def get_sample_sheet(directory):
"""
gets the sample sheet file path from a given run directory
:param directory:
:return:
"""
logging.info("Looking for sample sheet in {}".format(directory))
# Checks if we can access to the given directory, return empty and log a warning if we cannot.
if not os.access(directory, os.W_OK):
logging.error((
"The directory is not accessible, can not parse samples from this directory {}"
"".format(directory), directory))
raise exceptions.DirectoryError(
"The directory is not accessible, "
"can not parse samples from this directory {}".format(
directory), directory)
sample_sheet_file_name = Parser.SAMPLE_SHEET_FILE_NAME
file_list = common.get_file_list(directory)
if sample_sheet_file_name not in file_list:
logging.error(
"No sample sheet file in the Directory Upload format found")
raise exceptions.DirectoryError(
"The directory {} has no sample sheet file in the Directory Upload format "
"with the name {}"
"".format(directory, sample_sheet_file_name), directory)
else:
logging.debug("Sample sheet found")
return os.path.join(directory, sample_sheet_file_name)
def get_file_list(directory):
"""
Get the list of file names in the data directory
:param directory: directory to search for files
:return: list of file names in data directory
"""
# verify that directory exists
if not os.path.exists(directory):
raise exceptions.DirectoryError("Could not list files, as directory does not exist.", directory)
# Create a file list of the directory, only hit the os once
file_list = next(os.walk(directory))[2]
return file_list
def find_directory_list(directory):
"""Find and return all directories in the specified directory.
Arguments:
directory -- the directory to find directories in
Returns: a list of directories including current directory
"""
# Checks if we can access to the given directory, return empty and log a warning if we cannot.
if not os.access(directory, os.W_OK):
raise exceptions.DirectoryError("The directory is not writeable, "
"can not upload samples from this directory {}".format(directory),
directory)
dir_list = next(os.walk(directory))[1] # Gets the list of directories in the directory
full_dir_list = []
for d in dir_list:
full_dir_list.append(os.path.join(directory, d))
return full_dir_list
def get_full_data_directory(sample_sheet):
"""
Returns the path to where the sequence data files can be found, including the sample_sheet directory
Note, this hits the os, and as such is not to be used with cloud solutions.
For cloud solutions, use get_relative_data_directory() and solve the actual path for your cloud environment
:param sample_sheet: Sample sheet acts as the starting point for the data directory
:return: a string which represents the concatenated path components, as per os.path.join
"""
sample_sheet_dir = os.path.dirname(sample_sheet)
partial_data_dir = os.path.join(sample_sheet_dir, "Alignment_1")
# Verify the partial path exits, path could not exist if there was a sequencing error
# Also, if someone runs the miniseq parser on a miseq directory, this is the failure point
if not os.path.exists(partial_data_dir):
raise exceptions.DirectoryError((
"The uploader was unable to find the data directory, Verify that the run directory is "
"undamaged, and that it is a MiniSeq sequencing run."),
partial_data_dir)
# get the directories [1] get the first directory [0]
data_dir = os.path.join(partial_data_dir,
next(os.walk(partial_data_dir))[1][0], "Fastq")
return data_dir