Former-commit-id: 3bb1449efe6725785ffb74bfe250e27913e47885
--- a/documents/datagov-export.py
+++ b/documents/datagov-export.py
@@ -3,7 +3,7 @@
from ckanclient import CkanApiError
import re
import html2text # aaronsw :(
-import ckanapi
+import ckanapi # https://github.com/open-data/ckanapi
class LoaderError(Exception):
@@ -11,9 +11,9 @@
# Instantiate the CKAN client.
#ckan = ckanclient.CkanClient(base_location='http://localhost:5000/api', api_key='b47b24cd-591d-40c1-8677-d73101d56d1b')
-api_key = 'aeab80a6-0a58-4026-96a4-c07e57f9c434'
+api_key = 'b3ab75e4-afbb-465b-a09d-8171c8c69a7a'
ckan = ckanclient.CkanClient(base_location='http://data.disclosurelo.gs/api',
- api_key=api_key)
+ api_key=api_key)
ckandirect = ckanapi.RemoteCKAN('http://data.disclosurelo.gs', api_key=api_key)
couch = couchdb.Server('http://127.0.0.1:5984/')
#couch = couchdb.Server('http://192.168.1.113:5984/')
@@ -27,6 +27,7 @@
'iec_ext': ('byte', 'kibi', 'mebi', 'gibi', 'tebi', 'pebi', 'exbi',
'zebi', 'yobi'),
}
+
def human2bytes(s):
"""
@@ -91,12 +92,9 @@
name = re.sub('__', '_', name).lower()
return name
-#todo "{'name': [u'Url must be purely lowercase alphanumeric (ascii) characters and these symbols: -_']}"
-# http://data.gov.au/dataset/australian-domestic-regional-and-international-airline-activity-%E2%80%93-time-series/
+
def name_munge(input_name):
- return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and'))
- #[:100]
- #return input_name.replace(' ', '').replace('.', '_').replace('&', 'and')
+ return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and'))
def get_licence_id(licencename):
@@ -117,15 +115,46 @@
raise Exception(licencename + " not found");
return map[licencename];
+
docsdb = couch['disclosr-documents']
if __name__ == "__main__":
orgs_list = []
+ orgs_ids = {}
for doc in docsdb.view('app/datasets'):
+ print " --- "
print doc.id
+
if doc.value['url'] != "http://data.gov.au/data/" and doc.value['agencyID'] != "qld":
+
+
# Collect the package metadata.
- pkg_name = filter( lambda x: x in '0123456789abcdefghijklmnopqrstuvwxyz-_',doc.value['url'].replace("http://data.gov.au/dataset/",'').replace('/','')[:100]);
+ pkg_name = filter(lambda x: x in '0123456789abcdefghijklmnopqrstuvwxyz-_',
+ doc.value['url'].replace("http://data.gov.au/dataset/", '').replace('/', '')[:100]);
+ print pkg_name
+ #add to or create organization using direct API
+ org_name = name_munge(doc.value['metadata']["Agency"][:100])
+ if org_name not in orgs_list:
+ orgs_list = ckandirect.action.organization_list()['result']
+ #print orgs_list
+ if org_name not in orgs_list:
+ try:
+ print "org not found, creating " + org_name
+ ckandirect.action.organization_create(name=org_name, title=doc.value['metadata']["Agency"],
+ description=doc.value['metadata']["Agency"])
+ orgs_list.append(org_name)
+ except ckanapi.ValidationError, e:
+ print e
+ raise LoaderError('Unexpected status')
+ else:
+ print "org found, adding dataset to " + org_name
+
+ # cache org names -> id mapping
+ if org_name not in orgs_ids:
+ org = ckandirect.action.organization_show(id=org_name)
+ orgs_ids[org_name] = org["result"]["id"]
+ org_id = orgs_ids[org_name]
+ print "org id is "+org_id
tags = []
if doc.value['agencyID'] == "AGIMO":
if len(doc.value['metadata']["Keywords / Tags"]) > 0:
@@ -133,11 +162,7 @@
tags = tags + doc.value['metadata']["Keywords / Tags"]
else:
tags = tags + [doc.value['metadata']["Keywords / Tags"]]
- if 'data.gov.au Category' in doc.value['metadata'].keys() and len(doc.value['metadata']['data.gov.au Category']) > 0:
- if hasattr(doc.value['metadata']['data.gov.au Category'], '__iter__'):
- tags = tags + doc.value['metadata']['data.gov.au Category']
- else:
- tags = tags + [doc.value['metadata']['data.gov.au Category']]
+
tags = [re.sub('[^a-zA-Z0-9-_.]', '', tag.replace('&', 'and')).lower() for tag in tags if tag]
#print tags
package_entity = {
@@ -149,6 +174,8 @@
'maintainer': doc.value['metadata']["DCTERMS.Creator"],
'licence_id': get_licence_id(doc.value['metadata']['DCTERMS.License']),
'notes': html2text.html2text(doc.value['metadata']['Description']),
+ 'owner_org': org_id
+ #todo add missing key values like jurasdiction
}
