def create_annotation_table(connection,
table_name,
column_names,
column_descriptions,
values,
namespace=None,
table_description=None):
"""Creates a table annotation from a list of lists"""
table_name = f'{table_name}_{"".join([choice(ascii_letters) for n in range(32)])}.h5'
columns = _create_table(column_names=column_names,
columns_descriptions=column_descriptions,
values=values)
resources = connection.c.sf.sharedResources()
repository_id = resources.repositories().descriptions[0].getId().getValue()
table = resources.newTable(repository_id, table_name)
table.initialize(columns)
table.addData(columns)
original_file = table.getOriginalFile()
table.close() # when we are done, close.
file_ann = gw.FileAnnotationWrapper(connection)
file_ann.setNs(namespace)
file_ann.setDescription(table_description)
file_ann.setFile(model.OriginalFileI(
original_file.id.val, False)) # TODO: try to get this with a wrapper
file_ann.save()
return file_ann
def create_table(self, dataset_id, dataframe, table_name):
columns = dataframe.columns
resources = self.SESSION.sharedResources()
# resources = conn.c.sf.sharedResources()
repository_id = resources.repositories().descriptions[0].getId(
).getValue()
# create columns and data types, and populate with data
data_types = dataframe.dtypes
table_data = []
init_cols = []
for index, col in enumerate(dataframe.columns):
if dataframe[col].dtype == object:
max_len = dataframe[col].str.len().max()
init_col = grid.StringColumn(col, '', max_len, [])
init_cols.append(init_col)
data_col = grid.StringColumn(
col, '', max_len, list(dataframe.iloc[:, index].values))
table_data.append(data_col)
table = resources.newTable(repository_id,
''.join(["/", table_name, ".h5"]))
# table = resources.newTable(dataset_id, table_name)
table.initialize(init_cols)
table.addData(table_data)
# note that this worked after the table.close() statement was invoked in 5.4.10
orig_file = table.getOriginalFile()
table.close() # when we are done, close.
orig_file_id = orig_file.id.val
# ...so you can attach this data to an object e.g. Dataset
file_ann = model.FileAnnotationI()
# use unloaded OriginalFileI
file_ann.setFile(model.OriginalFileI(orig_file_id, False))
file_ann = self.SESSION.getUpdateService().saveAndReturnObject(
file_ann)
link = model.DatasetAnnotationLinkI()
link.setParent(model.DatasetI(dataset_id, False))
link.setChild(model.FileAnnotationI(file_ann.getId().getValue(),
False))
table = self.SESSION.getUpdateService().saveAndReturnObject(link)
return table
def save_original_file(connection, name, path, mimetype, size, sha1):
of = get_original_file(connection, name, mimetype)
if of is None:
of = om.OriginalFileI()
of.setName(ot.wrap(name))
of.setPath(ot.wrap(path))
of.setMimetype(ot.wrap(mimetype))
of.setSize(ot.rlong(size))
of.setHash(ot.wrap(sha1))
hasher = om.ChecksumAlgorithmI()
hasher.setValue(ot.wrap(ChecksumAlgorithmSHA1160))
of.setHasher(hasher)
of = connection.getUpdateService().saveAndReturnObject(of)
return of.getId().getValue()
else:
raise DuplicatedEntryError(
'OriginalFile with name %s and mimetype %s already exists' %
(name, mimetype))