def raw_data(self):
if not self._check_pargs(["project"]):
return
# if necessary, reformat flowcell identifier
if self.pargs.flowcell:
self.pargs.flowcell = self.pargs.flowcell.split("_")[-1]
# get the uid and gid to use for destination files
uid = os.getuid()
gid = os.getgid()
if self.pargs.group is not None and len(self.pargs.group) > 0:
gid = grp.getgrnam(group).gr_gid
self.log.debug("Connecting to project database")
p_con = ProjectSummaryConnection(**vars(self.pargs))
assert p_con, "Could not get connection to project databse"
self.log.debug("Connecting to samples database")
s_con = SampleRunMetricsConnection(**vars(self.pargs))
assert s_con, "Could not get connection to samples databse"
# Fetch the Uppnex project to deliver to
if not self.pargs.uppmax_project:
self.pargs.uppmax_project = p_con.get_entry(self.pargs.project, "uppnex_id")
if not self.pargs.uppmax_project:
self.log.error("Uppmax project was not specified and could not be fetched from project database")
return
# Extract the list of samples and runs associated with the project and sort them
samples = sorted(s_con.get_samples(fc_id=self.pargs.flowcell, sample_prj=self.pargs.project), key=lambda k: (k.get('project_sample_name','NA'), k.get('flowcell','NA'), k.get('lane','NA')))
# Setup paths and verify parameters
self._meta.production_root = self.app.config.get("production", "root")
self._meta.root_path = self._meta.production_root
proj_base_dir = os.path.join(self._meta.root_path, self.pargs.project)
assert os.path.exists(self._meta.production_root), "No such directory {}; check your production config".format(self._meta.production_root)
assert os.path.exists(proj_base_dir), "No project {} in production path {}".format(self.pargs.project,self._meta.root_path)
try:
self._meta.uppnex_project_root = self.app.config.get("deliver", "uppnex_project_root")
except Exception as e:
self.log.warn("{}, will use '/proj' as uppnext_project_root".format(e))
self._meta.uppnex_project_root = '/proj'
try:
self._meta.uppnex_delivery_dir = self.app.config.get("deliver", "uppnex_project_delivery_path")
except Exception as e:
self.log.warn("{}, will use 'INBOX' as uppnext_project_delivery_path".format(e))
self._meta.uppnex_delivery_dir = 'INBOX'
destination_root = os.path.join(self._meta.uppnex_project_root,self.pargs.uppmax_project,self._meta.uppnex_delivery_dir)
assert os.path.exists(destination_root), "Delivery destination folder {} does not exist".format(destination_root)
destination_root = os.path.join(destination_root,self.pargs.project)
# If interactively select, build a list of samples to skip
if self.pargs.interactive:
to_process = []
for sample in samples:
sname = sample.get("project_sample_name")
index = sample.get("sequence")
fcid = sample.get("flowcell")
lane = sample.get("lane")
date = sample.get("date")
self.log.info("Sample: {}, Barcode: {}, Flowcell: {}, Lane: {}, Started on: {}".format(sname,
index,
fcid,
lane,
date))
if query_yes_no("Deliver sample?", default="no"):
to_process.append(sample)
samples = to_process
# Find uncompressed fastq
uncompressed = self._find_uncompressed_fastq_files(proj_base_dir,samples)
if len(uncompressed) > 0:
self.log.warn("The following samples have uncompressed *.fastq files that cannot be delivered: {}".format(",".join(uncompressed)))
if not query_yes_no("Continue anyway?", default="no"):
return
self.log.info("Will deliver data for {} samples from project {} to {}".format(len(samples),self.pargs.project,destination_root))
if not query_yes_no("Continue?"):
return
# Get the list of files to transfer and the destination
self.log.debug("Gathering list of files to copy")
to_copy = self.get_file_copy_list(proj_base_dir,
destination_root,
samples)
# Make sure that transfer will be with rsync
if not self.pargs.rsync:
self.log.warn("Files must be transferred using rsync")
if not query_yes_no("Do you wish to continue delivering using rsync?", default="yes"):
return
self.pargs.rsync = True
# Process each sample run
for id, files in to_copy.items():
# get the sample database object
[sample] = [s for s in samples if s.get('_id') == id]
