be permissive with datagov resource file size
be permissive with datagov resource file size


Former-commit-id: c0a71c3c09369965e8748c2a8f7062eb2abbe01a

import ckanclient import ckanclient
import couchdb import couchdb
from ckanclient import CkanApiError from ckanclient import CkanApiError
import re import re
import html2text # aaronsw :( import html2text # aaronsw :(
  import ckanapi # https://github.com/open-data/ckanapi
   
   
class LoaderError(Exception): class LoaderError(Exception):
pass pass
   
# Instantiate the CKAN client. # Instantiate the CKAN client.
#ckan = ckanclient.CkanClient(base_location='http://localhost:5000/api', api_key='b47b24cd-591d-40c1-8677-d73101d56d1b') api_key = 'ff34526e-f794-4068-8235-fcbba38cd8bc'
ckan = ckanclient.CkanClient(base_location='http://data.disclosurelo.gs/api', ckan = ckanclient.CkanClient(base_location='http://data.disclosurelo.gs/api',
api_key='482a9dd2-a976-4adf-ac77-d71d92a98a52') api_key=api_key)
#couch = couchdb.Server('http://127.0.0.1:5984/') ckandirect = ckanapi.RemoteCKAN('http://data.disclosurelo.gs', api_key=api_key)
couch = couchdb.Server('http://192.168.1.113:5984/') couch = couchdb.Server('http://127.0.0.1:5984/')
  #couch = couchdb.Server('http://192.168.1.113:5984/')
   
# http://code.activestate.com/recipes/578019-bytes-to-human-human-to-bytes-converter/ # http://code.activestate.com/recipes/578019-bytes-to-human-human-to-bytes-converter/
SYMBOLS = { SYMBOLS = {
'customary': ('B', 'KB', 'MB', 'GB', 'T', 'P', 'E', 'Z', 'Y'), 'customary': ('B', 'KB', 'MB', 'GB', 'T', 'P', 'E', 'Z', 'Y'),
'customary_ext': ('byte', 'kilo', 'mega', 'giga', 'tera', 'peta', 'exa', 'customary_ext': ('byte', 'kilo', 'mega', 'giga', 'tera', 'peta', 'exa',
'zetta', 'iotta'), 'zetta', 'iotta'),
'iec': ('Bi', 'Ki', 'Mi', 'Gi', 'Ti', 'Pi', 'Ei', 'Zi', 'Yi'), 'iec': ('Bi', 'Ki', 'Mi', 'Gi', 'Ti', 'Pi', 'Ei', 'Zi', 'Yi'),
'iec_ext': ('byte', 'kibi', 'mebi', 'gibi', 'tebi', 'pebi', 'exbi', 'iec_ext': ('byte', 'kibi', 'mebi', 'gibi', 'tebi', 'pebi', 'exbi',
'zebi', 'yobi'), 'zebi', 'yobi'),
} }
   
   
def human2bytes(s): def human2bytes(s):
""" """
Attempts to guess the string format based on default symbols Attempts to guess the string format based on default symbols
set and return the corresponding bytes as an integer. set and return the corresponding bytes as an integer.
When unable to recognize the format ValueError is raised. When unable to recognize the format ValueError is raised.
   
>>> human2bytes('0 B') >>> human2bytes('0 B')
0 0
>>> human2bytes('1 K') >>> human2bytes('1 K')
1024 1024
>>> human2bytes('1 M') >>> human2bytes('1 M')
1048576 1048576
>>> human2bytes('1 Gi') >>> human2bytes('1 Gi')
1073741824 1073741824
>>> human2bytes('1 tera') >>> human2bytes('1 tera')
1099511627776 1099511627776
   
