def create_dataset():
"""Create a new dataset in the datastore for the project. The dataset
schema and rows are given in the request body. Dataset annotations are
optional. The expected request body format is:
{
"columns": [
{
"id": 0,
"name": "string",
"type": "string"
}
],
"rows": [
{
"id": 0,
"values": [
"string"
]
}
],
"annotations": [
{
"columnId": 0,
"rowId": 0,
"key": "string",
"value": "string"
}
]
}
"""
# Validate the request
obj = srv.validate_json_request(
request,
required=[labels.COLUMNS, labels.ROWS],
optional=[labels.ANNOTATIONS]
)
columns = deserialize.DATASET_COLUMNS(obj[labels.COLUMNS])
rows = [deserialize.DATASET_ROW(row) for row in obj[labels.ROWS]]
annotations = None
if labels.ANNOTATIONS in obj:
annotations = DatasetMetadata()
for anno in obj[labels.ANNOTATIONS]:
a = deserialize.ANNOTATION(anno)
if a.column_id is None:
annotations.rows.append(a)
elif a.row_id is None:
annotations.columns.append(a)
else:
annotations.cells.append(a)
try:
dataset = api.datasets.create_dataset(
project_id=config.project_id,
columns=columns,
rows=rows,
annotations=annotations
)
return jsonify(dataset)
except ValueError as ex:
raise srv.InvalidRequest(str(ex))
def get_dataset_descriptor(dataset_id):
"""Get the descriptor for the dataset with given identifier."""
try:
dataset = api.datasets.get_dataset_descriptor(
project_id=config.project_id, dataset_id=dataset_id)
if not dataset is None:
return jsonify(dataset)
except ValueError as ex:
raise srv.InvalidRequest(str(ex))
raise srv.ResourceNotFound('unknown dataset \'' + dataset_id + '\'')
def get_dataset(dataset_id):
"""Get the dataset with given identifier that has been generated by a
curation workflow.
"""
# Get dataset rows with offset and limit parameters
try:
dataset = api.datasets.get_dataset(
project_id=config.project_id,
dataset_id=dataset_id,
offset=request.args.get(PAGE_OFFSET),
limit=request.args.get(PAGE_LIMIT))
if not dataset is None:
return jsonify(dataset)
except ValueError as ex:
raise srv.InvalidRequest(str(ex))
raise srv.ResourceNotFound('unknown dataset \'' + dataset_id + '\'')
def get_dataset_chart_view(branch_id, workflow_id, module_id, chart_id):
"""Get content of a dataset chart view for a given workflow module.
"""
try:
view = api.views.get_dataset_chart_view(project_id=config.project_id,
branch_id=branch_id,
workflow_id=workflow_id,
module_id=module_id,
chart_id=chart_id)
except ValueError as ex:
raise srv.InvalidRequest(str(ex))
if not view is None:
return jsonify(view)
raise srv.ResourceNotFound(''.join([
'unknown branch \'' + branch_id, '\', workflow \'' + workflow_id,
'\', module \'' + module_id, '\' or chart \'' + chart_id + '\''
]))
def update_dataset_annotation(dataset_id):
"""Update an annotation that is associated with a component of the given
dataset.
Request
-------
{
"columnId": 0,
"rowId": 0,
"key": "string",
"oldValue": "string", or "int", or "float"
"newValue": "string", or "int", or "float"
}
"""
# Validate the request
obj = srv.validate_json_request(
request,
required=['key'],
optional=['columnId', 'rowId', 'key', 'oldValue', 'newValue']
)
# Create update statement and execute. The result is None if no dataset with
# given identifier exists.
key = obj[labels.KEY] if labels.KEY in obj else None
column_id = obj[labels.COLUMN_ID] if labels.COLUMN_ID in obj else None
row_id = obj[labels.ROW_ID] if labels.ROW_ID in obj else None
old_value = obj[labels.OLD_VALUE] if labels.OLD_VALUE in obj else None
new_value = obj[labels.NEW_VALUE] if labels.NEW_VALUE in obj else None
try:
annotations = api.datasets.update_annotation(
project_id=config.project_id,
dataset_id=dataset_id,
key=key,
column_id=column_id,
row_id=row_id,
old_value=old_value,
new_value=new_value
)
if not annotations is None:
return jsonify(annotations)
except ValueError as ex:
raise srv.InvalidRequest(str(ex))
raise srv.ResourceNotFound('unknown dataset \'' + dataset_id + '\'')