self.log.info("Processing sample {} and flowcell {}".format(sample.get("project_sample_name","NA"),sample.get("flowcell","NA")))
# transfer files
self.log.debug("Transferring {} fastq files".format(len(files)))
self._transfer_files([f[0] for f in files], [f[1] for f in files])
passed = True
if self.pargs.link or self.pargs.dry_run:
passed = False
else:
# calculate md5sums on the source side and write it on the destination
md5 = []
for f in files:
m = md5sum(f[0])
mfile = "{}.md5".format(f[1])
md5.append([m,mfile,f[2],f[0]])
self.log.debug("md5sum for source file {}: {}".format(f[0],m))
# write the md5sum to a file at the destination and verify the transfer
for m, mfile, read, srcpath in md5:
dstfile = os.path.splitext(mfile)[0]
self.log.debug("Writing md5sum to file {}".format(mfile))
self.app.cmd.write(mfile,"{} {}".format(m,os.path.basename(dstfile)),True)
self.log.debug("Verifying md5sum for file {}".format(dstfile))
dm = md5sum(dstfile)
self.log.debug("md5sum for destination file {}: {}".format(dstfile,dm))
if m != dm:
self.log.warn("md5sum verification FAILED for {}. Source: {}, Target: {}".format(dstfile,m,dm))
self.log.warn("Improperly transferred file {} is removed from destination, please retry transfer of this file".format(dstfile))
self.app.cmd.safe_unlink(dstfile)
self.app.cmd.safe_unlink(mfile)
passed = False
continue
# Modify the permissions to ug+rw
for f in [dstfile, mfile]:
self.app.cmd.chmod(f,stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP | stat.S_IWGRP)
# touch the flag to trigger uppmax inbox permission fix
self.app.cmd.safe_touchfile(os.path.join("/sw","uppmax","var","inboxfix","schedule",self.pargs.uppmax_project))
# log the transfer to statusdb if verification passed
if passed:
self.log.info("Logging delivery to StatusDB document {}".format(id))
data = {'raw_data_delivery': {'timestamp': utc_time(),
'files': {'R{}'.format(read):{'md5': m,
'path': os.path.splitext(mfile)[0],
'size_in_bytes': self._getsize(os.path.splitext(mfile)[0]),
'source_location': srcpath} for m, mfile, read, srcpath in md5},
}
}
jsonstr = json.dumps(data)
jsonfile = os.path.join(os.path.dirname(md5[0][3]),
"{}_{}_{}_{}_L{}_raw_data_delivery.json".format(sample.get("date"),
sample.get("flowcell"),
sample.get("project_sample_name"),
sample.get("sequence"),
sample.get("lane")))
self.log.debug("Writing delivery to json file {}".format(jsonfile))
self.app.cmd.write(jsonfile,data=jsonstr,overwrite=True)
self.log.debug("Saving delivery in StatusDB document {}".format(id))
sample.update(data)
self._save(s_con,sample)
self.log.debug(jsonstr)
def raw_data(self):
if not self._check_pargs(["project"]):
return
# if necessary, reformat flowcell identifier
if self.pargs.flowcell:
self.pargs.flowcell = self.pargs.flowcell.split("_")[-1]
# get the uid and gid to use for destination files
uid = os.getuid()
gid = os.getgid()
if self.pargs.group is not None and len(self.pargs.group) > 0:
gid = grp.getgrnam(group).gr_gid
self.log.debug("Connecting to project database")
p_con = ProjectSummaryConnection(**vars(self.pargs))
assert p_con, "Could not get connection to project databse"
self.log.debug("Connecting to samples database")
s_con = SampleRunMetricsConnection(**vars(self.pargs))
assert s_con, "Could not get connection to samples databse"
# Fetch the Uppnex project to deliver to
if not self.pargs.uppmax_project:
self.pargs.uppmax_project = p_con.get_entry(
self.pargs.project, "uppnex_id")
if not self.pargs.uppmax_project:
self.log.error(
"Uppmax project was not specified and could not be fetched from project database"
)
return
# Extract the list of samples and runs associated with the project and sort them
samples = sorted(s_con.get_samples(fc_id=self.pargs.flowcell,
sample_prj=self.pargs.project),
key=lambda k:
(k.get('project_sample_name', 'NA'),
k.get('flowcell', 'NA'), k.get('lane', 'NA')))
# Setup paths and verify parameters
self._meta.production_root = self.app.config.get("production", "root")
self._meta.root_path = self._meta.production_root
proj_base_dir = os.path.join(self._meta.root_path, self.pargs.project)
