fix broken urls on datagov export
[disclosr.git] / documents / datagov-export.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
import ckanclient
import couchdb
from ckanclient import CkanApiError
import re
import html2text # aaronsw :(
import ckanapi # https://github.com/open-data/ckanapi
 
 
class LoaderError(Exception):
    pass
 
# Instantiate the CKAN client.
api_key = 'ff34526e-f794-4068-8235-fcbba38cd8bc'
server = 'data.disclosurelo.gs'
 
ckan = ckanclient.CkanClient(base_location='http://'+server+'api',
                             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://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/
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'
    """
    if s == None:
        return 0
    s = s.replace(',', '')
    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
def munge(name):
    # convert spaces to underscores
    name = re.sub(' ', '_', name).lower()
    # convert symbols to dashes
    name = re.sub('[:]', '_-', name).lower()
    name = re.sub('[/]', '-', name).lower()
    # take out not-allowed characters
    name = re.sub('[^a-zA-Z0-9-_]', '', name).lower()
    # remove double underscores
    name = re.sub('__', '_', name).lower()
    return name
 
 
def name_munge(input_name):
    return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and'))
 
 
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']
 
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]);
            print pkg_name
            #add to or create organization using direct API
            agency = doc.value['metadata']["Agency"]
            if agency == "APS":
                agency = "Australian Public Service Commission"
            if agency == "Shared Services, Treasury Directorate":
                agency = "Shared Services Procurement, Treasury Directorate"
            if agency == "Treasury - Shared Services":
                agency = "Shared Services Procurement, Treasury Directorate"
            if agency == "Territory and Municipal Services (TAMS)":
                agency = "Territory and Municipal Services Directorate"
            if agency == "State Library of NSW":
                agency = "State Library of New South Wales"
            org_name = name_munge(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=agency,
                                                              description=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 = []
            creator = doc.value['metadata']["DCTERMS.Creator"]
            if doc.value['agencyID'] == "AGIMO":
                if len(doc.value['metadata']["Keywords / Tags"]) > 0:
                    if hasattr(doc.value['metadata']["Keywords / Tags"], '__iter__'):
                        tags = tags + doc.value['metadata']["Keywords / Tags"]
                    else:
                        tags = tags + [doc.value['metadata']["Keywords / Tags"]]
 
                tags = [re.sub('[^a-zA-Z0-9-_.]', '', tag.replace('&', 'and')).lower() for tag in tags if tag]
                #print tags
                extras = []
 
                for extra_key in doc.value['metadata'].keys():
                    if extra_key not in ["Description","Content-Language","DCTERMS.Description", "Keywords / Tags" ,"data.gov.au Category", "Download", "Permalink","DCTERMS.Identifier"]:
                        if doc.value['metadata'][extra_key] != None and doc.value['metadata'][extra_key] != "":
                                extras.append([extra_key, doc.value['metadata'][extra_key]])
 
                package_entity = {
                    'name': pkg_name,
                    'title': doc.value['metadata']['DCTERMS.Title'],
                    'url': doc.value['metadata']['DCTERMS.Source.URI'],
                    'tags': tags, #tags are mandatory?
                    'author': creator,
                    'maintainer': creator,
                    'licence_id': get_licence_id(doc.value['metadata']['DCTERMS.License']),
                    'notes': html2text.html2text(doc.value['metadata']['Description']),
                    'owner_org': org_id,
                    'extras': extras
                }
 
 
            try:
                #print package_entity
                ckan.package_register_post(package_entity)
            except CkanApiError, e:
                if ckan.last_message == "{\"name\": [\"That URL is already in use.\"]}":
                    print "package already exists"
                else:
                    print ckan.last_message
                    raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % (
                        ckan.last_status, pkg_name, e.args))
            pkg = ckan.package_entity_get(pkg_name)
 
 
            # add resources (downloadable data files)
            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'
                            if resource['format'] == '(Shapefile)':
                                format = 'shp'
                            if resource['format'] == '(KML/KMZ)':
                                format = 'kml'
                            name = resource['href']
                            if 'name' in resource.keys():
                                name = resource['name']
                            print resource
                            ckan.add_package_resource(pkg_name, url_fix(resource['href']), name=name, resource_type='data',
                                                      format=format,
                                                      size=human2bytes(resource.get('size','0B')))
                    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))