reference_config)
assert max_count == 4
assert min_count == 1
assert len(reference_list) == 2
def test_main(fs):
# write the necessary input files:
# configuration file
# reference files
# sample files
# prm file
# then check for the expected output files
# plot pdf
# table txt
# sample fit files
fs.CreateFile(
'ref_1.e', '''\
1000.0\t0.01
1000.1\t0.02
1000.2\t0.03
1000.3\t0.04
1000.4\t0.05
''')
fs.CreateFile(
'ref_2.e', '''\
1000.0\t0.05
1000.1\t0.04
1000.2\t0.03
1000.3\t0.02
1000.4\t0.01
''')
fs.CreateFile(
'ref_3.e', '''\
1000.0\t0.01
1000.1\t0.01
1000.2\t0.01
1000.3\t0.01
1000.4\t0.01
''')
fs.CreateFile(
'test_main.prm', '''\
NbCompoMax=3
NbCompoMin=1
ref=ref_1.e
ref=ref_2.e
ref=ref_3.e
''')
# sample 1 is twice ref_1
fs.CreateFile('sample_1.e', '''\
1000.1\t0.02
1000.2\t0.04
1000.3\t0.06
''')
# sample 2 is half ref_2
fs.CreateFile('sample_2.e', '''\
1000.1\t0.015
1000.2\t0.010
1000.3\t0.005
''')
# sample 3 is ref_1 plus ref_3
fs.CreateFile('sample_3.e', '''\
1000.1\t0.03
1000.2\t0.04
1000.3\t0.05
''')
test_main_config = get_config_parser()
fs.CreateFile(
'test_main.cfg', '''\
[reference_spectra]
prm = test_main.prm
[data]
sample*.e
[output]
best_fit_files_dir = .
plots_pdf_fp = test_main_plots.pdf
table_fp = test_main_table.txt
reference_plots_pdf = test_main_reference_plots.pdf
''')
result = cli(['test_main.cfg'])
assert result.exit_code == 0
assert os.path.exists('test_main_plots.pdf')
assert os.path.exists('test_main_table.txt')
assert os.path.exists('sample_1_fit.txt')