assert os.path.exists(
self._meta.production_root
), "No such directory {}; check your production config".format(
self._meta.production_root)
assert os.path.exists(
proj_base_dir), "No project {} in production path {}".format(
self.pargs.project, self._meta.root_path)
try:
self._meta.uppnex_project_root = self.app.config.get(
"deliver", "uppnex_project_root")
except Exception as e:
self.log.warn(
"{}, will use '/proj' as uppnext_project_root".format(e))
self._meta.uppnex_project_root = '/proj'
try:
self._meta.uppnex_delivery_dir = self.app.config.get(
"deliver", "uppnex_project_delivery_path")
except Exception as e:
self.log.warn(
"{}, will use 'INBOX' as uppnext_project_delivery_path".format(
e))
self._meta.uppnex_delivery_dir = 'INBOX'
destination_root = os.path.join(self._meta.uppnex_project_root,
self.pargs.uppmax_project,
self._meta.uppnex_delivery_dir)
assert os.path.exists(
destination_root
), "Delivery destination folder {} does not exist".format(
destination_root)
destination_root = os.path.join(destination_root, self.pargs.project)
# If interactively select, build a list of samples to skip
if self.pargs.interactive:
to_process = []
for sample in samples:
sname = sample.get("project_sample_name")
index = sample.get("sequence")
fcid = sample.get("flowcell")
lane = sample.get("lane")
date = sample.get("date")
self.log.info(
"Sample: {}, Barcode: {}, Flowcell: {}, Lane: {}, Started on: {}"
.format(sname, index, fcid, lane, date))
if query_yes_no("Deliver sample?", default="no"):
to_process.append(sample)
samples = to_process
# Find uncompressed fastq
uncompressed = self._find_uncompressed_fastq_files(
proj_base_dir, samples)
if len(uncompressed) > 0:
self.log.warn(
"The following samples have uncompressed *.fastq files that cannot be delivered: {}"
.format(",".join(uncompressed)))
if not query_yes_no("Continue anyway?", default="no"):
return
self.log.info(
"Will deliver data for {} samples from project {} to {}".format(
len(samples), self.pargs.project, destination_root))
if not query_yes_no("Continue?"):
return
# Get the list of files to transfer and the destination
self.log.debug("Gathering list of files to copy")
to_copy = self.get_file_copy_list(proj_base_dir, destination_root,
samples)
# Make sure that transfer will be with rsync
if not self.pargs.rsync:
self.log.warn("Files must be transferred using rsync")
if not query_yes_no(
"Do you wish to continue delivering using rsync?",
default="yes"):
return
self.pargs.rsync = True
# Process each sample run
for id, files in to_copy.items():
# get the sample database object
[sample] = [s for s in samples if s.get('_id') == id]
self.log.info("Processing sample {} and flowcell {}".format(
sample.get("project_sample_name", "NA"),
sample.get("flowcell", "NA")))
# calculate md5sums on the source side and write it on the destination
md5 = []
for f in files:
m = md5sum(f[0])
mfile = "{}.md5".format(f[1])
md5.append([m, mfile, f[2], f[0]])
self.log.debug("md5sum for source file {}: {}".format(f[0], m))
# transfer files
self.log.debug("Transferring {} fastq files".format(len(files)))
self._transfer_files([f[0] for f in files], [f[1] for f in files])
# write the md5sum to a file at the destination and verify the transfer
passed = True
for m, mfile, read, srcpath in md5:
dstfile = os.path.splitext(mfile)[0]
self.log.debug("Writing md5sum to file {}".format(mfile))
self.app.cmd.write(
mfile, "{} {}".format(m, os.path.basename(dstfile)), True)
self.log.debug("Verifying md5sum for file {}".format(dstfile))
# if dry-run, make sure verification pass
if self.pargs.dry_run:
dm = m
else:
dm = md5sum(dstfile)
self.log.debug("md5sum for destination file {}: {}".format(
dstfile, dm))
if m != dm:
self.log.warn(
"md5sum verification FAILED for {}. Source: {}, Target: {}"
.format(dstfile, m, dm))
self.log.warn(
"Improperly transferred file {} is removed from destination, please retry transfer of this file"
.format(dstfile))
self.app.cmd.safe_unlink(dstfile)
self.app.cmd.safe_unlink(mfile)
passed = False
continue
# Modify the permissions to ug+rw
for f in [dstfile, mfile]:
