help="explicitly specify subdirectory of DIR with "
"Fastq files to run the QC on.")
p.add_argument('--link',
action='store_true',
dest='link',
default=False,
help="hard link files instead of copying")
options, args = p.parse_known_args()
# Get analysis dir
try:
dirn = args[0]
print("Loading data for analysis dir %s" % dirn)
except IndexError:
p.error("Need to supply the path to an analysis dir")
sys.exit(1)
analysis_dir = AnalysisDir(dirn)
# Get specified project
try:
project_name = args[1]
except IndexError:
# List projects and exit
print("Projects:")
for project in analysis_dir.projects:
print("%s" % project.name)
if len(project.fastq_dirs) > 1:
# List the fastq sets if there are more than one
# and flag the primary set with an asterisk
for d in project.fastq_dirs:
is_primary = (d == project.info.primary_fastq_dir)
print("- %s%s" % (d, (" *" if is_primary else "")))
if analysis_dir.undetermined:
help="initialise metadata file for the selected project "
"(nb can only be applied to one project at a time)")
p.add_argument("-u","--update",
action="append",dest="update",
help="update the metadata in the selected project by "
"specifying key=value pairs e.g. user='******' "
"(nb can only be applied to one project at a time)")
p.add_argument("dir",metavar="DIR",
help="analysis directory to update metadata for")
p.add_argument("project",metavar="PROJECT",
help="project within the analysis directory to update "
"metadata for")
args = p.parse_args()
# Try and load the data
try:
run = AnalysisDir(args.dir)
print("Loaded data from %s" % args.dir)
print("Found %d projects" % run.n_projects)
if run.undetermined is not None:
print("Found 'undetermined' analysis")
print("Found %d sequencing data directories" % run.n_sequencing_data)
except Exception as ex:
sys.stderr.write("Failed to load data for %s: %s\n" % (args.dir,ex))
sys.exit(1)
# Aquire the projects
target_project = args.project
projects = run.get_projects(pattern=target_project)
if not projects:
print("No projects found")
sys.exit(1)
print("%d projects selected" % len(projects))
p.add_option("--pi",action='store',dest="pi_name",default=None,
help="List data for PI(s) matching PI_NAME (can use glob-style "
"patterns)")
p.add_option("--unassigned",action='store_true',dest="unassigned",default=False,
help="List data for projects where PI is not assigned")
opts,args = p.parse_args()
# Collect data
audit_data = {}
unassigned = []
undetermined = []
for d in args:
for dirn in utils.list_dirs(d):
dirn = os.path.join(d,dirn)
#print "Examining %s" % dirn
try:
run = AnalysisDir(dirn)
for p in run.get_projects():
if p.name == "undetermined":
undetermined.append((p,get_size(p.dirn)))
continue
pi = p.info.PI
if pi is None:
# PI is not assigned
p.info['run'] = os.path.basename(dirn)
unassigned.append(p)
continue
elif opts.pi_name is not None:
if not fnmatch.fnmatch(pi,opts.pi_name):
# Skip non-matching name
continue
#print "\t%s: %s" % (p.name,pi)
# Collect arguments
if len(args) != 4:
p.error("Wrong args")
galaxy_url = args[0]
library_name = args[1]
dest = args[2]
rundir = args[3]
# Subset of projects
if opts.projects is not None:
projects = opts.projects.split(',')
else:
projects = None
# Get analysis dir
analysis_dir = AnalysisDir(rundir)
print "Run name: %s" % analysis_dir.run_name
print "Run number: %s" % analysis_dir.metadata.run_number
print "Date stamp: %s" % analysis_dir.date_stamp
try:
# Create the directory structure on the server
build_library_directory(analysis_dir, dest, projects=projects)
# Upload to the Galaxy instance
create_data_library(galaxy_url,
library_name,
analysis_dir,
dest,
projects=projects,
no_verify=opts.no_verify)
except Exception as ex:
logging.critical("Export failed: %s" % ex)