def __check_usergroup(self):
LOGGER.debug('Checking if user is in docker usergroup')
process = subprocess.Popen(['groups'],
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT)
output, _ = process.communicate()
groups = output.decode('utf-8').split()
if 'docker' in groups:
return True
else:
return False
def __check_docker_client(self):
LOGGER.debug('Checking if docker client exist.')
process = subprocess.Popen(['which', 'docker'],
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT)
output, _ = process.communicate()
decoded = output.decode('utf-8')
if decoded != '':
return True
else:
return False
def __update_replacements(self, cdecode, is_raw_header=False):
self.func_replace_trg_listbox.delete(0, tk.END)
if is_raw_header:
cdename = self.parent.cdemapper.get_cde_mipmap_header(cdecode)
else:
cdename = cdecode
current_replacements = self.parent.cdemapper.get_corr_replacements(
cdename)
if current_replacements:
LOGGER.debug('Replacements found for cde "{}": {}'.format(
cdecode, len(current_replacements)))
for rep in current_replacements:
self.__add_replacement(rep.source, rep.target)
def add_replacement_expr(self):
reps = self.func_replace_trg_listbox.get(0, tk.END)
replacemnts = []
source_col = self.parent.csv_name.replace(
".csv", "") + '.' + self.selected_column
LOGGER.debug('the repleacements are: {}'.format(reps))
if len(reps) > 0:
self.expressions_text.delete("1.0", "end-1c")
for rep in reps:
s = rep.split('->')
replacemnts.append(Replacement(s[0], s[1]))
expr = ifstr(source_col, replacemnts)
LOGGER.info('the expression is: {}'.format(expr))
self.expressions_text.insert('1.0', expr)
def __create_db_container(self):
"""Creates a postgres 9.6 container.
"""
self.__check_db_container(mode='running')
self.__check_db_container(mode='exist')
if self.__is_db_running:
LOGGER.info('db container ({}) is already up and'
' running. Skipping creation step...'.format(
self.__db_cont_name))
self.__remove_create_db()
pass
elif self.__is_db_exist and not self.__is_db_running:
LOGGER.info('db container({}) already exists. '
'Restarting db container'.format(self.__db_cont_name))
subprocess.run(['docker', 'restart', self.__db_cont_name])
time.sleep(10)
self.__remove_create_db()
else:
# create the db container
LOGGER.debug('Creating db container with name {}'.format(
self.__db_cont_name))
arg_port = ['-p', '{}:5432'.format(self.__dbport)]
arg_name = ['--name', self.__db_cont_name]
arg_env1 = ['-e', 'POSTGRES_PASSWORD={}'.format(self.__dbpassword)]
arg_env2 = ['-e', 'POSTGRES_USER={}'.format(self.__dbuser)]
arg_img = ['-d', self.__db_image]
command2 = ['docker', 'run'
] + arg_port + arg_name + arg_env1 + arg_env2 + arg_img
try:
createproc = subprocess.run(command2)
time.sleep(50)
self.__remove_create_db()
except subprocess.CalledProcessError:
LOGGER.warning(
'There was an error while attempting creating the db container.'
)
raise DockerExecError(
'There was an error while attempting creating the db container.'
)
def suggest_cde(self, columnreport, threshold=0.6):
"""Suggests the most similar CDE for the column.
Arguments:
:param columnreport: ColumnReport object with info of a datset column
:param threshold: 0-1 similarity threshold, below that not a cde is suggested
:returns: a CdeVariable object
"""
name = columnreport.name
val_range = columnreport.value_range
mip_type = columnreport.miptype
LOGGER.debug('The incoming column name is: {}'.format(name))
# select cdes with tha same type and calculate similarity
canditates = [
cde for cde in self.__cdes.values() if cde.miptype == mip_type
]
LOGGER.debug('Number of cdes with miptype {} is: {}'.format(
mip_type, len(canditates)))
if canditates:
canditates.sort(key=lambda x: x.similarity(name, val_range),
reverse=True)
canditate = canditates[0]
similarity = canditate.similarity(name, val_range)
LOGGER.debug(
'The simirarity between "{}" and cde "{}" is: {}'.format(
name, canditate.code, similarity))
if similarity >= threshold:
return canditate
else:
LOGGER.info(
'No cde match found for incoming column "{}"'.format(name))
return None
else:
LOGGER.info(
'No cde match found for incoming column "{}"'.format(name))
return None
def __is_mipmap_container_exist(self):
proc_docker = subprocess.Popen(['docker', 'ps', '-a'],
stdout=subprocess.PIPE)
proc_grep = subprocess.Popen(['grep', self.__name],
stdin=proc_docker.stdout,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
stdout, _ = proc_grep.communicate()
output = str(stdout).split()
LOGGER.debug(output)
try:
container_name = output[-1]
# remove new line spacial character
container_name = container_name.rstrip("\\n'")
except IndexError:
container_name = None
if container_name == self.__name:
return True
else:
return False
def __init__(self, rootfolder, username):
""" Arguments:
:param rootfolder: folder path with DICOMs subfolders
:param username: str with the username
"""
start_time = time.time()
self.reportdata = None
self.rootfolder = rootfolder
self.subfolders = getsubfolders(rootfolder)
self.username = username
self.dataset = {
'version': __version__,
'date_qc_ran':
datetime.datetime.now().strftime('%d/%m/%Y %H:%M:%S'),
'username': username,
'dicomfolder': str(os.path.abspath(rootfolder))
}
# files that are not DICOMS (.dcm)
# list of (folder, filename)
self.__notprocessed = []
# MRISequences objects
self.__invalidseq = []
# MRIPatient objects
self.__patients = []
# statistic info
self.__totalvalidseq = 0
self.__totalstudies = 0
self.__totalinvaliddicoms = 0
self.__seriesdescriptions_valid = set()
self.__seriesdescriptions_invalid = set()
self.__seqperpatient_distr = {
's1': 0,
's2': 0,
's3-s5': 0,
's6more': 0,
}
self.__patientid_with_invalids = set()
# Read all the dcm files and calc QC stats
self.__readicoms_parallel(mp.cpu_count() + 1)
self.__collect_stats()
LOGGER.debug('Dicom analysis running time: %s seconds' %
(time.time() - start_time))
LOGGER.debug('Folders read: %i' % len(self.subfolders))
LOGGER.debug('Total MRI sequences found: %i' %
(self.totalvalidsequences + self.totalinvalidsequences))
LOGGER.info('Patients with good seq: %i' % self.totalpatients)
LOGGER.info('Total visits: %i' % self.totalvisits)
LOGGER.info('Good seq: %i' % self.totalvalidsequences)
LOGGER.info('Bad seq: %i' % self.totalinvalidsequences)
LOGGER.info('Files not processed: %i' % self.totalbadfiles)
def __add_replacement(self, sourcevalue, targetvalue):
LOGGER.debug('Target value is: {} and the source value is: {}'.format(
targetvalue, sourcevalue))
stringforbox = '->'.join([sourcevalue, targetvalue])
self.func_replace_trg_listbox.insert(tk.END, stringforbox)