self.app.cmd.chmod(
f, stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP
| stat.S_IWGRP)
# touch the flag to trigger uppmax inbox permission fix
self.app.cmd.safe_touchfile(
os.path.join("/sw", "uppmax", "var", "inboxfix", "schedule",
self.pargs.uppmax_project))
# log the transfer to statusdb if verification passed
if passed:
self.log.info(
"Logging delivery to StatusDB document {}".format(id))
data = {
'raw_data_delivery': {
'timestamp': utc_time(),
'files': {
'R{}'.format(read): {
'md5':
m,
'path':
os.path.splitext(mfile)[0],
'size_in_bytes':
self._getsize(os.path.splitext(mfile)[0]),
'source_location':
srcpath
}
for m, mfile, read, srcpath in md5
},
}
}
jsonstr = json.dumps(data)
jsonfile = os.path.join(
os.path.dirname(md5[0][3]),
"{}_{}_{}_{}_L{}_raw_data_delivery.json".format(
sample.get("date"), sample.get("flowcell"),
sample.get("project_sample_name"),
sample.get("sequence"), sample.get("lane")))
self.log.debug(
"Writing delivery to json file {}".format(jsonfile))
self.app.cmd.write(jsonfile, data=jsonstr, overwrite=True)
self.log.debug(
"Saving delivery in StatusDB document {}".format(id))
sample.update(data)
self._save(s_con, sample)
self.log.debug(jsonstr)
def raw_data(self):
if not self._check_pargs(["project"]):
return
# if necessary, reformat flowcell identifier
if self.pargs.flowcell:
self.pargs.flowcell = self.pargs.flowcell.split("_")[-1]
# get the uid and gid to use for destination files
uid = os.getuid()
gid = os.getgid()
if self.pargs.group is not None and len(self.pargs.group) > 0:
gid = grp.getgrnam(group).gr_gid
self.log.debug("Connecting to project database")
p_con = ProjectSummaryConnection(**vars(self.pargs))
assert p_con, "Could not get connection to project database"
self.log.debug("Connecting to flowcell database")
f_con = FlowcellRunMetricsConnection(**vars(self.pargs))
assert f_con, "Could not get connection to flowcell database"
self.log.debug("Connecting to x_flowcell database")
x_con = X_FlowcellRunMetricsConnection(**vars(self.pargs))
assert x_con, "Could not get connection to x_flowcell database"
# Fetch the Uppnex project to deliver to
if not self.pargs.uppmax_project:
self.pargs.uppmax_project = p_con.get_entry(
self.pargs.project, "uppnex_id")
if not self.pargs.uppmax_project:
self.log.error(
"Uppmax project was not specified and could not be fetched from project database"
)
return
# Setup paths and verify parameters
self._meta.production_root = self.pargs.root if self.pargs.root else self.app.config.get(
"production", "root")
self._meta.root_path = self._meta.production_root
proj_base_dir = os.path.join(self._meta.root_path, self.pargs.project)
assert os.path.exists(
self._meta.production_root
), "No such directory {}; check your production config".format(
self._meta.production_root)
assert os.path.exists(
proj_base_dir), "No project {} in production path {}".format(
self.pargs.project, self._meta.root_path)
try:
self._meta.uppnex_project_root = self.app.config.get(
"deliver", "uppnex_project_root")
except Exception as e:
self.log.warn(
"{}, will use '/proj' as uppnext_project_root".format(e))
self._meta.uppnex_project_root = '/proj'
try:
self._meta.uppnex_delivery_dir = self.app.config.get(
"deliver", "uppnex_project_delivery_path")
except Exception as e:
self.log.warn(
"{}, will use 'INBOX' as uppnext_project_delivery_path".format(
e))
self._meta.uppnex_delivery_dir = 'INBOX'
destination_root = os.path.join(self._meta.uppnex_project_root,
self.pargs.uppmax_project,
self._meta.uppnex_delivery_dir)
assert os.path.exists(
destination_root
), "Delivery destination folder {} does not exist".format(
destination_root)
destination_root = os.path.join(destination_root, self.pargs.project)
# Find uncompressed fastq
uncompressed = self._find_uncompressed_fastq_files(
proj_base_dir=proj_base_dir,
sample=self.pargs.sample,
flowcell=self.pargs.flowcell)
if len(uncompressed) > 0:
self.log.error(
"There are uncompressed fastq file for project, kindly check all files are compressed properly before delivery"
)
return
# Extract the list of samples and runs associated with the project and sort them
samples = self.samples_to_copy(
