def test_munge_name_multiple_pass(self):
'''Munging name multiple times produces same result.'''
for org, exp in self.munge_list:
first_munge = munge_name(org)
assert_equal(first_munge, exp)
second_munge = munge_name(first_munge)
assert_equal(second_munge, exp)
def test_munge_name_multiple_pass(self):
'''Munging name multiple times produces same result.'''
for org, exp in self.munge_list:
first_munge = munge_name(org)
nose_tools.assert_equal(first_munge, exp)
second_munge = munge_name(first_munge)
nose_tools.assert_equal(second_munge, exp)
def make_userdict(self):
return {
"email": self.email,
"name": munge.munge_name(self.name),
"sysadmin": self.is_sysadmin(),
"plugin_extras": {"drupal_idp": dataclasses.asdict(self)},
}
def taxonomy_create(context, data_dict):
"""
Creates a new taxonomy. Terms are not created here, they must be
created using taxonomy_term_create with the taxonomy id from this
call.
:param owner_org: the id of the dataset's owning organization, see
:returns: The newly created taxonomy
:rtype: A dictionary.
"""
_check_access('taxonomy_create', context, data_dict)
model = context['model']
name = data_dict.get('name')
title = logic.get_or_bust(data_dict, 'title')
uri = logic.get_or_bust(data_dict, 'uri')
if not name:
name = munge_name(title)
# Check the name has not been used
if model.Session.query(Taxonomy).filter(Taxonomy.name == name).count() > 0:
raise logic.ValidationError("Name is already in use")
t = Taxonomy(name=name, title=title, uri=uri)
model.Session.add(t)
model.Session.commit()
return t.as_dict()
def taxonomy_create(context, data_dict):
"""
Creates a new taxonomy. Terms are not created here, they must be
created using taxonomy_term_create with the taxonomy id from this
call.
:param owner_org: the id of the dataset's owning organization, see
:returns: The newly created taxonomy
:rtype: A dictionary.
"""
_check_access('taxonomy_create', context, data_dict)
model = context['model']
name = data_dict.get('name')
title = logic.get_or_bust(data_dict, 'title')
uri = logic.get_or_bust(data_dict, 'uri')
if not name:
name = munge_name(title)
# Check the name has not been used
if model.Session.query(Taxonomy).filter(Taxonomy.name == name).count() > 0:
raise logic.ValidationError("Name is already in use")
t = Taxonomy(name=name, title=title, uri=uri)
model.Session.add(t)
model.Session.commit()
return t.as_dict()
def _handle_rights_holder(self, dataset_dict, temp_dict, job):
try:
config = json.loads(job.source.config) if job.source.config else {}
except (TypeError, ValueError) as err:
log.warning('Cannot parse job config to get rights holder: %s',
err,
exc_info=err)
config = {}
orgs_conf = config.get('remote_orgs', None)
ctx = {'ignore_auth': True, 'user': self._get_user_name()}
if orgs_conf in ('create', ):
holder_name = dataset_dict.get('holder_name', None)
holder_identifier = dataset_dict.get('holder_identifier', None)
if holder_identifier and holder_name:
org = dcatapit_helpers\
.get_organization_by_identifier(ctx,
holder_identifier)
if not org:
org_dict = {
'identifier': holder_identifier,
'name': munge_name(holder_name),
'title': holder_name
}
act = p.toolkit.get_action('organization_create')
org = act(context=ctx, data_dict=org_dict)
dataset_dict['owner_org'] = org['name']
# remove holder fields, as this info will be handled in org
dataset_dict.pop('holder_name', None)
dataset_dict.pop('holder_identifier', None)
def _get_group(self, base_url, group_name):
url = base_url + self._get_rest_api_offset() + '/group/' + munge_name(group_name)
