fix broken urls on datagov export
fix broken urls on datagov export


Former-commit-id: 3c41d2694ab3235f5b6e99681b28dc21dda4e617

import ckanclient import ckanclient
import couchdb import couchdb
from ckanclient import CkanApiError from ckanclient import CkanApiError
import re import re
   
   
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') #ckan = ckanclient.CkanClient(base_location='http://localhost:5000/api', api_key='b47b24cd-591d-40c1-8677-d73101d56d1b')
api_key = 'b3ab75e4-afbb-465b-a09d-8171c8c69a7a' 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=api_key) api_key=api_key)
couch = couchdb.Server('http://127.0.0.1:5984/') couch = couchdb.Server('http://127.0.0.1:5984/')
#couch = couchdb.Server('http://192.168.1.113:5984/') #couch = couchdb.Server('http://192.168.1.113:5984/')
   
# 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'))
   
   
docsdb = couch['disclosr-documents'] docsdb = couch['disclosr-documents']
   
if __name__ == "__main__": if __name__ == "__main__":
groups = {} groups = {}
for doc in docsdb.view('app/datasetGroups'): for doc in docsdb.view('app/datasetGroups'):
group_name = doc.key group_name = doc.key
if group_name != "Not specified": if group_name != "Not specified":
pkg_name = filter(lambda x: x in '0123456789abcdefghijklmnopqrstuvwxyz-_', pkg_name = filter(lambda x: x in '0123456789abcdefghijklmnopqrstuvwxyz-_',
doc.value.replace("http://data.gov.au/dataset/", '').replace('/', '')[:100]); doc.value.replace("http://data.gov.au/dataset/", '').replace('/', '')[:100]);
if group_name in groups.keys(): if group_name in groups.keys():
groups[group_name] = list(set(groups[group_name] + [pkg_name])) groups[group_name] = list(set(groups[group_name] + [pkg_name]))
else: else:
groups[group_name] = [pkg_name] groups[group_name] = [pkg_name]
   
# add dataset to group(s) # add dataset to group(s)
for group_name in groups.keys(): for group_name in groups.keys():
if group_name != "Not specified": if group_name != "Not specified":
group_url = name_munge(group_name[:100]) group_url = name_munge(group_name[:100])
print group_name print group_name
print groups[group_name] print groups[group_name]
try: try:
# Update the group details # Update the group details
group_entity = ckan.group_entity_get(group_url) group_entity = ckan.group_entity_get(group_url)
print "group "+group_name+" exists" print "group "+group_name+" exists"
if 'packages' in group_entity.keys(): if 'packages' in group_entity.keys():
group_entity['packages'] = list(set(group_entity['packages'] + groups[group_name])) group_entity['packages'] = list(set(group_entity['packages'] + groups[group_name]))
else: else:
group_entity['packages'] = groups[group_name] group_entity['packages'] = groups[group_name]
ckan.group_entity_put(group_entity) ckan.group_entity_put(group_entity)
except CkanApiError, e: except CkanApiError, e:
if ckan.last_status == 404: if ckan.last_status == 404:
print "group "+group_name+" does not exist, creating" print "group "+group_name+" does not exist, creating"
group_entity = { group_entity = {
'name': group_url, 'name': group_url,
'title': group_name, 'title': group_name,
'description': group_name, 'description': group_name,
'packages': groups[group_name] 'packages': groups[group_name]
} }
#print group_entity #print group_entity
ckan.group_register_post(group_entity) ckan.group_register_post(group_entity)
elif ckan.last_status == 409: elif ckan.last_status == 409:
print "group already exists" print "group already exists"
else: else:
raise LoaderError('Unexpected status %s adding to group under \'%s\': %r' % ( raise LoaderError('Unexpected status %s adding to group under \'%s\': %r' % (
ckan.last_status, pkg_name, e.args)) ckan.last_status, pkg_name, e.args))
   
