def create_table(summary_csv):
"""Long Amplicon Analysis results table"""
columns = []
columns.append(Column(Constants.C_BC))
columns.append(Column(Constants.C_GOOD))
columns.append(Column(Constants.C_GOOD_PCT))
columns.append(Column(Constants.C_CHIM))
columns.append(Column(Constants.C_CHIM_PCT))
columns.append(Column(Constants.C_NOISE))
columns.append(Column(Constants.C_NOISE_PCT))
t = Table(Constants.T_R, columns=columns)
COL_IDS = [
Constants.C_GOOD, Constants.C_GOOD_PCT, Constants.C_CHIM,
Constants.C_CHIM_PCT, Constants.C_NOISE, Constants.C_NOISE_PCT
]
def add_column(barcode_id, n_good, n_chimera, n_noise):
pct_good = pct_chimera = pct_noise = 0
total = n_good + n_chimera + n_noise
if total > 0:
pct_good = n_good / float(total)
pct_chimera = n_chimera / float(total)
pct_noise = n_noise / float(total)
values = [n_good, pct_good, n_chimera, pct_chimera, n_noise, pct_noise]
t.add_data_by_column_id(Constants.C_BC, bc_id)
for column_id, value in zip(COL_IDS, values):
t.add_data_by_column_id(column_id, value)
with open(summary_csv) as csv_in:
reader = csv.reader(csv_in, delimiter=',')
reader.next()
for rec in reader:
assert len(rec) == 7, rec
bc_id = rec[0]
if bc_id == "All":
continue
add_column(bc_id, int(rec[1]), int(rec[3]), int(rec[5]))
n_good = sum(t.get_column_by_id(Constants.C_GOOD).values)
n_chimera = sum(t.get_column_by_id(Constants.C_CHIM).values)
n_noise = sum(t.get_column_by_id(Constants.C_NOISE).values)
add_column("All", n_good, n_chimera, n_noise)
return t
def create_table(summary_csv):
"""Long Amplicon Analysis results table"""
columns = []
columns.append(Column(Constants.C_BC))
columns.append(Column(Constants.C_GOOD))
columns.append(Column(Constants.C_GOOD_PCT))
columns.append(Column(Constants.C_CHIM))
columns.append(Column(Constants.C_CHIM_PCT))
columns.append(Column(Constants.C_NOISE))
columns.append(Column(Constants.C_NOISE_PCT))
t = Table(Constants.T_R, columns=columns)
COL_IDS = [Constants.C_GOOD, Constants.C_GOOD_PCT, Constants.C_CHIM,
Constants.C_CHIM_PCT, Constants.C_NOISE, Constants.C_NOISE_PCT]
def add_column(barcode_id, n_good, n_chimera, n_noise):
pct_good = pct_chimera = pct_noise = 0
total = n_good + n_chimera + n_noise
if total > 0:
pct_good = n_good / float(total)
pct_chimera = n_chimera / float(total)
pct_noise = n_noise / float(total)
values = [n_good, pct_good, n_chimera, pct_chimera, n_noise, pct_noise]
t.add_data_by_column_id(Constants.C_BC, bc_id)
for column_id, value in zip(COL_IDS, values):
t.add_data_by_column_id(column_id, value)
with open(summary_csv) as csv_in:
reader = csv.reader(csv_in, delimiter=',')
reader.next()
for rec in reader:
assert len(rec) == 7, rec
bc_id = rec[0]
if bc_id == "All":
continue
add_column(bc_id, int(rec[1]), int(rec[3]), int(rec[5]))
n_good = sum(t.get_column_by_id(Constants.C_GOOD).values)
n_chimera = sum(t.get_column_by_id(Constants.C_CHIM).values)
n_noise = sum(t.get_column_by_id(Constants.C_NOISE).values)
add_column("All", n_good, n_chimera, n_noise)
return t
def _to_report(bg,
job_output_dir,
job_id,
state,
was_successful,
run_time,
error_message=None,
report_uuid=None):
""" High Level Report of the workflow state
Write the output of workflow datastore to pbreports report object
Workflow summary .dot/svg (collapsed workflow)
Workflow details .dot/svg (chunked workflow)
To add:
- Resolved WorkflowSettings (e.g., nproc, max_workers)
-
:type bg: BindingsGraph
"""
emsg = "" if error_message is None else error_message
columns = [
Column('task_id', header='Task id'),
Column('was_successful', header='Was Successful'),
Column('state', header="Task State"),
Column('run_time_sec', header="Run Time (sec)"),
Column('nproc', header="# of procs"),
Column("num_core_hours", header="Core Hours")
]
tasks_table = Table('tasks', title="Tasks", columns=columns)
for tnode in bg.all_task_type_nodes():
nproc = bg.node[tnode]['nproc']
# the task might not be completed.
run_time_sec = bg.node[tnode]['run_time']
if run_time_sec is None:
core_hours = 0.0
else:
core_hours = (run_time_sec / 60.0 / 60.0) * nproc
tasks_table.add_data_by_column_id('task_id', str(tnode))
tasks_table.add_data_by_column_id('nproc', bg.node[tnode]['nproc'])
tasks_table.add_data_by_column_id('state', bg.node[tnode]['state'])
tasks_table.add_data_by_column_id(
'was_successful', bg.node[tnode]['state'] == TaskStates.SUCCESSFUL)
# rt_ = bg.node[tnode]['run_time']
# rtime = None if rt_ is None else int(rt_)
tasks_table.add_data_by_column_id('run_time_sec',
bg.node[tnode]['run_time'])
tasks_table.add_data_by_column_id('num_core_hours',
round(core_hours, 4))
total_core_hours = sum(
tasks_table.get_column_by_id('num_core_hours').values)
attributes = [
Attribute('was_successful', was_successful, name="Was Successful"),
Attribute('total_run_time_sec', int(run_time), name="Walltime (sec)"),
Attribute('error_message', emsg, name="Error Message"),
Attribute('job_id', job_id, name="Job Id"),
Attribute('job_state', state, name="Job State"),
Attribute('job_output_dir',
job_output_dir,
name="Job Output Directory"),
Attribute('pbsmrtpipe_version',
pbsmrtpipe.get_version(),
name="pbsmrtpipe Version"),
Attribute('total_core_hours', round(total_core_hours, 4),
"Total core hours")
]
ep_table = _to_table("entry_points", bg, bg.entry_binding_nodes(),
"Entry Points")
fnodes_table = _to_table("file_node", bg, bg.file_nodes(), "File Nodes")
# this would be nice if the DataSet UUIDs of the entry-points are added to the
# dataset_uuids of the report.
report = Report('pbsmrtpipe',
tables=[tasks_table, ep_table, fnodes_table],
attributes=attributes,
uuid=report_uuid)
return report
