def test_load_file_to_disk_and_db7(self):
au = Analysis_collection_utils(dbsession_class=self.session_class,
analysis_name='AnalysisA',
tag_name='TagA',
collection_name='RunA',
collection_type='AnalysisA_Files',
collection_table='run',
base_path=self.temp_base_dir)
input_file_list = [
os.path.join(self.temp_work_dir, file_name)
for file_name in self.input_list
]
output_list = au.load_file_to_disk_and_db(
input_file_list=input_file_list,
withdraw_exisitng_collection=False
) # loading all files to same collection
base = BaseAdaptor(**{'session_class': self.session_class})
base.start_session()
ca = CollectionAdaptor(**{'session': base.session})
ca_files = ca.get_collection_files(collection_name='RunA',
collection_type='AnalysisA_Files',
output_mode='dataframe')
file_list = list(ca_files['file_path'].to_dict().values())
datestamp = get_datestamp_label()
test_file = os.path.join(
self.temp_base_dir, 'ProjectA', 'SampleA', 'ExperimentA', 'RunA',
'AnalysisA', '{0}_{1}_{2}_{3}.{4}'.format('RunA', 'AnalysisA',
'TagA', datestamp,
'cram'))
test_file = preprocess_path_name(input_path=test_file)
self.assertTrue(test_file in file_list)
self.assertTrue(test_file in output_list)
base.close_session()
def test_load_file_to_disk_and_db8(self):
au = Analysis_collection_utils(dbsession_class=self.session_class,
analysis_name='AnalysisA',
tag_name='TagA',
collection_name='RunA',
collection_type='AnalysisA_Files',
collection_table='run')
input_file = os.path.join(self.temp_work_dir, 'a.cram')
input_file = preprocess_path_name(input_path=input_file)
new_file_name = au.get_new_file_name(input_file=input_file)
datestamp = get_datestamp_label()
test_file_name = '{0}_{1}_{2}_{3}.{4}'.format('RunA', 'AnalysisA',
'TagA', datestamp,
'cram')
self.assertEqual(new_file_name, test_file_name)
def load_file_to_disk_and_db(self,
input_file_list,
withdraw_exisitng_collection=True,
autosave_db=True,
file_suffix=None,
force=True,
remove_file=False):
'''
A method for loading analysis results to disk and database. File will be moved to a new path if base_path is present.
Directory structure of the final path is based on the collection_table information.
Following will be the final directory structure if base_path is present
project - base_path/project_igf_id/analysis_name
sample - base_path/project_igf_id/sample_igf_id/analysis_name
experiment - base_path/project_igf_id/sample_igf_id/experiment_igf_id/analysis_name
run - base_path/project_igf_id/sample_igf_id/experiment_igf_id/run_igf_id/analysis_name
:param input_file_list: A list of input file to load, all using the same collection info
:param withdraw_exisitng_collection: Remove existing collection group, DO NOT use this while loading a list of files
:param autosave_db: Save changes to database, default True
:param file_suffix: Use a specific file suffix, use None if it should be same as original file
e.g. input.vcf.gz to output.vcf.gz
:param force: Toggle for removing existing file, default True
:param remove_file: A toggle for removing existing file from disk, default False
:returns: A list of final filepath
'''
try:
project_igf_id = None
sample_igf_id = None
experiment_igf_id = None
experiment_igf_id = None
run_igf_id = None
output_path_list = list() # define empty output list
dbconnected = False
if self.collection_name is None or \
self.collection_type is None or \
self.collection_table is None:
raise ValueError('File collection information is incomplete'
) # check for collection information
base = BaseAdaptor(**{'session_class': self.dbsession_class})
base.start_session() # connect to db
dbconnected = True
if self.base_path is not None:
if self.collection_table == 'sample':
sa = SampleAdaptor(**{'session': base.session})
sample_igf_id = self.collection_name
sample_exists = sa.check_sample_records_igf_id(
sample_igf_id=sample_igf_id)
if not sample_exists:
raise ValueError('Sample {0} not found in db'.\
format(sample_igf_id))
project_igf_id = \
sa.fetch_sample_project(sample_igf_id=sample_igf_id) # fetch project id for sample