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 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'
api_key = 'b3ab75e4-afbb-465b-a09d-8171c8c69a7a' server = 'data.disclosurelo.gs'
ckan = ckanclient.CkanClient(base_location='http://data.disclosurelo.gs/api',  
  ckan = ckanclient.CkanClient(base_location='http://'+server+'api',
api_key=api_key) api_key=api_key)
ckandirect = ckanapi.RemoteCKAN('http://data.disclosurelo.gs', api_key=api_key) ckandirect = ckanapi.RemoteCKAN('http://'+server, api_key=api_key)
couch = couchdb.Server('http://127.0.0.1:5984/') couch = couchdb.Server('http://127.0.0.1:5984/')
#couch = couchdb.Server('http://192.168.1.113:5984/') #couch = couchdb.Server('http://192.168.1.113:5984/')
   
  import urllib
  import urlparse
   
  def url_fix(s, charset='utf-8'):
  """Sometimes you get an URL by a user that just isn't a real
  URL because it contains unsafe characters like ' ' and so on. This
  function can fix some of the problems in a similar way browsers
  handle data entered by the user:
   
  >>> url_fix(u'http://de.wikipedia.org/wiki/Elf (Begriffsklärung)')
  'http://de.wikipedia.org/wiki/Elf%20%28Begriffskl%C3%A4rung%29'
   
  :param charset: The target charset for the URL if the url was
  given as unicode string.
  """
  if isinstance(s, unicode):
  s = s.encode(charset, 'ignore')
  if not urlparse.urlparse(s).scheme:
  s = "http://"+s
  scheme, netloc, path, qs, anchor = urlparse.urlsplit(s)
  path = urllib.quote(path, '/%')
  qs = urllib.quote_plus(qs, ':&=')
  return urlparse.urlunsplit((scheme, netloc, path, qs, anchor))
   
# 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'))
   
   
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_list = []
orgs_ids = {} orgs_ids = {}
for doc in docsdb.view('app/datasets'): for doc in docsdb.view('app/datasets'):
print " --- " print " --- "
print doc.id print doc.id
   
if doc.value['url'] != "http://data.gov.au/data/" and doc.value['agencyID'] != "qld": if doc.value['url'] != "http://data.gov.au/data/" and doc.value['agencyID'] != "qld":
   
   
# Collect the package metadata. # Collect the package metadata.
pkg_name = filter(lambda x: x in '0123456789abcdefghijklmnopqrstuvwxyz-_', pkg_name = filter(lambda x: x in '0123456789abcdefghijklmnopqrstuvwxyz-_',
doc.value['url'].replace("http://data.gov.au/dataset/", '').replace('/', '')[:100]); doc.value['url'].replace("http://data.gov.au/dataset/", '').replace('/', '')[:100]);
print pkg_name print pkg_name
#add to or create organization using direct API #add to or create organization using direct API
agency = doc.value['metadata']["Agency"] agency = doc.value['metadata']["Agency"]
if agency == "APS": if agency == "APS":
agency = "Australian Public Service Commission" agency = "Australian Public Service Commission"
if agency == "Shared Services, Treasury Directorate": if agency == "Shared Services, Treasury Directorate":
agency = "Shared Services Procurement, Treasury Directorate" agency = "Shared Services Procurement, Treasury Directorate"
if agency == "Treasury - Shared Services": if agency == "Treasury - Shared Services":
agency = "Shared Services Procurement, Treasury Directorate" agency = "Shared Services Procurement, Treasury Directorate"
if agency == "Territory and Municipal Services (TAMS)": if agency == "Territory and Municipal Services (TAMS)":
agency = "Territory and Municipal Services Directorate" agency = "Territory and Municipal Services Directorate"
if agency == "State Library of NSW": if agency == "State Library of NSW":
agency = "State Library of New South Wales" agency = "State Library of New South Wales"
org_name = name_munge(agency[:100]) org_name = name_munge(agency[:100])
if org_name not in orgs_list: if org_name not in orgs_list:
orgs_list = ckandirect.action.organization_list()['result'] orgs_list = ckandirect.action.organization_list()['result']
#print orgs_list #print orgs_list
if org_name not in orgs_list: if org_name not in orgs_list:
try: try:
print "org not found, creating " + org_name print "org not found, creating " + org_name
ckandirect.action.organization_create(name=org_name, title=agency, ckandirect.action.organization_create(name=org_name, title=agency,
description=agency) description=agency)
orgs_list.append(org_name) orgs_list.append(org_name)
except ckanapi.ValidationError, e: except ckanapi.ValidationError, e:
print e print e
raise LoaderError('Unexpected status') raise LoaderError('Unexpected status')
else: else:
print "org found, adding dataset to " + org_name print "org found, adding dataset to " + org_name
   
