/
reports.py
246 lines (197 loc) · 8.17 KB
/
reports.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
from collections import namedtuple
from sqlalchemy import or_, and_, func
import ckan.model as model
import ckan.plugins as p
import ckan.lib.dictization.model_dictize as model_dictize
resource_dictize = model_dictize.resource_dictize
def five_stars(id=None):
"""
Return a list of dicts: 1 for each dataset that has an openness score.
Each dict is of the form:
{'name': <string>, 'title': <string>, 'openness_score': <int>}
"""
if id:
pkg = model.Package.get(id)
if not pkg:
return "Not found"
# take the maximum openness score among dataset resources to be the
# overall dataset openness core
query = model.Session.query(model.Package.name, model.Package.title,
model.Resource.id,
model.TaskStatus.value.label('value'))\
.join(model.ResourceGroup, model.Package.id == model.ResourceGroup.package_id)\
.join(model.Resource)\
.join(model.TaskStatus, model.TaskStatus.entity_id == model.Resource.id)\
.filter(model.TaskStatus.key==u'openness_score')\
.group_by(model.Package.name, model.Package.title, model.Resource.id, model.TaskStatus.value)\
.distinct()
if id:
query = query.filter(model.Package.id == pkg.id)
results = []
for row in query:
results.append({
'name': row.name,
'title': row.title + u' ' + row.id,
'openness_score': row.value
})
return results
def resource_five_stars(id):
"""
Return a dict containing the QA results for a given resource
Each dict is of the form:
{'openness_score': <int>, 'openness_score_reason': <string>, 'failure_count': <int>}
"""
if id:
r = model.Resource.get(id)
if not r:
return {} # Not found
context = {'model': model, 'session': model.Session}
data = {'entity_id': r.id, 'task_type': 'qa'}
try:
data['key'] = 'openness_score'
status = p.toolkit.get_action('task_status_show')(context, data)
openness_score = int(status.get('value'))
openness_score_updated = status.get('last_updated')
data['key'] = 'openness_score_reason'
status = p.toolkit.get_action('task_status_show')(context, data)
openness_score_reason = status.get('value')
openness_score_reason_updated = status.get('last_updated')
data['key'] = 'openness_score_failure_count'
status = p.toolkit.get_action('task_status_show')(context, data)
openness_score_failure_count = int(status.get('value'))
openness_score_failure_count_updated = status.get('last_updated')
last_updated = max(
openness_score_updated,
openness_score_reason_updated,
openness_score_failure_count_updated )
result = {
'openness_score': openness_score,
'openness_score_reason': openness_score_reason,
'openness_score_failure_count': openness_score_failure_count,
'openness_score_updated': openness_score_updated,
'openness_score_reason_updated': openness_score_reason_updated,
'openness_score_failure_count_updated': openness_score_failure_count_updated,
'openness_updated': last_updated
}
except p.toolkit.ObjectNotFound:
result = {}
return result
def broken_resource_links_by_dataset():
"""
Return a list of named tuples, one for each dataset that contains
broken resource links (defined as resources with an openness score of 0).
The named tuple is of the form:
(name (str), title (str), resources (list of dicts))
"""
query = model.Session.query(model.Package.name, model.Package.title, model.Resource)\
.join(model.ResourceGroup, model.Package.id == model.ResourceGroup.package_id)\
.join(model.Resource)\
.join(model.TaskStatus, model.TaskStatus.entity_id == model.Resource.id)\
.filter(model.TaskStatus.key == u'openness_score')\
.filter(model.TaskStatus.value == u'0')\
.distinct()
context = {'model': model, 'session': model.Session}
results = {}
for name, title, resource in query:
resource = resource_dictize(resource, context)
data = {'entity_id': resource['id'], 'task_type': 'qa', 'key': 'openness_score_reason'}
status = p.toolkit.get_action('task_status_show')(context, data)
resource['openness_score_reason'] = status.get('value')
if name in results:
results[name].resources.append(resource)
else:
DatasetTuple = namedtuple('DatasetTuple', ['name', 'title', 'resources'])
results[name] = DatasetTuple(name, title or name, [resource])
return results.values()
def broken_resource_links_by_dataset_for_organisation(organisation_id):
result = _get_broken_resource_links(organisation_id)
if result:
return {
'id': result.keys()[0][1],
'title': result.keys()[0][0],
'packages': result.values()[0]
}
else:
return {
'id': None,
'title': None,
'packages': []
}
def organisations_with_broken_resource_links_by_name():
result = _get_broken_resource_links().keys()
result.sort()
return result
def organisations_with_broken_resource_links():
return _get_broken_resource_links()
def _get_broken_resource_links(organisation_id=None):
organisation_id = None
query = model.Session.query(model.Package.name, model.Package.title,
model.PackageExtra.value, model.Resource)\
.join(model.PackageExtra)\
.join(model.ResourceGroup, model.Package.id == model.ResourceGroup.package_id)\
.join(model.Resource)\
.join(model.TaskStatus, model.TaskStatus.entity_id == model.Resource.id)\
.filter(model.TaskStatus.key == u'openness_score')\
.filter(model.TaskStatus.value == u'0')\
.filter(or_(
and_(model.PackageExtra.key=='published_by',
model.PackageExtra.value.like('%%[%s]' % (organisation_id is None and '%' or organisation_id))),
and_(model.PackageExtra.key=='published_via',
model.PackageExtra.value.like('%%[%s]' % (organisation_id is None and '%' or organisation_id))),
)\
)\
.distinct()
context = {'model': model, 'session': model.Session}
data = []
for row in query:
resource = resource_dictize(row.Resource, context)
task_data = {'entity_id': resource['id'], 'task_type': 'qa', 'key': 'openness_score_reason'}
status = p.toolkit.get_action('task_status_show')(context, task_data)
resource['openness_score'] = u'0'
resource['openness_score_reason'] = status.get('value')
data.append([row.name, row.title, row.value, resource])
return _collapse(data, [_extract_publisher, _extract_dataset])
def _collapser(data, key_func=None):
result = {}
for row in data:
if key_func:
row = key_func(row)
key = row[0]
if len(row) == 2:
row = row[1]
else:
row = row[1:]
if key in result:
result[key].append(row)
else:
result[key] = [row]
return result
def _collapse(data, fn):
first = _collapser(data, fn[0])
result = {}
for k, v in first.items():
result[k] = _collapser(v, fn[1])
return result
def _extract_publisher(row):
"""
Extract publisher info from a query result row.
Each row should be a list of the form [name, title, value, Resource]
Returns a list of the form:
[<publisher tuple>, <other elements in row tuple>]
"""
publisher = row[2]
parts = publisher.split('[')
try:
pub_parts = (parts[0].strip(), parts[1][:-1])
except:
raise Exception('Could not get the ID from %r' % publisher)
else:
return [pub_parts] + [row[0], row[1], row[3]]
def _extract_dataset(row):
"""
Extract dataset info form a query result row.
Each row should be a list of the form [name, title, Resource]
Returns a list of the form:
[(name, title), Resource]
"""
return [(row[0], row[1]), row[2]]