def find(self, resource, req, lookup):
"""
Called to execute a search against the AAP Mulitmedia API. It attempts to translate the search request
passed in req to a suitable form for a search request against the API. It parses the response into a
suitable ElasticCursor, the front end will never know.
:param resource:
:param req:
:param lookup:
:return:
"""
if self._headers is None:
self.__set_auth_cookie(self._app)
url = self._app.config['AAP_MM_SEARCH_URL'] + '/Assets/search'
query_keywords = '*:*'
if 'query' in req['query']['filtered']:
query_keywords = req['query']['filtered']['query']['query_string']['query']
query_keywords = query_keywords.replace('slugline:', 'objectname:')
query_keywords = query_keywords.replace('description_text:', 'captionabstract:')
fields = {}
for criterion in req.get('post_filter', {}).get('and', {}):
# parse out the date range if possible
if 'range' in criterion:
start = None
end = None
daterange = None
if 'firstcreated' in criterion.get('range', {}):
if 'gte' in criterion['range']['firstcreated']:
start = criterion['range']['firstcreated']['gte'][0:10]
if 'lte' in criterion['range']['firstcreated']:
end = criterion['range']['firstcreated']['lte'][0:10]
# if there is a special start and no end it's one of the date buttons
if start and not end:
if start == 'now-24H':
daterange = {'Dates': ['[NOW/HOUR-24HOURS TO NOW/HOUR]']}
if start == 'now-1w':
daterange = {'Dates': ['[NOW/DAY-7DAYS TO NOW/DAY]']}
if start == 'now-1M':
daterange = {'Dates': ['[NOW/DAY-1MONTH TO NOW/DAY]']}
# we've got something but no daterange set above
if (start or end) and not daterange:
daterange = {'DateRange': [{'Start': start, 'End': end}], 'DateCreatedFilter': 'true'}
if daterange:
fields.update(daterange)
if 'terms' in criterion:
if 'type' in criterion.get('terms', {}):
fields.update({'MediaTypes': criterion['terms']['type']})
if 'credit' in criterion.get('terms', {}):
fields.update({'Credits': criterion['terms']['credit']})
if 'anpa_category.name' in criterion.get('terms', {}):
cat_list = []
for cat in criterion['terms']['anpa_category.name']:
qcode = [key for key, value in subject_codes.items() if value == cat]
if qcode:
for code in qcode:
cat_list.append(code)
else:
cat_list.append(cat)
fields.update({'Categories': cat_list})
size = int(req.get('size', '25')) if int(req.get('size', '25')) > 0 else 25
query = {'Query': query_keywords, 'pageSize': str(size),
'pageNumber': str(int(req.get('from', '0')) // size + 1)}
r = self._http.urlopen('POST', url + '?' + urllib.parse.urlencode(query),
body=json.dumps(fields), headers=self._headers)
hits = self._parse_hits(json.loads(r.data.decode('UTF-8')))
return ElasticCursor(docs=hits['docs'], hits={'hits': hits, 'aggregations': self._parse_aggregations(hits)})
def find(self, resource, req, lookup):
"""
Called to execute a search against the AAP Mulitmedia API. It attempts to translate the search request
passed in req to a suitable form for a search request against the API. It parses the response into a
suitable ElasticCursor, the front end will never know.
:param resource:
:param req:
:param lookup:
:return:
"""
if self._headers is None:
self.__set_auth_cookie(self._app)
url = self._app.config['AAP_MM_SEARCH_URL'] + '/Assets/search'
query_keywords = '*:*'
if 'query' in req['query']['filtered']:
query_keywords = req['query']['filtered']['query']['query_string'][
'query']
query_keywords = query_keywords.replace('slugline:', 'objectname:')
query_keywords = query_keywords.replace('description:',
'captionabstract:')
fields = {}
for criterion in req.get('post_filter', {}).get('and', {}):
# parse out the date range if possible
if 'range' in criterion:
start = None
end = None
daterange = None
if 'firstcreated' in criterion.get('range', {}):
if 'gte' in criterion['range']['firstcreated']:
start = criterion['range']['firstcreated']['gte'][0:10]
if 'lte' in criterion['range']['firstcreated']:
end = criterion['range']['firstcreated']['lte'][0:10]
# if there is a special start and no end it's one of the date buttons
if start and not end:
if start == 'now-24H':
daterange = {
'Dates': ['[NOW/HOUR-24HOURS TO NOW/HOUR]']
}
if start == 'now-1w':
daterange = {'Dates': ['[NOW/DAY-7DAYS TO NOW/DAY]']}
if start == 'now-1M':
daterange = {'Dates': ['[NOW/DAY-1MONTH TO NOW/DAY]']}
# we've got something but no daterange set above
if (start or end) and not daterange:
daterange = {
'DateRange': [{
'Start': start,
'End': end
}],
'DateCreatedFilter': 'true'
}
if daterange:
fields.update(daterange)
if 'terms' in criterion:
if 'type' in criterion.get('terms', {}):
fields.update({'MediaTypes': criterion['terms']['type']})
if 'credit' in criterion.get('terms', {}):
fields.update({'Credits': criterion['terms']['credit']})
if 'anpa_category.name' in criterion.get('terms', {}):
cat_list = []
for cat in criterion['terms']['anpa_category.name']:
qcode = [
key for key, value in subject_codes.items()
if value == cat
]
if qcode:
for code in qcode:
cat_list.append(code)
else:
cat_list.append(cat)
fields.update({'Categories': cat_list})
size = int(req.get('size',
'25')) if int(req.get('size', '25')) > 0 else 25
query = {
'Query': query_keywords,
'pageSize': str(size),
'pageNumber': str(int(req.get('from', '0')) // size + 1)
}
r = self._http.urlopen('POST',
url + '?' + urllib.parse.urlencode(query),
body=json.dumps(fields),
headers=self._headers)
hits = self._parse_hits(json.loads(r.data.decode('UTF-8')))
return ElasticCursor(docs=hits['docs'],
hits={
'hits': hits,
'aggregations': self._parse_aggregations(hits)
})
def test_all_subjects_map_to_a_media_topic(self):
for (k, v) in sorted(subject_codes.items()):
topic = self.topics.get_media_topic(k)
self.assertIsNotNone(topic)