# cache org names -> id mapping # cache org names -> id mapping
if org_name not in orgs_ids: if org_name not in orgs_ids:
org = ckandirect.action.organization_show(id=org_name) org = ckandirect.action.organization_show(id=org_name)
orgs_ids[org_name] = org["result"]["id"] orgs_ids[org_name] = org["result"]["id"]
org_id = orgs_ids[org_name] org_id = orgs_ids[org_name]
print "org id is "+org_id print "org id is "+org_id
tags = [] tags = []
creator = doc.value['metadata']["DCTERMS.Creator"] creator = doc.value['metadata']["DCTERMS.Creator"]
if doc.value['agencyID'] == "AGIMO": if doc.value['agencyID'] == "AGIMO":
if len(doc.value['metadata']["Keywords / Tags"]) > 0: if len(doc.value['metadata']["Keywords / Tags"]) > 0:
if hasattr(doc.value['metadata']["Keywords / Tags"], '__iter__'): if hasattr(doc.value['metadata']["Keywords / Tags"], '__iter__'):
tags = tags + doc.value['metadata']["Keywords / Tags"] tags = tags + doc.value['metadata']["Keywords / Tags"]
else: else:
tags = tags + [doc.value['metadata']["Keywords / Tags"]] tags = tags + [doc.value['metadata']["Keywords / Tags"]]
   
tags = [re.sub('[^a-zA-Z0-9-_.]', '', tag.replace('&', 'and')).lower() for tag in tags if tag] tags = [re.sub('[^a-zA-Z0-9-_.]', '', tag.replace('&', 'and')).lower() for tag in tags if tag]
#print tags #print tags
extras = [] extras = []
   
for extra_key in doc.value['metadata'].keys(): for extra_key in doc.value['metadata'].keys():
if extra_key != "Keywords / Tags" and extra_key != "data.gov.au Category" and extra_key != "Download" : if extra_key not in ["Description","Content-Language","DCTERMS.Description", "Keywords / Tags" ,"data.gov.au Category", "Download", "Permalink","DCTERMS.Identifier"]:
extras.append({'key':extra_key, 'value':doc.value['metadata'][extra_key]}) if doc.value['metadata'][extra_key] != None and doc.value['metadata'][extra_key] != "":
  extras.append([extra_key, doc.value['metadata'][extra_key]])
   
package_entity = { package_entity = {
'name': pkg_name, 'name': pkg_name,
'title': doc.value['metadata']['DCTERMS.Title'], 'title': doc.value['metadata']['DCTERMS.Title'],
'url': doc.value['metadata']['DCTERMS.Source.URI'], 'url': doc.value['metadata']['DCTERMS.Source.URI'],
'tags': tags, #tags are mandatory? 'tags': tags, #tags are mandatory?
'author': creator, 'author': creator,
'maintainer': creator, 'maintainer': creator,
'licence_id': get_licence_id(doc.value['metadata']['DCTERMS.License']), 'licence_id': get_licence_id(doc.value['metadata']['DCTERMS.License']),
'notes': html2text.html2text(doc.value['metadata']['Description']),