>>> human2bytes('0.5kilo') >>> human2bytes('0.5kilo')
512 512
>>> human2bytes('0.1 byte') >>> human2bytes('0.1 byte')
0 0
>>> human2bytes('1 k') # k is an alias for K >>> human2bytes('1 k') # k is an alias for K
1024 1024
>>> human2bytes('12 foo') >>> human2bytes('12 foo')
Traceback (most recent call last): Traceback (most recent call last):
... ...
ValueError: can't interpret '12 foo' ValueError: can't interpret '12 foo'
""" """
  if s == None:
  return 0
  s = s.replace(',', '')
init = s init = s
num = "" num = ""
while s and s[0:1].isdigit() or s[0:1] == '.': while s and s[0:1].isdigit() or s[0:1] == '.':
num += s[0] num += s[0]
s = s[1:] s = s[1:]
num = float(num) num = float(num)
letter = s.strip() letter = s.strip()
for name, sset in SYMBOLS.items(): for name, sset in SYMBOLS.items():
if letter in sset: if letter in sset:
break break
else: else:
if letter == 'k': if letter == 'k':
# treat 'k' as an alias for 'K' as per: http://goo.gl/kTQMs # treat 'k' as an alias for 'K' as per: http://goo.gl/kTQMs
sset = SYMBOLS['customary'] sset = SYMBOLS['customary']
letter = letter.upper() letter = letter.upper()
else: else:
raise ValueError("can't interpret %r" % init) raise ValueError("can't interpret %r" % init)
prefix = {sset[0]: 1} prefix = {sset[0]: 1}
for i, s in enumerate(sset[1:]): for i, s in enumerate(sset[1:]):
prefix[s] = 1 << (i + 1) * 10 prefix[s] = 1 << (i + 1) * 10
return int(num * prefix[letter]) return int(num * prefix[letter])
   
# https://github.com/okfn/ckanext-importlib # https://github.com/okfn/ckanext-importlib
def munge(name): def munge(name):
# convert spaces to underscores # convert spaces to underscores
name = re.sub(' ', '_', name).lower() name = re.sub(' ', '_', name).lower()
# convert symbols to dashes # convert symbols to dashes
name = re.sub('[:]', '_-', name).lower() name = re.sub('[:]', '_-', name).lower()
name = re.sub('[/]', '-', name).lower() name = re.sub('[/]', '-', name).lower()
# take out not-allowed characters # take out not-allowed characters
name = re.sub('[^a-zA-Z0-9-_]', '', name).lower() name = re.sub('[^a-zA-Z0-9-_]', '', name).lower()
# remove double underscores # remove double underscores
name = re.sub('__', '_', name).lower() name = re.sub('__', '_', name).lower()
return name return name
   
   
def name_munge(input_name): def name_munge(input_name):
return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and')) return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and'))
#return input_name.replace(' ', '').replace('.', '_').replace('&', 'and')  
   
   
def get_licence_id(licencename): def get_licence_id(licencename):
map = { map = {
"Creative Commons - Attribution-Share Alike 2.0 Australia (CC-SA)\nThe downloadable version of the database is licensed under CC-BY-SA Creative Commons Attribution Share Alike and contains only the database fields that are released under that license. These fields are object title, object number, object description as well as temporal, spatial and dimension details. It also contains a persistent URL for each record.": 'cc-by-sa', "Creative Commons - Attribution-Share Alike 2.0 Australia (CC-SA)\nThe downloadable version of the database is licensed under CC-BY-SA Creative Commons Attribution Share Alike and contains only the database fields that are released under that license. These fields are object title, object number, object description as well as temporal, spatial and dimension details. It also contains a persistent URL for each record.": 'cc-by-sa',
"CreativeCommonsAttributionNonCommercial30AustraliaCCBYNC30": 'cc-nc', "CreativeCommonsAttributionNonCommercial30AustraliaCCBYNC30": 'cc-nc',
'Otherpleasespecify': 'notspecified', 'Otherpleasespecify': 'notspecified',
'': 'notspecified', '': 'notspecified',
"Publicly available data": 'notspecified', "Publicly available data": 'notspecified',
"CreativeCommonsAttributionNoDerivativeWorks30AustraliaCCBYND30": "other-closed", "CreativeCommonsAttributionNoDerivativeWorks30AustraliaCCBYND30": "other-closed",
"CreativeCommonsAttributionNonCommercialNoDerivs30AustraliaCCBYNCND30": "other-closed", "CreativeCommonsAttributionNonCommercialNoDerivs30AustraliaCCBYNCND30": "other-closed",
'CreativeCommonsAttribution30AustraliaCCBY30': 'cc-by', 'CreativeCommonsAttribution30AustraliaCCBY30': 'cc-by',
"Creative Commons - Attribution 2.5 Australia (CC-BY)": 'cc-by', "Creative Commons - Attribution 2.5 Australia (CC-BY)": 'cc-by',
'CreativeCommonsAttributionCCBY25': 'cc-by', 'CreativeCommonsAttributionCCBY25': 'cc-by',
"PublicDomain": 'other-pd', "PublicDomain": 'other-pd',
} }
if licencename not in map.keys(): if licencename not in map.keys():
raise Exception(licencename + " not found"); raise Exception(licencename + " not found");
return map[licencename]; return map[licencename];
   