try:
content = self._get_content(url)
return json.loads(content)
except (ContentFetchError, ValueError):
log.debug('Could not fetch/decode remote group');
raise RemoteResourceError('Could not fetch/decode remote group')
def _get_group(self, base_url, group_name):
url = base_url + self._get_action_api_offset() + "/group_show?id=" + munge_name(group_name)
try:
content = self._get_content(url)
return json.loads(content)
except (ContentFetchError, ValueError):
log.debug("Could not fetch/decode remote group")
raise RemoteResourceError("Could not fetch/decode remote group")
def _get_group(self, base_url, group_name):
url = (base_url + self._get_rest_api_offset() + "/group/" +
munge_name(group_name))
try:
content = self._get_content(url)
return json.loads(content)
except (ContentFetchError, ValueError):
log.debug("Could not fetch/decode remote group")
raise RemoteResourceError("Could not fetch/decode remote group")
def _get_group(self, base_url, group_name):
url = base_url + self._get_action_api_offset() + '/group_show?id=' + \
munge_name(group_name)
try:
content = self._get_content(url)
return json.loads(content)
except (ContentFetchError, ValueError):
log.debug('Could not fetch/decode remote group')
raise RemoteResourceError('Could not fetch/decode remote group')
def improve_pkg_dict(self, pkg_dict, params, data=None):
if pkg_dict['name'] != '':
pkg_dict['name'] = munge_name(pkg_dict['name']).replace('_', '-')
else:
pkg_dict['name'] = munge_title_to_name(pkg_dict['title'])
if pkg_dict['url'] == '':
pkg_dict.pop('url', None)
# override the 'id' as this never matches the CKAN internal ID
pkg_dict['id'] = pkg_dict['name']
if params is not None and params.get(license, None) is not None:
pkg_dict['license_id'] = params['license']
else:
pkg_dict['license_id'] = config.get('ckanext.ddi.default_license')
# TODO: move all this to a interface method in ckanext-unhcr
# Save as draft to enable the stages on the form
# TODO: what about updates (if overriding)
pkg_dict['state'] = 'draft'
if data:
for field in ('owner_org', 'private'):
if field in data:
pkg_dict[field] = data[field]
pkg_dict['archived'] = 'False'
if pkg_dict.get('tags'):
pkg_dict['keywords'] = [tag['name'] for tag in pkg_dict['tags']]
if pkg_dict.get('unit_of_analysis'):
pkg_dict['unit_of_measurement'] = pkg_dict['unit_of_analysis']
if pkg_dict.get('data_collector'):
pkg_dict['data_collector'] = _get_data_collector_values(
pkg_dict['data_collector'])
if pkg_dict.get('data_collection_technique'):
pkg_dict[
'data_collection_technique'] = _get_data_collection_technique_value(
pkg_dict['data_collection_technique'])
if pkg_dict.get('id_number'):
pkg_dict['original_id'] = pkg_dict['id_number']
if pkg_dict.get('abstract'):
pkg_dict['notes'] = pkg_dict['abstract']
if pkg_dict.get('abbreviation'):
pkg_dict['short_title'] = pkg_dict['abbreviation']
pkg_dict['ddi'] = True
return pkg_dict
def test_into_user(self, details_data):
details = utils.Details(**details_data)
userdict = details.make_userdict()
assert userdict == {
"email": details_data["email"],
"name": munge.munge_name(details_data["name"]),
"sysadmin": False,
"plugin_extras": {
"drupal_idp": details_data
},
}
def improve_pkg_dict(self, pkg_dict, params):
if pkg_dict['name'] != '':
pkg_dict['name'] = munge_name(pkg_dict['name']).replace('_', '-')
else:
pkg_dict['name'] = munge_title_to_name(pkg_dict['title'])
if pkg_dict['url'] == '':
pkg_dict.pop('url', None)
# override the 'id' as this never matches the CKAN internal ID
pkg_dict['id'] = pkg_dict['name']
if params is not None and params.get(license, None) is not None:
pkg_dict['license_id'] = params['license']
else:
pkg_dict['license_id'] = config.get('ckanext.ddi.default_license')
return pkg_dict
def _create_or_update_package(self, base_name, metadata):
pkg_name = munge_name(base_name.replace('#', '_'))
extras_list = self._generate_extras(metadata)
pkg_dict = {
'name': pkg_name,
'title': metadata.get('title', base_name),
'notes': metadata.get('doc_excerpt', None),
'tags': metadata.get('tags', None),
'maintainer': metadata.get('publisher', None),
'author': metadata.get('creator', None),
'extras': extras_list,
'resources': [],
}
pprint(pkg_dict)
try:
print "pkg_name: %s" % pkg_name
pkg = self.ckan.action.package_show(id=pkg_name)
pkg.update(pkg_dict)
self.ckan.call_action('package_update', pkg)
except ckanapi.NotFound:
pkg = self.ckan.call_action('package_create', pkg_dict)
return pkg
class DCATAPITHarvesterPlugin(p.SingletonPlugin):
p.implements(IDCATRDFHarvester, inherit=True)
def before_download(self, url, harvest_job):
return url, []
def update_session(self, session):
return session
def after_download(self, content, harvest_job):
return content, []
def before_update(self, harvest_object, dataset_dict, temp_dict):
self._before(dataset_dict, temp_dict, harvest_object)
def after_update(self, harvest_object, dataset_dict, temp_dict):
return self._after(dataset_dict, temp_dict)
def before_create(self, harvest_object, dataset_dict, temp_dict):
self._before_create(harvest_object, dataset_dict)
self._before(dataset_dict, temp_dict, harvest_object)
def after_create(self, harvest_object, dataset_dict, temp_dict):
return self._after(dataset_dict, temp_dict)
def _before_create(self, harvest_object, dataset_dict):
title = dataset_dict['title']
name = HarvesterBase._gen_new_name(title)
if not name:
raise Exception('Could not generate a unique name '
'from the title or the GUID. Please '
'choose a more unique title.')
dataset_dict['name'] = name
def _before(self, dataset_dict, temp_dict, job):
loc_dict = dataset_dict.pop(LOCALISED_DICT_NAME_BASE, {})
res_dict = dataset_dict.pop(LOCALISED_DICT_NAME_RESOURCES, {})
if loc_dict or res_dict:
temp_dict['dcatapit'] = {
LOCALISED_DICT_NAME_BASE: loc_dict,
LOCALISED_DICT_NAME_RESOURCES: res_dict
}
self._handle_rights_holder(dataset_dict, temp_dict, job)
def _handle_rights_holder(self, dataset_dict, temp_dict, job):
try:
config = json.loads(job.source.config) if job.source.config else {}
except (
TypeError,
ValueError,
), err:
log.warning("Cannot parse job config to get rights holder: %s",
err,
exc_info=err)
config = {}
orgs_conf = config.get('remote_orgs', None)
ctx = {'ignore_auth': True, 'user': self._get_user_name()}
if orgs_conf in ('create', ):
holder_name = dataset_dict.get('holder_name', None)
holder_identifier = dataset_dict.get('holder_identifier', None)
if holder_identifier and holder_name:
org = dcatapit_helpers\
.get_organization_by_identifier(ctx,
holder_identifier)
if not org:
org_dict = {
'identifier': holder_identifier,
'name': munge_name(holder_name),
'title': holder_name
}
act = p.toolkit.get_action('organization_create')
org = act(context=ctx, data_dict=org_dict)
dataset_dict['owner_org'] = org['name']
# remove holder fields, as this info will be handled in org
dataset_dict.pop('holder_name', None)
dataset_dict.pop('holder_identifier', None)
def test_munge_name(self):
'''Munge a list of names gives expected results.'''
