def get_current_ckan_resources_from_api(harvest_source_id):
results_json_path = config.get_ckan_results_cache_path()
logger.info(f'Extracting from harvest source id: {harvest_source_id}')
cpa = CKANPortalAPI(base_url=config.CKAN_CATALOG_URL)
resources = 0
page = 0
for datasets in cpa.search_harvest_packages(
harvest_source_id=harvest_source_id):
# getting resources in pages of packages
page += 1
logger.info('PAGE {} from harvest source id: {}'.format(
page, harvest_source_id))
for dataset in datasets:
pkg_resources = len(dataset['resources'])
resources += pkg_resources
yield (dataset)
# we don't need to save this
# save_dict_as_data_packages(data=package, path=config.get_data_packages_folder_path(),
# prefix='ckan-result',
# identifier_field='id')
logger.info('{} total resources in harvest source id: {}'.format(
resources, harvest_source_id))
cpa.save_packages_list(path=results_json_path)
def test_load_from_url(self):
cpa = CKANPortalAPI(base_url=CKAN_BASE_URL)
resources = 0
page = 0
for packages in cpa.search_harvest_packages(
harvest_source_id=HARVEST_SOURCE_ID):
page += 1
print(f'API packages search page {page}')
self.assertGreater(cpa.total_packages,
0) # has resources in the first page
break # do not need more
def get_current_ckan_resources_from_api(harvest_source_id=None):
logger.info('Extracting from harvest source id: {}'.format(harvest_source_id))
cpa = CKANPortalAPI()
resources = 0
page = 0
for packages in cpa.search_harvest_packages(harvest_source_id=harvest_source_id):
# getting resources in pages of packages
page += 1
logger.info('PAGE {} from harvest source id: {}'.format(page, harvest_source_id))
for package in packages:
pkg_resources = len(package['resources'])
resources += pkg_resources
yield(package)
logger.info('{} total resources'.format(resources))
def test_create_harvest_source(self):
logger.info('Creating harvest source')
cpa = CKANPortalAPI(base_url=CKAN_BASE_URL, api_key=CKAN_API_KEY)
cpa.delete_all_harvest_sources(harvest_type='harvest',
source_type='datajson')
title = 'Energy JSON test {}'.format(random.randint(1, 999999))
url = 'http://www.energy.gov/data-{}.json'.format(
random.randint(1, 999999))
res = cpa.create_harvest_source(
title=title,
url=url,
owner_org_id=CKAN_ORG_ID,
source_type='datajson',
notes='Some tests about local harvesting sources creation',
frequency='WEEKLY')
self.assertTrue(res['success'])
harvest_source = res['result']
logger.info('Created: {}'.format(res['success']))
# read it
res = cpa.show_package(ckan_package_id_or_name=harvest_source['id'])
self.assertTrue(res['success'])
self.assertEqual(harvest_source['url'], url)
self.assertEqual(harvest_source['title'], title)
self.assertEqual(harvest_source['type'], 'harvest')
self.assertEqual(harvest_source['source_type'], 'datajson')
# search for it
results = cpa.search_harvest_packages(rows=1000,
harvest_type='harvest',
source_type='datajson')
created_ok = False
for datasets in results:
for dataset in datasets:
# print('FOUND: {}'.format(dataset['name']))
if dataset['name'] == harvest_source['name']:
created_ok = True
logger.info('Found!')
else:
logger.info('Other harvest source: {}'.format(
dataset['name']))
assert created_ok == True
# create a dataset with this harvest_soure_id
dataset_title = 'Dataset number {}'.format(random.randint(1, 999999))
dataset_name = slugify(dataset_title)
tags = [{'name': 'tag81'}, {'name': 'tag82'}]
randval = random.randint(1, 999)
extras = [
{
'key': 'harvest_source_id',
'value': harvest_source['id']
},
{
'key': 'harvest_source_title',
'value': harvest_source['title']
},
# {'key': 'harvest_object_id', 'value': harvest_source['id']}, # ? not sure
{
'key': 'harvest_ng_source_id',
'value': harvest_source['id']
},
{
'key': 'harvest_ng_source_title',
'value': harvest_source['title']
},
{
'key': 'try_a_extra',
'value': randval
}
]
package = {
'name': dataset_name,
'title': dataset_title,
'owner_org': CKAN_ORG_ID,
'tags': tags,
'extras': extras
}
res2 = cpa.create_package(ckan_package=package)
self.assertTrue(res2['success'])
logger.info('Package with harvest source: {}'.format(res2['success']))
# read full dataset
res3 = cpa.show_package(ckan_package_id_or_name=dataset_name)
self.assertTrue(res3['success'])
ckan_dataset = res3['result']
logger.info(
'Package with harvest source readed: {}'.format(ckan_dataset))
assert 'extras' in ckan_dataset
assert [str(randval)] == [
extra['value'] for extra in ckan_dataset['extras']
if extra['key'] == 'try_a_extra'
]
# my custom ID (not connected to a real harvest ID)
assert [harvest_source['id']] == [
extra['value'] for extra in ckan_dataset['extras']
if extra['key'] == 'harvest_ng_source_id'
]
# check if this package is related to harvest source
total_datasets_in_source = 0
datasets_from_source = cpa.search_harvest_packages(
harvest_source_id=harvest_source['id'])
connected_ok = False
for datasets in datasets_from_source:
for dataset in datasets:
total_datasets_in_source += 1
if dataset['name'] == dataset_name:
connected_ok = True
logger.info('Found!')