if doc.value['agencyID'] == "qld":
package_entity = doc.value['metadata']
@@ -165,76 +192,44 @@
ckan.last_status, pkg_name, e.args))
pkg = ckan.package_entity_get(pkg_name)
- #add to group
-
- group_name = name_munge(doc.value['metadata']["Agency"][:100])
- try:
- print ckan.group_entity_get(group_name)
-
- # Update the group details
- group_entity = ckan.last_message
- print "group exists"
- if 'packages' in group_entity.keys():
- group_entity['packages'] = list(set(group_entity['packages'] + [pkg_name]))
+ # add dataset to group(s)
+ groups = []
+ if 'data.gov.au Category' in doc.value['metadata'].keys() and len(
+ doc.value['metadata']['data.gov.au Category']) > 0:
+ if hasattr(doc.value['metadata']['data.gov.au Category'], '__iter__'):
+ groups = groups + doc.value['metadata']['data.gov.au Category']
else:
- group_entity['packages'] = [pkg_name]
- ckan.group_entity_put(group_entity)
- except CkanApiError, e:
- if ckan.last_status == 404:
- print "group does not exist, creating"
- group_entity = {
- 'name': group_name,
- 'title': doc.value['metadata']["Agency"],
- 'description': doc.value['metadata']["Agency"],
- 'packages': [pkg_name],
- # 'type': "organization" # not allowed via API, use database query
- # update "group" set type = 'organization';
+ groups = groups + [doc.value['metadata']['data.gov.au Category']]
+
+ for group_name in groups:
+ group_url = name_munge(group_name[:100])
+ try:
+ # Update the group details
+ group_entity = ckan.group_entity_get(group_url)
+ print "group "+group_name+" exists"
+ if 'packages' in group_entity.keys():
+ group_entity['packages'] = list(set(group_entity['packages'] + [pkg_name]))
+ else:
+ group_entity['packages'] = [pkg_name]
+ ckan.group_entity_put(group_entity)
+ except CkanApiError, e:
+ if ckan.last_status == 404:
+ print "group "+group_name+" does not exist, creating"
+ group_entity = {
+ 'name': group_url,
+ 'title': group_name,
+ 'description': group_name,
+ 'packages': [pkg_name]
}
- print group_entity
- ckan.group_register_post(group_entity)
- elif ckan.last_status == 409:
- print "group already exists"
- else:
- raise LoaderError('Unexpected status %s adding to group under \'%s\': %r' % (
- ckan.last_status, pkg_name, e.args))
-
- # #add to or create organization using direct API
- # org_name = name_munge(doc.value['metadata']["Agency"][:100])
- # if org_name not in orgs_list:
- # orgs_list = ckandirect.action.organization_list()['result']
- # print orgs_list
- # if org_name not in orgs_list:
- # try:
- # print "org not found, creating "+org_name
- # ckandirect.action.organization_create(name = org_name, title= doc.value['metadata']["Agency"],
- # description= doc.value['metadata']["Agency"])
- # orgs_list.append(org_name)
- # except ckanapi.ValidationError, e:
- # print e
- # raise LoaderError('Unexpected status')
- #
- # try:
- # print "linking package to "+org_name
- #
- # #brute force
- # org = ckandirect.action.organization_show(id=org_name)
- # #print org["result"]["packages"]
- # #print pkg
- # org["result"]["packages"].append({"id":pkg['id'], "name": pkg_name})
- # #print org["result"]
- # ckandirect.call_action(action="organization_update", data_dict=org["result"])
- #
- # # if only this worked
- # print ckandirect.action.package_owner_org_update(id=pkg['id'],organization_id=org["result"]["id"])
- #
- # except ckanapi.ValidationError, e:
- # print e
- # raise LoaderError('Unexpected status')
- # except ckanapi.CKANAPIError, e:
- # print "lol"
-
-
-
+ #print group_entity
+ ckan.group_register_post(group_entity)
+ elif ckan.last_status == 409:
+ print "group already exists"
+ else:
+ raise LoaderError('Unexpected status %s adding to group under \'%s\': %r' % (
+ ckan.last_status, pkg_name, e.args))
+
+ # add resources (downloadable data files)
if 'Download' in doc.value['metadata'].keys():
try:
@@ -254,7 +249,8 @@
name = resource['name']
print resource
ckan.add_package_resource(pkg_name, resource['href'], name=name, resource_type='data',
- format=format, size=human2bytes(resource['size'].replace(',', '')))
+ format=format,
+ size=human2bytes(resource['size'].replace(',', '')))
else:
print "resources already exist"
except CkanApiError, e: