def get_record_type(cls, xml):
'''
For a given ISO19139 record, returns the "type"
e.g. "dataset", "series", "service"
xml - etree of the ISO19139 XML record
'''
gemini = GeminiDocument(xml_tree=xml)
return gemini.read_value('resource-type')
def detect(cls):
'''Finds datasets that are coupled and adds their
harvest_source_reference to the HarvestObject and package extras.
'''
from ckan.lib.base import json
from ckan import model
from ckanext.harvest.model import HarvestObject
from ckanext.spatial.model import GeminiDocument
from ckanext.spatial.lib.coupled_resource import extract_guid
# Find service records
for service_record in model.Session.query(model.Package).\
filter_by(state='active').\
join(model.PackageExtra).\
filter_by(state='active').\
filter_by(key='resource-type').\
filter_by(value='service'):
# Find coupled dataset records
service_type = service_record.extras['resource-type']
if not 'coupled-resource' in service_record.extras:
if service_type in ('view', 'download'):
service_stats.add('No coupled-resource extra for %s type (where it is mandatory)', service_record.name, service_type)
else:
service_stats.add('No coupled-resource extra (but not mandatory for this service type)', service_record.name)
continue
coupled_resources_str = service_record.extras['coupled-resource']
coupled_resources = json.loads(coupled_resources_str)
log.info('%s has %i coupled resources',
service_record.name, len(coupled_resources))
couples_all_detected = True
couples_detected = False
for i, coupled_resource in enumerate(coupled_resources):
couple_id = '%s.%s' % (service_record.name, i)
href = coupled_resource['href']
# For tests only
#if href != ['http://www.ordnancesurvey.co.uk/oswebsite/xml/products/Topo.xml']:
# break
if len(href) <> 1:
log.error('Coupled resource href is not a list of 1: %r couple=%s',
href, couple_id)
couple_stats.add('Couple href is length %i' % len(href), couple_id)
couples_all_detected = False
continue
href = href[0]
if not href.strip():
log.error('Coupled resource href is blank. couple=%s',
couple_id)
couple_stats.add('Couple href is blank', couple_id)
couples_all_detected = False
continue
# Look for the equivalent dataset resource
# If it is CSW, we must extract the guid
# Example CSW url: http://ogcdev.bgs.ac.uk/geonetwork/srv/en/csw?SERVICE=CSW&REQUEST=GetRecordById&ID=9df8df52-d788-37a8-e044-0003ba9b0d98&elementSetName=full&OutputSchema=http://www.isotc211.org/2005/gmd
guid = extract_guid(href)
if guid:
if not guid.strip():
couple_stats.add('Guid was blank', couple_id)
log.error('Guid was blank. href=%s', href, couple_id)
try:
harvest_object = cls.find_harvest_object_by_guid(guid)
except FindError, e:
log.error('%s guid=%s couple=%s', e, guid, couple_id)
couple_stats.add(str(e), couple_id)
couples_all_detected = False
continue
dataset_record = harvest_object.package #res.resource_group.package
couple_stats.add('Couple completed', couple_id)
log.info('Couple completed %s <-> %s',
service_record.name, dataset_record.name)
cls.add_coupling(service_record, dataset_record, harvest_object, guid)
couples_detected = True
continue
# Known bad couples are weeded out
bad_couples = ('GetCapabilities', 'CEH:EIDC',
'ceh:eidc',
'http://data.nbn.org.uk#',
'www.geostore.com/OGC/OGCInterface',
'spatialni.gov.uk/arcgis/services/LPS/CadastreNI/MapServer/WMSServer',
'Please enter a valid url',
)
bad_couple_detected = False
for bad_couple in bad_couples:
if bad_couple in href:
couple_stats.add('Invalid couple (%s)' % bad_couple, couple_id)
log.info('Invalid couple (%s): %s couple=%s', bad_couple, href, couple_id)
bad_couple_detected = True
if bad_couple_detected:
couples_all_detected = False
continue
# Try as a WAF
# Try the URL to download the gemini again, to find the
# GUID of the dataset
log.info('Trying possible WAF href: %s' % href)
try:
res = requests.get(href, timeout=10)
except Exception, e:
couple_stats.add('Connecting to href failed: %s' % \
e, couple_id)
log.warning('Connecting to href failed: %s href:"%s"', \
e, href)
couples_all_detected = False
break
if not res.ok:
couple_stats.add('Resolving href failed: %s' % \
res.reason, couple_id)
log.warning('Resolving href failed: %s %s href:"%s"', \
res.status_code, res.reason, href)
couples_all_detected = False
break
gemini = GeminiDocument(res.content)
try:
guid = gemini.read_value('guid')
except KeyError, e:
couple_stats.add('Could not get GUID from Gemini downloaded' % \
href, couple_id)
log.warning('Could not get GUID from Gemini downloaded href:"%s"', \
href)
couples_all_detected = False
break
def detect(cls):
'''Finds datasets that are coupled and adds their
harvest_source_reference to the HarvestObject and package extras.