   
docsdb = couch['disclosr-documents'] docsdb = couch['disclosr-documents']
   
if __name__ == "__main__": if __name__ == "__main__":
  orgs_list = []
  orgs_ids = {}
for doc in docsdb.view('app/datasets'): for doc in docsdb.view('app/datasets'):
  print " --- "
print doc.id print doc.id
if doc.value['url'] != "http://data.gov.au/data/":  
  if doc.value['url'] != "http://data.gov.au/data/" and doc.value['agencyID'] != "qld":
   
   
# Collect the package metadata. # Collect the package metadata.
pkg_name = doc.value['url'].replace("http://data.gov.au/dataset/",'').replace('/',''); _ 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 = [] tags = []
if len(doc.value['metadata']["Keywords / Tags"]) > 0: if doc.value['agencyID'] == "AGIMO":
if hasattr(doc.value['metadata']["Keywords / Tags"], '__iter__'): if len(doc.value['metadata']["Keywords / Tags"]) > 0:
tags = tags + doc.value['metadata']["Keywords / Tags"] if hasattr(doc.value['metadata']["Keywords / Tags"], '__iter__'):
else: tags = tags + doc.value['metadata']["Keywords / Tags"]
tags = tags + [doc.value['metadata']["Keywords / Tags"]] else:
if 'data.gov.au Category' in doc.value['metadata'].keys() and len(doc.value['metadata']['data.gov.au Category']) > 0: tags = tags + [doc.value['metadata']["Keywords / Tags"]]
if hasattr(doc.value['metadata']['data.gov.au Category'], '__iter__'):  
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]
else: #print tags
tags = tags + [doc.value['metadata']['data.gov.au Category']] package_entity = {
tags = [re.sub('[^a-zA-Z0-9-_.]', '', tag.replace('&', 'and')).lower() for tag in tags if tag] 'name': pkg_name,
print tags 'title': doc.value['metadata']['DCTERMS.Title'],
package_entity = { 'url': doc.value['metadata']['DCTERMS.Source.URI'],
'name': pkg_name, 'tags': tags, #tags are mandatory?
'title': doc.value['metadata']['DCTERMS.Title'], 'author': doc.value['metadata']["DCTERMS.Creator"],
'url': doc.value['metadata']['DCTERMS.Source.URI'], 'maintainer': doc.value['metadata']["DCTERMS.Creator"],
'tags': tags, #tags are mandatory? 'licence_id': get_licence_id(doc.value['metadata']['DCTERMS.License']),
'author': doc.value['metadata']["DCTERMS.Creator"], 'notes': html2text.html2text(doc.value['metadata']['Description']),
'maintainer': doc.value['metadata']["DCTERMS.Creator"], 'owner_org': org_id
'licence_id': get_licence_id(doc.value['metadata']['DCTERMS.License']), #todo add missing key values like jurasdiction
'notes': html2text.html2text(doc.value['metadata']['Description']), }
} if doc.value['agencyID'] == "qld":
  package_entity = doc.value['metadata']
   
try: try:
print package_entity #print package_entity
ckan.package_register_post(package_entity) ckan.package_register_post(package_entity)
except CkanApiError, e: except CkanApiError, e:
if ckan.last_status == 409: if ckan.last_message == "{\"name\": [\"That URL is already in use.\"]}":
print "package already exists" print "package already exists"
else: else:
  print ckan.last_message
raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % ( raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % (
ckan.last_status, pkg_name, e.args)) ckan.last_status, pkg_name, e.args))
  pkg = ckan.package_entity_get(pkg_name)
   
#add to group # add dataset to group(s)
  groups = []
group_name = name_munge(doc.value['metadata']["Agency"][:100]) if 'data.gov.au Category' in doc.value['metadata'].keys() and len(
try: doc.value['metadata']['data.gov.au Category']) > 0:
print ckan.group_entity_get(group_name) if hasattr(doc.value['metadata']['data.gov.au Category'], '__iter__'):
  groups = groups + doc.value['metadata']['data.gov.au Category']
# 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]))  
else: else:
group_entity['packages'] = [pkg_name] groups = groups + [doc.value['metadata']['data.gov.au Category']]
ckan.group_entity_put(group_entity)  
except CkanApiError, e: for group_name in groups:
if ckan.last_status == 404: group_url = name_munge(group_name[:100])
print "group does not exist, creating" try:
group_entity = { # Update the group details
'name': group_name, group_entity = ckan.group_entity_get(group_url)
'title': doc.value['metadata']["Agency"], print "group "+group_name+" exists"
'description': doc.value['metadata']["Agency"], if 'packages' in group_entity.keys():
'packages': [pkg_name], group_entity['packages'] = list(set(group_entity['packages'] + [pkg_name]))
# 'type': "organization" # not allowed via API, use database query else:
# update "group" set type = 'organization'; 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 #print group_entity
ckan.group_register_post(group_entity) ckan.group_register_post(group_entity)
else: elif ckan.last_status == 409:
raise LoaderError('Unexpected status %s adding to group under \'%s\': %r' % ( print "group already exists"
ckan.last_status, pkg_name, e.args)) 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(): if 'Download' in doc.value['metadata'].keys():
try: try:
pkg = ckan.package_entity_get(pkg_name)  
resources = pkg.get('resources', []) resources = pkg.get('resources', [])
if len(resources) < len(doc.value['metadata']['Download']): if len(resources) < len(doc.value['metadata']['Download']):
for resource in doc.value['metadata']['Download']: for resource in doc.value['metadata']['Download']:
   
# http://docs.ckan.org/en/ckan-1.7/domain-model-resource.html # http://docs.ckan.org/en/ckan-1.7/domain-model-resource.html
# (KML/KMZ) / (Shapefile) /(Other) # (KML/KMZ) / (Shapefile) /(Other)
format = "plain" format = "plain"
if resource['format'] == '(XML)': if resource['format'] == '(XML)':
format = 'xml' format = 'xml'
if resource['format'] == '(CSV/XLS)': if resource['format'] == '(CSV/XLS)':
format = 'csv' format = 'csv'
  if resource['format'] == '(Shapefile)':
  format = 'shp'
  if resource['format'] == '(KML/KMZ)':
  format = 'kml'
name = resource['href'] name = resource['href']
if 'name' in resource.keys(): if 'name' in resource.keys():
name = resource['name'] name = resource['name']
print resource print resource
ckan.add_package_resource(pkg_name, resource['href'], name=name, resource_type='data', 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.get('size','0B')))
else: else:
print "resources already exist" print "resources already exist"
except CkanApiError, e: except CkanApiError, e:
if ckan.last_status == 404: if ckan.last_status == 404:
print "parent dataset does not exist" print "parent dataset does not exist"
else: else:
raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % ( raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % (
ckan.last_status, pkg_name, e.args)) ckan.last_status, pkg_name, e.args))