projects / disclosr.git / commitdiff
search:
summary | shortlog | log | commit | commitdiff | tree
unified | plain
enforce valid package url
enforce valid package url


Former-commit-id: 26accf46eff2958223ffae9a1234d4291379bce4

file:a/documents/datagov-export.py -> file:b/documents/datagov-export.py
import ckanclient import ckanclient
import couchdb import couchdb
from ckanclient import CkanApiError from ckanclient import CkanApiError
import re import re
  import html2text # aaronsw :(
  import ckanapi
   
   
class LoaderError(Exception): class LoaderError(Exception):
pass pass
   
  # 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'
  ckan = ckanclient.CkanClient(base_location='http://data.disclosurelo.gs/api',
  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/')
   
  # http://code.activestate.com/recipes/578019-bytes-to-human-human-to-bytes-converter/
  SYMBOLS = {
  'customary': ('B', 'KB', 'MB', 'GB', 'T', 'P', 'E', 'Z', 'Y'),
  'customary_ext': ('byte', 'kilo', 'mega', 'giga', 'tera', 'peta', 'exa',
  'zetta', 'iotta'),
  'iec': ('Bi', 'Ki', 'Mi', 'Gi', 'Ti', 'Pi', 'Ei', 'Zi', 'Yi'),
  'iec_ext': ('byte', 'kibi', 'mebi', 'gibi', 'tebi', 'pebi', 'exbi',
  'zebi', 'yobi'),
  }
   
  def human2bytes(s):
  """
  Attempts to guess the string format based on default symbols
  set and return the corresponding bytes as an integer.
  When unable to recognize the format ValueError is raised.
   
  >>> human2bytes('0 B')
  0
  >>> human2bytes('1 K')
  1024
  >>> human2bytes('1 M')
  1048576
  >>> human2bytes('1 Gi')
  1073741824
  >>> human2bytes('1 tera')
  1099511627776
   
  >>> human2bytes('0.5kilo')
  512
  >>> human2bytes('0.1 byte')
  0
  >>> human2bytes('1 k') # k is an alias for K
  1024
  >>> human2bytes('12 foo')
  Traceback (most recent call last):
  ...
  ValueError: can't interpret '12 foo'
  """
  init = s
  num = ""
  while s and s[0:1].isdigit() or s[0:1] == '.':
  num += s[0]
  s = s[1:]
  num = float(num)
  letter = s.strip()
  for name, sset in SYMBOLS.items():
  if letter in sset:
  break
  else:
  if letter == 'k':
  # treat 'k' as an alias for 'K' as per: http://goo.gl/kTQMs
  sset = SYMBOLS['customary']
  letter = letter.upper()
  else:
  raise ValueError("can't interpret %r" % init)
  prefix = {sset[0]: 1}
  for i, s in enumerate(sset[1:]):
  prefix[s] = 1 << (i + 1) * 10
  return int(num * prefix[letter])
   
# https://github.com/okfn/ckanext-importlib # https://github.com/okfn/ckanext-importlib
# Instantiate the CKAN client.  
ckan = ckanclient.CkanClient(base_location='http://localhost:5000/api',  
api_key='b47b24cd-591d-40c1-8677-d73101d56d1b')  
# (use your own api_key from http://thedatahub.org/user/me )  
   
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
   
  #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): def name_munge(input_name):
return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and')) return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and'))
  #[:100]
#return input_name.replace(' ', '').replace('.', '_').replace('&', 'and') #return input_name.replace(' ', '').replace('.', '_').replace('&', 'and')
   
couch = couchdb.Server('http://127.0.0.1:5984/')  
  def get_licence_id(licencename):
  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',
  "CreativeCommonsAttributionNonCommercial30AustraliaCCBYNC30": 'cc-nc',
  'Otherpleasespecify': 'notspecified',
  '': 'notspecified',
  "Publicly available data": 'notspecified',
  "CreativeCommonsAttributionNoDerivativeWorks30AustraliaCCBYND30": "other-closed",
  "CreativeCommonsAttributionNonCommercialNoDerivs30AustraliaCCBYNCND30": "other-closed",
  'CreativeCommonsAttribution30AustraliaCCBY30': 'cc-by',
  "Creative Commons - Attribution 2.5 Australia (CC-BY)": 'cc-by',
  'CreativeCommonsAttributionCCBY25': 'cc-by',
  "PublicDomain": 'other-pd',
  }
  if licencename not in map.keys():
  raise Exception(licencename + " not found");
  return map[licencename];
   