else:
# we just expect one dataset
error = '{} != {} ------ {}'.format(
dataset['name'], dataset_name, dataset)
logger.error(error)
assert error == False
assert connected_ok == True
assert total_datasets_in_source == 1
logger.info(
f' +++++++++++++ total_datasets_in_source={total_datasets_in_source}'
)
# this fails, harvest process is more complex that just add an extra
# assert [harvest_source['id']] == [extra['value'] for extra in ckan_dataset['extras'] if extra['key'] == 'harvest_source_id']
# delete both
logger.info('Delete CKAN package: {}'.format(ckan_dataset['id']))
res4 = cpa.delete_package(ckan_package_id_or_name=ckan_dataset['id'])
self.assertTrue(res4['success'])
logger.info('Delete Harvest source: {}'.format(harvest_source['id']))
res5 = cpa.delete_package(ckan_package_id_or_name=harvest_source['id'])
self.assertTrue(res5['success'])
'schema_version', 'total_dataset', 'total_resources', 'dataset_types',
'resource_types'
]
writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
writer.writeheader()
results = []
colections_ids = set()
c = 0
urls = []
with_configs = 0
with_config_filters = 0
with_config_defaults = 0
for results in cpa.search_harvest_packages(harvest_type='harvest',
method='GET'
#,source_type='datajson'
):
for local_harvest_source in results:
url = local_harvest_source['url']
if url in urls:
logger.error(
'------------------\n ALREADY READED\n------------------')
continue
else:
urls.append(url)
c += 1
name = local_harvest_source.get('name', 'UNNAMED')
hspath = config.get_harvest_sources_path(hs_name=name)
f = open(hspath, 'w')
parser.add_argument("--source_type",
type=str,
default='datajson',
help="Tipe of harvest source: datajson|csw|waf etc")
parser.add_argument("--method",
type=str,
default='GET',
help="POST fails on CKAN 2.3, now is working")
args = parser.parse_args()
cpa = CKANPortalAPI(base_url=CKAN_BASE_URL, api_key=CKAN_API_KEY)
total_sources = cpa.import_harvest_sources(catalog_url=args.import_from_url,
method=args.method,
on_duplicated='DELETE',
harvest_type=args.harvest_type,
source_type=args.source_type,
delete_local_harvest_sources=True)
# search
total_searched = 0
for harvest_sources in cpa.search_harvest_packages(
method='POST',
harvest_type=args.harvest_type,
source_type=args.source_type):
for harvest_source in harvest_sources:
total_searched += 1
assert total_sources == total_searched
templated_harvest_command = """
source {{ params.env_path }}/bin/activate
cd {{ params.app_path }}
python harvest.py \
--name {{ params.name }} \
--url {{ params.data_json_url }} \
--harvest_source_id {{ params.harvest_source_id }} \
--ckan_owner_org_id {{ params.ckan_org_id }} \
--catalog_url {{ params.catalog_url }} \
--ckan_api_key {{ params.ckan_api_key }} \
--limit_dataset 10 # limit for test, remove for production
"""
results = cpa.search_harvest_packages(rows=1000,
harvest_type='harvest',
source_type='datajson')
for datasets in results:
for harvest_source in datasets:
frequency = harvest_source.get('frequency', 'MONTHLY').upper()
if frequency not in valid_frequencies:
raise Exception(f'Unknown frequency: {frequency}')
url = harvest_source['url']
if url in urls: # avoid duplicates
continue
urls.append(url)
organization = harvest_source['organization']
name = harvest_source['name']
