def _convert_to_tdf_resource(resource_dict, pkg_zipstream=None):
'''Convert a CKAN resource dict into a Tabular Data Format resource dict.
:param pkg_zipstream: If given and if the resource has a file uploaded to
the FileStore, then the file will be written to the zipstream and the
returned dict will contain "path" instead of "url".
:type pkg_zipstream: zipstream.ZipFile
'''
if pkg_zipstream and resource_dict.get('url_type') == 'upload':
name = resource_dict.get('name')
if not name:
# FIXME: Need to generate unique names (unique within the
# package) for unnamed files.
name = toolkit._('Unnamed file')
resource = {'path': name}
# Add the resource file itself into the ZIP file.
pkg_zipstream.write(util.get_path_to_resource_file(resource_dict),
arcname=resource['path'])
else:
resource = {'url': resource_dict['url']}
try:
schema_string = resource_dict.get('schema', '')
resource['schema'] = h.json.loads(schema_string)
except ValueError:
pass
return resource
def _convert_to_tdf_resource(resource_dict, pkg_zipstream=None):
'''Convert a CKAN resource dict into a Tabular Data Format resource dict.
:param pkg_zipstream: If given and if the resource has a file uploaded to
the FileStore, then the file will be written to the zipstream and the
returned dict will contain "path" instead of "url".
:type pkg_zipstream: zipstream.ZipFile
'''
if pkg_zipstream and resource_dict.get('url_type') == 'upload':
name = resource_dict.get('name')
if not name:
# FIXME: Need to generate unique names (unique within the
# package) for unnamed files.
name = toolkit._('Unnamed file')
resource = {'path': name}
# Add the resource file itself into the ZIP file.
pkg_zipstream.write(util.get_path_to_resource_file(resource_dict),
arcname=resource['path'])
else:
resource = {'url': resource_dict['url']}
try:
schema_string = resource_dict.get('schema', '')
resource['schema'] = h.json.loads(schema_string)
except ValueError:
pass
return resource
def csv_data(resource):
'''Return the CSV data for the given resource.
'''
try:
path = util.get_path_to_resource_file(resource)
except exceptions.ResourceFileDoesNotExistException:
return {'success': False,
'error': toolkit._("There's no uploaded file for this "
"resource")}
return _csv_data_from_file(open(path))
def csv_data(resource):
'''Return the CSV data for the given resource.
'''
try:
path = util.get_path_to_resource_file(resource)
except exceptions.ResourceFileDoesNotExistException:
return {
'success': False,
'error': toolkit._("There's no uploaded file for this "
"resource")
}
return _csv_data_from_file(open(path))
def test_get_path_to_resource_file_with_uploaded_file(self):
user = factories.User()
package = factories.Dataset(user=user)
api = ckanapi.TestAppCKAN(self.app, apikey=user['apikey'])
csv_file = custom_helpers.get_csv_file('test-data/datetimes.csv')
resource = api.action.resource_create(package_id=package['id'],
upload=csv_file)
path = util.get_path_to_resource_file(resource)
# Check that the path is correct by comparing the contents of the
# uploaded copy of the file to the original file.
assert open(path).read() == (
open(os.path.join(os.path.split(__file__)[0],
'../test-data/datetimes.csv')).read())
def test_get_path_to_resource_file_with_uploaded_file(self):
user = factories.User()
package = factories.Dataset(user=user)
api = ckanapi.TestAppCKAN(self.app, apikey=user['apikey'])
csv_file = custom_helpers.get_csv_file('test-data/datetimes.csv')
resource = api.action.resource_create(package_id=package['id'],
upload=csv_file)
path = util.get_path_to_resource_file(resource)
# Check that the path is correct by comparing the contents of the
# uploaded copy of the file to the original file.
assert open(path).read() == (open(
os.path.join(
os.path.split(__file__)[0],
'../test-data/datetimes.csv')).read())
def _infer_schema_for_resource(resource):
'''Return a JSON Table Schema for the given resource.
This will guess column headers and types from the resource's CSV file.
'''
# Note: Since this function is only called after uploading a file,
# we assume the resource does have an uploaded file and this line will not
# raise an exception.
path = util.get_path_to_resource_file(resource)
if not csv_utils.resource_is_csv_or_text_file(path):
helpers.flash_notice(
'This file does not seem to be a csv or text file. '
'You could try validating this file at http://csvlint.io'
)
try:
schema = csv_utils.infer_schema_from_csv_file(path)
except exceptions.CouldNotReadCSVException:
schema = {'fields': []}
return schema
data_dict, errors = dictization_functions.validate(data_dict,
schema.resource_schema_field_create_schema(), context)
except exceptions.InvalidResourceIDException, e:
raise toolkit.ValidationError(e)
if errors:
raise toolkit.ValidationError(errors)
resource_id = data_dict.pop('resource_id')
resource_dict = toolkit.get_action('resource_show')(context,
{'id': resource_id})
if data_dict.get('type') in ('date', 'time', 'datetime'):
try:
path = util.get_path_to_resource_file(resource_dict)
except exceptions.ResourceFileDoesNotExistException:
path = None
if path:
try:
data_dict['temporal_extent'] = csv_utils.temporal_extent(path,
column_num=data_dict['index'])
except ValueError:
pass
except TypeError:
pass
schema_ = toolkit.get_action('resource_schema_show')(context,
{'resource_id': resource_id})
schema_['fields'].append(data_dict)