'''
from ckan.lib.base import json
from ckan import model
from ckanext.harvest.model import HarvestObject
from ckanext.spatial.model import GeminiDocument
from ckanext.spatial.lib.coupled_resource import extract_guid
# Find service records
for service_record in model.Session.query(model.Package).\
filter_by(state='active').\
join(model.PackageExtra).\
filter_by(state='active').\
filter_by(key='resource-type').\
filter_by(value='service'):
# Find coupled dataset records
service_type = service_record.extras['resource-type']
if not 'coupled-resource' in service_record.extras:
if service_type in ('view', 'download'):
service_stats.add('No coupled-resource extra for %s type (where it is mandatory)', service_record.name, service_type)
else:
service_stats.add('No coupled-resource extra (but not mandatory for this service type)', service_record.name)
continue
coupled_resources_str = service_record.extras['coupled-resource']
coupled_resources = json.loads(coupled_resources_str)
log.info('%s has %i coupled resources',
service_record.name, len(coupled_resources))
couples_all_detected = True
couples_detected = False
for i, coupled_resource in enumerate(coupled_resources):
couple_id = '%s.%s' % (service_record.name, i)
href = coupled_resource['href']
# For tests only
#if href != ['http://www.ordnancesurvey.co.uk/oswebsite/xml/products/Topo.xml']:
# break
if len(href) <> 1:
log.error('Coupled resource href is not a list of 1: %r couple=%s',
href, couple_id)
couple_stats.add('Couple href is length %i' % len(href), couple_id)
couples_all_detected = False
continue
href = href[0]
if not href.strip():
log.error('Coupled resource href is blank. couple=%s',
couple_id)
couple_stats.add('Couple href is blank', couple_id)
couples_all_detected = False
continue
# Look for the equivalent dataset resource
# If it is CSW, we must extract the guid
# Example CSW url: http://ogcdev.bgs.ac.uk/geonetwork/srv/en/csw?SERVICE=CSW&REQUEST=GetRecordById&ID=9df8df52-d788-37a8-e044-0003ba9b0d98&elementSetName=full&OutputSchema=http://www.isotc211.org/2005/gmd
guid = extract_guid(href)
if guid:
if not guid.strip():
couple_stats.add('Guid was blank', couple_id)
log.error('Guid was blank. href=%s', href, couple_id)
try:
harvest_object = cls.find_harvest_object_by_guid(guid)
except FindError, e:
log.error('%s guid=%s couple=%s', e, guid, couple_id)
couple_stats.add(str(e), couple_id)
couples_all_detected = False
continue
dataset_record = harvest_object.package #res.resource_group.package
couple_stats.add('Couple completed', couple_id)
log.info('Couple completed %s <-> %s',
service_record.name, dataset_record.name)
cls.add_coupling(service_record, dataset_record, harvest_object, guid)
couples_detected = True
continue
# Known bad couples are weeded out
bad_couples = ('GetCapabilities', 'CEH:EIDC',
'ceh:eidc',
'http://data.nbn.org.uk#',
'www.geostore.com/OGC/OGCInterface',
'spatialni.gov.uk/arcgis/services/LPS/CadastreNI/MapServer/WMSServer',
'Please enter a valid url',
)
bad_couple_detected = False
for bad_couple in bad_couples:
if bad_couple in href:
couple_stats.add('Invalid couple (%s)' % bad_couple, couple_id)
log.info('Invalid couple (%s): %s couple=%s', bad_couple, href, couple_id)
bad_couple_detected = True
if bad_couple_detected:
couples_all_detected = False
continue
# Try as a WAF
# Try the URL to download the gemini again, to find the
# GUID of the dataset
log.info('Trying possible WAF href: %s' % href)
try:
res = requests.get(href, timeout=10)
except Exception, e:
couple_stats.add('Connecting to href failed: %s' % \
e, couple_id)
log.warning('Connecting to href failed: %s href:"%s"', \
e, href)
couples_all_detected = False
break
if not res.ok:
couple_stats.add('Resolving href failed: %s' % \
res.reason, couple_id)
log.warning('Resolving href failed: %s %s href:"%s"', \
res.status_code, res.reason, href)
couples_all_detected = False
break
gemini = GeminiDocument(res.content)
try:
guid = gemini.read_value('guid')
except KeyError, e:
couple_stats.add('Could not get GUID from Gemini downloaded' % \
href, couple_id)
log.warning('Could not get GUID from Gemini downloaded href:"%s"', \
href)
couples_all_detected = False
break