elif self.collection_table == 'experiment':
ea = ExperimentAdaptor(**{'session': base.session})
experiment_igf_id = self.collection_name
experiment_exists = \
ea.check_experiment_records_id(
experiment_igf_id=experiment_igf_id)
if not experiment_exists:
raise ValueError('Experiment {0} not present in database'.\
format(experiment_igf_id))
(project_igf_id,sample_igf_id) = \
ea.fetch_project_and_sample_for_experiment(
experiment_igf_id=experiment_igf_id) # fetch project and sample id for experiment
elif self.collection_table == 'run':
ra = RunAdaptor(**{'session': base.session})
run_igf_id = self.collection_name
run_exists = ra.check_run_records_igf_id(
run_igf_id=run_igf_id)
if not run_exists:
raise ValueError('Run {0} not found in database'.\
format(run_igf_id))
(project_igf_id,sample_igf_id,experiment_igf_id) = \
ra.fetch_project_sample_and_experiment_for_run(
run_igf_id=run_igf_id) # fetch project, sample and experiment id for run
elif self.collection_table == 'project':
pa = ProjectAdaptor(**{'session': base.session})
project_igf_id = self.collection_name
project_exists = \
pa.check_project_records_igf_id(
project_igf_id=project_igf_id)
if not project_exists:
raise ValueError('Project {0} not found in database'.\
format(project_igf_id))
if self.rename_file and self.analysis_name is None:
raise ValueError('Analysis name is required for renaming file'
) # check analysis name
for input_file in input_file_list:
final_path = ''
if self.base_path is None: # do not move file if base_path is absent
final_path = os.path.dirname(input_file)
else: # move file path
if self.collection_table == 'project':
if project_igf_id is None:
raise ValueError('Missing project id for collection {0}'.\
format(self.collection_name))
final_path = \
os.path.join(
self.base_path,
project_igf_id,
self.analysis_name) # final path for project
elif self.collection_table == 'sample':
if project_igf_id is None or \
sample_igf_id is None:
raise ValueError('Missing project and sample id for collection {0}'.\
format(self.collection_name))
final_path = \
os.path.join(
self.base_path,
project_igf_id,
sample_igf_id,
self.analysis_name) # final path for sample
elif self.collection_table == 'experiment':
if project_igf_id is None or \
sample_igf_id is None or \
experiment_igf_id is None:
raise ValueError('Missing project,sample and experiment id for collection {0}'.\
format(self.collection_name))
final_path = \
os.path.join(
self.base_path,
project_igf_id,
sample_igf_id,
experiment_igf_id,
self.analysis_name) # final path for experiment
elif self.collection_table == 'run':
if project_igf_id is None or \
sample_igf_id is None or \
experiment_igf_id is None or \
run_igf_id is None:
raise ValueError('Missing project,sample,experiment and run id for collection {0}'.\
format(self.collection_name))
final_path = \
os.path.join(\
self.base_path,
project_igf_id,
sample_igf_id,
experiment_igf_id,
run_igf_id,
self.analysis_name) # final path for run
if self.rename_file:
new_filename = \
self.get_new_file_name(
input_file=input_file,
file_suffix=file_suffix)
final_path = \
os.path.join(
final_path,
new_filename) # get new filepath
else:
final_path = \
os.path.join(
final_path,
os.path.basename(input_file))
if final_path != input_file: # move file if its required
final_path = preprocess_path_name(
input_path=final_path
) # remove unexpected characters from file path
move_file(source_path=input_file,
destinationa_path=final_path,
force=force
) # move or overwrite file to destination dir
output_path_list.append(
final_path) # add final path to the output list
self.create_or_update_analysis_collection(
file_path=final_path,
dbsession=base.session,
withdraw_exisitng_collection=withdraw_exisitng_collection,
remove_file=remove_file,
autosave_db=autosave_db) # load new file collection in db
if autosave_db:
base.commit_session() # save changes to db for each file
base.commit_session() # save changes to db
base.close_session() # close db connection
return output_path_list
except:
if dbconnected:
base.rollback_session()
base.close_session()
raise