docsdb = couch['disclosr-documents'] docsdb = couch['disclosr-documents']
   
if __name__ == "__main__": if __name__ == "__main__":
  orgs_list = []
for doc in docsdb.view('app/datasets'): for doc in docsdb.view('app/datasets'):
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 = name_munge(doc.value['metadata']['DCTERMS.Title'][:100]) pkg_name = filter( lambda x: x in '0123456789abcdefghijklmnopqrstuvwxyz-_',doc.value['url'].replace("http://data.gov.au/dataset/",'').replace('/','')[:100]);
package_entity = { tags = []
'name': pkg_name, if doc.value['agencyID'] == "AGIMO":
'title': doc.value['metadata']['DCTERMS.Title'], if len(doc.value['metadata']["Keywords / Tags"]) > 0:
'url': doc.value['metadata']['DCTERMS.Source.URI'], if hasattr(doc.value['metadata']["Keywords / Tags"], '__iter__'):
'tags': doc.value['metadata']["Keywords / Tags"], #todo must be alphanumeric characters or symbols tags = tags + doc.value['metadata']["Keywords / Tags"]
  else:
'author': doc.value['metadata']["DCTERMS.Creator"], tags = tags + [doc.value['metadata']["Keywords / Tags"]]
'maintainer': doc.value['metadata']["DCTERMS.Creator"], if 'data.gov.au Category' in doc.value['metadata'].keys() and len(doc.value['metadata']['data.gov.au Category']) > 0:
'licence_id': doc.value['metadata']['DCTERMS.License'], if hasattr(doc.value['metadata']['data.gov.au Category'], '__iter__'):
'notes': doc.value['metadata']['Description'], 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 = {
  'name': pkg_name,
  'title': doc.value['metadata']['DCTERMS.Title'],
  'url': doc.value['metadata']['DCTERMS.Source.URI'],
  'tags': tags, #tags are mandatory?
  'author': doc.value['metadata']["DCTERMS.Creator"],
  'maintainer': doc.value['metadata']["DCTERMS.Creator"],
  'licence_id': get_licence_id(doc.value['metadata']['DCTERMS.License']),
  'notes': html2text.html2text(doc.value['metadata']['Description']),
} }
  if doc.value['agencyID'] == "qld":
  package_entity = doc.value['metadata']
   
try: try:
  #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 "already exists" print "package already exists"
else: else:
raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % (ckan.last_status, doc.id, e.args)) print ckan.last_message
  raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % (
print package_entity ckan.last_status, pkg_name, e.args))
ckan.add_package_resource(pkg_name, 'http://example.org/', name='Foo', resource_type='data', format='csv') 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]))
  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';
  }
  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"
   
   
   
  if 'Download' in doc.value['metadata'].keys():
  try:
   
  resources = pkg.get('resources', [])
  if len(resources) < len(doc.value['metadata']['Download']):
  for resource in doc.value['metadata']['Download']:
   
  # http://docs.ckan.org/en/ckan-1.7/domain-model-resource.html
  # (KML/KMZ) / (Shapefile) /(Other)
  format = "plain"
  if resource['format'] == '(XML)':
  format = 'xml'
  if resource['format'] == '(CSV/XLS)':
  format = 'csv'
  name = resource['href']
  if 'name' in resource.keys():
  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(',', '')))
  else:
  print "resources already exist"
  except CkanApiError, e:
  if ckan.last_status == 404:
  print "parent dataset does not exist"
  else:
  raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % (
  ckan.last_status, pkg_name, e.args))
   
file:b/documents/datagov-merge.php (new)
  <?php
 
  include_once("../include/common.inc.php");
 
 
  setlocale(LC_CTYPE, 'C');
 
  $db = $server->get_db('disclosr-documents');
  $datasets = Array();
  try {
  $rows = $db->get_view("app", "datasets", null, true)->rows;
 
  foreach ($rows as $row) {
  //print_r($row);
  if ($row->value->url != "http://data.gov.au/data/")
  $datasets[str_replace(Array("http://data.gov.au/dataset/","/"),"",$row->value->url)] = $row->id;
  }
  } catch (SetteeRestClientException $e) {
  setteErrorHandler($e);
  }
  ksort($datasets);
  foreach ($datasets as $datasetname => $datasetkey) {
  print "$datasetname => $datasetkey<br>\n";
  }
  ?>
 
file:a/documents/datagov.py -> file:b/documents/datagov.py
import sys, os import sys, os
import time import time
import scrape import scrape
from bs4 import BeautifulSoup from bs4 import BeautifulSoup