pid=p_con.get_entry(self.pargs.project, "project_id"),
pod=p_con.get_entry(self.pargs.project, "open_date"),
fc_dict={
'HiSeq2500': f_con.proj_list,
'HiSeqX': x_con.proj_list
},
proj_base_dir=proj_base_dir,
destination_root=destination_root,
sample=self.pargs.sample,
flowcell=self.pargs.flowcell)
# If interactively select, build a list of samples to skip
if self.pargs.interactive:
to_process = {}
for sample in samples:
if query_yes_no("Deliver sample {} ?".format(sample),
default="no"):
to_process[sample] = samples[sample]
samples = to_process
if self.pargs.sample:
sample = samples.get(self.pargs.sample)
if not sample:
self.log.error(
"There is no such sample {} for project {}".format(
self.pargs.sample, self.pargs.project))
return
samples = {self.pargs.sample: sample}
self.log.info(
"Will deliver data for {} samples from project {} to {}".format(
len(samples), self.pargs.project, destination_root))
if not query_yes_no("Continue?"):
return
# Make sure that transfer will be with rsync
if not self.pargs.rsync:
self.log.warn("Files must be transferred using rsync")
if not query_yes_no(
"Do you wish to continue delivering using rsync?",
default="yes"):
return
self.pargs.rsync = True
# Process each sample
for sample, flowcells in samples.iteritems():
for fc, files in flowcells.iteritems():
self.log.info("Processing sample {} and flowcell {}".format(
sample, fc))
# transfer files
self.log.debug("Transferring {} fastq files".format(
len(files['src'])))
self._transfer_files(sources=files['src'],
targets=files['dst'])
passed = True
if self.pargs.link or self.pargs.dry_run:
passed = False
else:
# calculate md5sums on the source side and write it on the destination
md5 = []
for s, d in zip(files['src'], files['dst']):
m = md5sum(s)
mfile = "{}.md5".format(d)
md5.append([m, mfile, s])
self.log.debug("md5sum for source file {}: {}".format(
s, m))
# write the md5sum to a file at the destination and verify the transfer
for m, mfile, srcpath in md5:
dstfile = os.path.splitext(mfile)[0]
self.log.debug(
"Writing md5sum to file {}".format(mfile))
self.app.cmd.write(
mfile, "{} {}".format(m,
os.path.basename(dstfile)),
True)
self.log.debug(
"Verifying md5sum for file {}".format(dstfile))
dm = md5sum(dstfile)
self.log.debug(
"md5sum for destination file {}: {}".format(
dstfile, dm))
if m != dm:
self.log.warn(
"md5sum verification FAILED for {}. Source: {}, Target: {}"
.format(dstfile, m, dm))
self.log.warn(
"Improperly transferred file {} is removed from destination, please retry transfer of this file"
.format(dstfile))
self.app.cmd.safe_unlink(dstfile)
self.app.cmd.safe_unlink(mfile)
passed = False
continue
# Modify the permissions to ug+rw
for f in [dstfile, mfile]:
self.app.cmd.chmod(
f, stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP
| stat.S_IWGRP)
# touch the flag to trigger uppmax inbox permission fix
self.app.cmd.safe_touchfile(
os.path.join("/sw", "uppmax", "var", "inboxfix",
"schedule", self.pargs.uppmax_project))
# log the transfer to statusdb if verification passed
if passed:
data = {
'raw_data_delivery': {
'timestamp': utc_time(),
'files': {
os.path.splitext(
(os.path.basename(srcpath)))[0]:
{
'md5':
m,
'path':
os.path.splitext(mfile)[0],
'size_in_bytes':
self._getsize(os.path.splitext(mfile)[0]),
'source_location':
srcpath
}
for m, mfile, srcpath in md5
}
}
}
jsonstr = json.dumps(data)
jsonfile = os.path.join(
proj_base_dir, sample, fc,
"{}_{}_raw_data_delivery.json".format(sample, fc))
self.log.debug(
"Writing delivery to json file {}".format(jsonfile))
self.app.cmd.write(jsonfile, data=jsonstr, overwrite=True)
self.log.debug(
"Saving delivery in StatusDB document {}".format(id))
if self.proj_flowcells[fc]['type'] == 'HiSeqX':
fc_con = x_con
else:
fc_con = f_con
fc_obj = fc_con.get_entry(fc)
self.log.info(
"Logging delivery to StatusDB document {}".format(
fc_obj.get('_id')))
fc_raw_data = fc_obj.get('raw_data_delivery', {})
fc_raw_data.update(data['raw_data_delivery'])
fc_obj['raw_data_delivery'] = fc_raw_data
self._save(fc_con, fc_obj)
self.log.debug(jsonstr)