for org, exp in self.munge_list:
munge = munge_name(org)
nose_tools.assert_equal(munge, exp)
def test_munge(title, expected_munge):
munge = munge_name(title)
assert_equal(munge, expected_munge)
def munge_package_name(self):
name = request.params.get('name')
munged_name = munge.munge_name(name)
return self._finish_ok(munged_name)
orgs_conf = config.get('remote_orgs', None)
ctx = {'ignore_auth': True, 'user': self._get_user_name()}
if orgs_conf in ('create', ):
holder_name = dataset_dict.get('holder_name', None)
holder_identifier = dataset_dict.get('holder_identifier', None)
if holder_identifier and holder_name:
org = dcatapit_helpers.get_organization_by_identifier(
ctx, holder_identifier)
if not org:
org_dict = {
'identifier': holder_identifier,
'name': munge_name(holder_name),
'title': holder_name
}
org = p.toolkit.get_action('organization_create')(
context=ctx, data_dict=org_dict)
dataset_dict['owner_org'] = org['name']
# remove holder fields, as this info will be handled in org
dataset_dict.pop('holder_name', None)
dataset_dict.pop('holder_identifier', None)
def _after(self, dataset_dict, temp_dict):
dcatapit_dict = temp_dict.get('dcatapit')
if not dcatapit_dict:
return None
def test_remote_orgs(self):
dataset = {'title': 'some title 2',
'owner_id': self.org['id'],
'id': 'sometitle2',
'name': 'somename',
'holder_name': 'test holder',
'holder_identifier': 'abcdef',
'notes': 'some notes',
'modified': '2000-01-01',
'theme': 'AGRI',
'frequency': 'UNKNOWN',
'publisher_name': 'publisher',
'identifier': 'identifier2',
'publisher_identifier': 'publisher',
}
# no org creation, holder_identifier should be assigned to dataset
data = json.dumps(dataset)
harvest_dict = self._create_harvest_obj('http://mock/source/a',
name='testpkg_2',
config=json.dumps({'remote_orgs': 'no-create'}),
owner_org=self.org['id'],
)
harvest_obj = HarvestObject.get(harvest_dict['id'])
harvest_obj.content = data
h = DCATRDFHarvester()
out = h.import_stage(harvest_obj)
self.assertTrue(out, harvest_obj.errors)
pkg = helpers.call_action('package_show', context={}, name_or_id='some-title-2')
for k in ('holder_name', 'holder_identifier',):
self.assertEqual(pkg.get(k), dataset[k])
# check for new org
dataset.update({'id': 'sometitle3',
'name': munge_name('some title 3'),
'title': 'some title 3',
'holder_name': 'test test holder',
'holder_identifier': 'abcdefg',
'identifier': 'identifier3',
})
harvest_dict = self._create_harvest_obj('http://mock/source/b',
name='testpkg_3',
config=json.dumps({'remote_orgs': 'create'}),
owner_org=self.org['id'],
)
harvest_obj = HarvestObject.get(harvest_dict['id'])
harvest_obj.content = json.dumps(dataset)
out = h.import_stage(harvest_obj)
self.assertTrue(out, harvest_obj.errors)
pkg = helpers.call_action('package_show', context={}, name_or_id='testpkg_3')
self.assertTrue(out)
self.assertTrue(isinstance(out, bool))
pkg = helpers.call_action('package_show', context={}, name_or_id=dataset['name'])
org_id = pkg['owner_org']
self.assertIsNotNone(org_id)
org = helpers.call_action('organization_show', context={}, id=org_id)
self.assertEqual(org['identifier'], dataset['holder_identifier'])
# package's holder should be updated with organization's data
for k in (('holder_name', 'title',), ('holder_identifier', 'identifier',)):
self.assertEqual(pkg.get(k[0]), org[k[1]])
# check for existing org
dataset.update({'id': 'sometitle4',
'name': munge_name('some title 4'),
'title': 'some title 4',
'identifier': 'identifier4',
})
harvest_dict = self._create_harvest_obj('http://mock/source/c',
name='testpkg_4',
config=json.dumps({'remote_orgs': 'create'}),
owner_org=self.org['id'],
)
harvest_obj = HarvestObject.get(harvest_dict['id'])
harvest_obj.content = json.dumps(dataset)
out = h.import_stage(harvest_obj)
self.assertTrue(out, harvest_obj.errors)
pkg = helpers.call_action('package_show', context={}, name_or_id='testpkg_4')
self.assertTrue(isinstance(out, bool))
pkg = helpers.call_action('package_show', context={}, name_or_id=dataset['name'])
org_id = pkg['owner_org']
self.assertIsNotNone(org_id)
org = helpers.call_action('organization_show', context={}, id=org_id)
self.assertEqual(org['identifier'], dataset['holder_identifier'])
def munge_package_name(self):
name = request.params.get('name')
munged_name = munge.munge_name(name)
return self._finish_ok(munged_name)
def test_munge(title, expected_munge):
munge = munge_name(title)
assert_equal(munge, expected_munge)
def dataworld_name(title):
cleaned_title = ' '.join(title.split()).replace('_', '-').replace(' ', '-')
return munge_name('-'.join(filter(None, cleaned_title.split('-'))))
def test_munge_name_pass(original, expected):
"""Munging name multiple times produces same result."""
