def mock_find(resource, req, lookup, p):
doc = {
'fetch_endpoint': 'search_providers_proxy',
'pubstatus': 'usable',
'slugline': 'Fish on a bike',
'byline': 'Fred Smith/AAP PHOTOS',
'_id': '20200108001362610429',
'_type': 'externalsource',
'original_source': 'AAP Image/AAP',
'description_text': 'Sydney to the Gong some years ago',
'guid': '20200108001362610429',
'type': 'picture',
'firstcreated': utcnow(),
'ednote': 'Not for publication',
'source': 'AAP Image',
'headline': 'Fish on a bike',
'versioncreated': utcnow(),
'archive_description': 'Sydney to the Gong some years ago'
}
hits = {'docs': [doc], 'total': 1}
return ElasticCursor(docs=hits['docs'],
hits={
'hits': hits,
'aggregations': None
})
def mock_find_progress(resource, req, lookup, p):
doc = {}
hits = {'docs': [doc], 'total': 0}
return ElasticCursor(docs=hits['docs'],
hits={
'hits': hits,
'aggregations': None
})
def _parse_hits(self, hits):
schemas = {
'planning': self._get_resource_schema('planning'),
'events': self._get_resource_schema('events')
}
docs = []
for hit in hits.get('hits', {}).get('hits', []):
item_type = hit.get('_type')
schema = schemas.get(item_type)
dates = get_dates(schema)
doc = format_doc(hit, schema, dates)
build_custom_hateoas({'self': {'title': doc['_type'], 'href': '/{}/{{_id}}'.format(doc['_type'])}}, doc)
docs.append(doc)
return ElasticCursor(hits, docs)
def find(self, resource, req, lookup):
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']
fields = {
'query':
query_keywords,
'pageSize':
str(req.get('size', '25')),
'pageNumber':
str(int(req.get('from', '0')) // int(req.get('size', '25')) + 1)
}
r = self._http.request('GET',
url,
fields=fields,
headers=self._headers)
hits = self._parse_hits(json.loads(r.data.decode('UTF-8')))
return ElasticCursor(docs=hits['docs'], hits={'hits': hits})
def get(self, req, lookup):
"""
Return a list of items related to the given item. The given item id is retrieved
from the lookup dictionary as 'item_id'
"""
if "item_id" not in lookup:
raise SuperdeskApiError.badRequestError(
_("The item identifier is required"))
item = get_resource_service("archive_autosave").find_one(
req=None, _id=lookup["item_id"])
if not item:
item = get_resource_service("archive").find_one(
req=None, _id=lookup["item_id"])
if not item:
raise SuperdeskApiError.notFoundError(
_("Invalid item identifer"))
keywords = self.provider.get_keywords(self._transform(item))
if not keywords:
return ElasticCursor([])
query = {
"query": {
"filtered": {
"query": {
"query_string": {
"query": " ".join(kwd["text"] for kwd in keywords)
}
}
}
}
}
req = ParsedRequest()
req.args = {
"source": json.dumps(query),
"repo": "archive,published,archived"
}
return get_resource_service("search").get(req=req, lookup=None)
def get(self, req, lookup):
"""
Return a list of items related to the given item. The given item id is retrieved
from the lookup dictionary as 'item_id'
"""
if 'item_id' not in lookup:
raise SuperdeskApiError.badRequestError(
_('The item identifier is required'))
item = get_resource_service('archive_autosave').find_one(
req=None, _id=lookup['item_id'])
if not item:
item = get_resource_service('archive').find_one(
req=None, _id=lookup['item_id'])
if not item:
raise SuperdeskApiError.notFoundError(
_('Invalid item identifer'))
keywords = self.provider.get_keywords(self._transform(item))
if not keywords:
return ElasticCursor([])
query = {
'query': {
'filtered': {
'query': {
'query_string': {
'query': ' '.join(kwd['text'] for kwd in keywords)
}
}
}
}
}
req = ParsedRequest()
req.args = {
'source': json.dumps(query),
'repo': 'archive,published,archived'
}
return get_resource_service('search').get(req=req, lookup=None)
def find(self, resource, req, lookup):
"""
Called to execute a search against the Scanpix 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.
:param resource:
:param req:
:param lookup:
:return:
"""
url = self._app.config['SCANPIX_SEARCH_URL'] + '/search'
data = {'mainGroup': 'any'}
if 'query' in req['query']['filtered']:
query = req['query']['filtered']['query']['query_string']['query'] \
.replace('slugline:', 'keywords:') \
.replace('description:', 'caption:')
# Black & White
try:
bw = bool(int(extract_params(query, 'bw')['bw']))
except KeyError:
pass
else:
if bw:
data['saturation'] = {'max': 1}
# Clear Edge
try:
clear_edge = bool(
int(extract_params(query, 'clear_edge')['clear_edge']))
except KeyError:
pass
else:
if clear_edge:
data['clearEdge'] = True
# subscription
try:
data['subscription'] = extract_params(
query, 'subscription')['subscription']
except KeyError:
data[
'subscription'] = 'subscription' # this is requested as a default value
if 'ntbtema' in resource and data['subscription'] == 'subscription':
# small hack for SDNTB-250
data['subscription'] = 'punchcard'
if data['subscription'] == 'all':
del data['subscription']
text_params = extract_params(
query, ('headline', 'keywords', 'caption', 'text'))
# combine all possible text params to use the q field.
data['searchString'] = ' '.join(text_params.values())
try:
ids = extract_params(query, 'id')['id'].split()
except KeyError:
pass
else:
data['refPtrs'] = ids
for criterion in req.get('post_filter', {}).get('and', {}):
if 'range' in criterion:
start = None
end = None
filter_data = criterion.get('range', {})
if 'firstcreated' in filter_data:
created = criterion['range']['firstcreated']
if 'gte' in created:
start = created['gte'][0:10]
if 'lte' in created:
end = created['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':
data['timeLimit'] = 'last24'
if start == 'now-1w':
data['timeLimit'] = 'lastweek'
if start == 'now-1M':
data['timeLimit'] = 'lastmonth'
elif start or end:
data['archived'] = {'min': '', 'max': ''}
if start:
data['archived']['min'] = start
if end:
data['archived']['max'] = end
if 'terms' in criterion:
if 'type' in criterion.get('terms', {}):
type_ = criterion['terms']['type']
if type_ == CONTENT_TYPE.VIDEO:
data['mainGroup'] = 'video'
offset, limit = int(req.get('from',
'0')), max(10, int(req.get('size', '25')))
data['offset'] = offset
data['showNumResults'] = limit
r = self._request(url, data)
hits = self._parse_hits(r.json())
return ElasticCursor(docs=hits['docs'], hits={'hits': 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)
})