first_munge = munge_name(original)
assert first_munge == expected
second_munge = munge_name(first_munge)
assert second_munge == expected
def get_package_dict(self, context, data_dict):
package_dict = data_dict['package_dict']
iso_values = data_dict['iso_values']
harvest_object = data_dict['harvest_object']
source_config = json.loads(data_dict['harvest_object'].source.config)
xml_location_url = self._get_object_extra(data_dict['harvest_object'],
'waf_location')
xml_modified_date = self._get_object_extra(data_dict['harvest_object'],
'waf_modified_date')
# convert extras key:value list to dictinary
extras = {x['key']: x['value'] for x in package_dict.get('extras', [])}
extras['xml_location_url'] = xml_location_url
if xml_modified_date:
extras['xml_modified_date'] = xml_modified_date
# copy some fields over from iso_values if they exist
if (iso_values.get('limitations-on-public-access')):
extras['limitations-on-public-access'] = iso_values.get(
'limitations-on-public-access')
if (iso_values.get('access-constraints')):
extras['access-constraints'] = iso_values.get('access-constraints')
if (iso_values.get('use-constraints')):
extras['use-constraints'] = iso_values.get('use-constraints')
if (iso_values.get('use-constraints-code')):
extras['use-constraints-code'] = iso_values.get(
'use-constraints-code')
if (iso_values.get('legal-constraints-reference-code')):
extras['legal-constraints-reference-code'] = iso_values.get(
'legal-constraints-reference-code')
if (iso_values.get('distributor')):
extras['distributor'] = iso_values.get('distributor')
# load remote xml content
package_dict = _extract_xml_from_harvest_object(
package_dict, harvest_object)
# Handle Scheming, Composit, and Fluent extensions
loaded_plugins = plugins.toolkit.config.get("ckan.plugins")
if 'scheming_datasets' in loaded_plugins:
# composite = 'composite' in loaded_plugins
fluent = 'fluent' in loaded_plugins
log.debug(
'#### Scheming, Composite, or Fluent extensions found, processing dictinary ####'
)
schema = plugins.toolkit.h.scheming_get_dataset_schema('dataset')
# Package name, default harvester uses title or guid in that order.
# we want to reverse that order, so guid or title. Also use english
# title only for name
title_as_name = self.from_json(package_dict.get(
'title', '{}')).get('en', package_dict['name'])
name = munge.munge_name(extras.get('guid', title_as_name)).lower()
package_dict['name'] = name
# populate license_id
package_dict['license_id'] = iso_values.get(
'legal-constraints-reference-code') or iso_values.get(
'use-constraints') or 'CC-BY-4.0'
# populate citation
package_dict['citation'] = iso_values.get('citation')
# populate trlanslation method for bilingual field
notes_translation_method = iso_values.get(
'abstract_translation_method')
title_translation_method = iso_values.get(
'title_translation_method')
if notes_translation_method:
extras['notes_translation_method'] = notes_translation_method
if title_translation_method:
extras['title_translation_method'] = title_translation_method
# iterate over schema fields and update package dictionary as needed
for field in schema['dataset_fields']:
handled_fields = []
self.handle_composite_harvest_dictinary(
field, iso_values, extras, package_dict, handled_fields)
if fluent:
self.handle_fluent_harvest_dictinary(
field, iso_values, package_dict, schema,
handled_fields, source_config)
self.handle_scheming_harvest_dictinary(field, iso_values,
extras, package_dict,
handled_fields)
# populate resource format if missing
for resource in package_dict.get('resources', []):
if not resource.get('format'):
if (resource.get('resource_locator_protocol').startswith(
'http') or resource.get('url').startswith('http')):
resource['format'] = 'text/html'
# set default values
package_dict['progress'] = extras.get('progress', 'onGoing')
package_dict['frequency-of-update'] = extras.get(
'frequency-of-update', 'asNeeded')
extras_as_list = []
for key, value in extras.items():
if package_dict.get(key, ''):
log.error('extras %s found in package dict: key:%s value:%s',
key, key, value)
if isinstance(value, (list, dict)):
extras_as_list.append({'key': key, 'value': json.dumps(value)})
else:
extras_as_list.append({'key': key, 'value': value})
package_dict['extras'] = extras_as_list
# update resource format
resources = package_dict.get('resources', [])
if len(resources):
for resource in resources:
url = resource.get('url', '').strip()
format = resource.get('format') or ''
if url:
format = self.cioos_guess_resource_format(url) or format
resource['format'] = format
package_dict['resources'] = resources
return self.trim_values(package_dict)
def improve_pkg_dict(self, pkg_dict, params, data=None):
if pkg_dict['name'] != '':
pkg_dict['name'] = munge_name(pkg_dict['name']).replace('_', '-')
else:
pkg_dict['name'] = munge_title_to_name(pkg_dict['title'])
if pkg_dict['url'] == '':
pkg_dict.pop('url', None)
# override the 'id' as this never matches the CKAN internal ID
pkg_dict['id'] = pkg_dict['name']
if params is not None and params.get(license, None) is not None:
pkg_dict['license_id'] = params['license']
else:
pkg_dict['license_id'] = tk.config.get('ckanext.ddi.default_license')
# TODO: move all this to a interface method in ckanext-unhcr
# Save as draft to enable the stages on the form
# TODO: what about updates (if overriding)
pkg_dict['state'] = 'draft'
if data:
for field in (
'owner_org',
'private',
'visibility', # for resources
'license_id',
'external_access_level',
):
if field in data:
pkg_dict[field] = data[field]
pkg_dict['archived'] = 'False'
if pkg_dict.get('keywords'):
pkg_dict['keywords'] = _get_keywords(pkg_dict['keywords'])
if pkg_dict.get('unit_of_analysis'):
pkg_dict['unit_of_measurement'] = pkg_dict['unit_of_analysis']
if pkg_dict.get('data_collector'):
pkg_dict['data_collector'] = _get_data_collector_values(
pkg_dict['data_collector'])
if pkg_dict.get('data_collection_technique'):
pkg_dict['data_collection_technique'] = _get_data_collection_technique_value(
pkg_dict['data_collection_technique'])
if pkg_dict.get('id_number'):
pkg_dict['original_id'] = pkg_dict['id_number']
if pkg_dict.get('abstract'):
pkg_dict['notes'] = pkg_dict['abstract']
if pkg_dict.get('abbreviation'):
pkg_dict['short_title'] = pkg_dict['abbreviation']
if not pkg_dict.get('country_codes'):
# geographies fields is now a required field.
# We need to ensure any dataset to have a "geographies" field filled
pkg_dict['country_codes'] = tk.config.get(
'ckanext.ddi.default_country_code',
'UNSPECIFIED'
)
pkg_dict['ddi'] = True
return pkg_dict
def test_munge_name(self):
'''Munge a list of names gives expected results.'''
for org, exp in self.munge_list:
munge = munge_name(org)
assert_equal(munge, exp)
