def append_to_document(self, document, nodevalue, nodeid, tile, provisional=False):
try:
assert isinstance(nodevalue, (list, tuple)) # assert nodevalue is an array
except AssertionError:
nodevalue = [nodevalue]
for valueid in nodevalue:
value = self.get_value(valueid)
date_range = self.get_concept_dates(value.concept)
if date_range is not None:
min_date = ExtendedDateFormat(date_range["min_year"]).lower
max_date = ExtendedDateFormat(date_range["max_year"]).upper
if {"gte": min_date, "lte": max_date} not in document["date_ranges"]:
document["date_ranges"].append(
{"date_range": {"gte": min_date, "lte": max_date}, "nodegroup_id": tile.nodegroup_id, "provisional": provisional}
)
document["domains"].append(
{
"label": value.value,
"conceptid": value.concept_id,
"valueid": valueid,
"nodegroup_id": tile.nodegroup_id,
"provisional": provisional,
}
)
document["strings"].append({"string": value.value, "nodegroup_id": tile.nodegroup_id, "provisional": provisional})
def parse(self, test_case):
i = test_case[0]
if isinstance(i, tuple):
i, o = i
else:
o = i
print("parsing '%s'" % i)
f = ExtendedDateFormat(i)
if len(test_case) == 2:
if f.result_set:
for i, item in enumerate(f.result_set):
self.assertEqual(str(item.lower), test_case[1][i][0])
self.assertEqual(str(item.upper), test_case[1][i][1])
else:
self.assertEqual(str(f.lower), test_case[1])
elif len(test_case) == 3:
self.assertEqual(str(f.lower), test_case[1])
self.assertEqual(str(f.upper), test_case[2])
elif len(test_case) == 4:
self.assertEqual(str(f.lower), test_case[1])
self.assertEqual(str(f.upper), test_case[1])
self.assertEqual(str(f.lower_fuzzy), test_case[2])
self.assertEqual(str(f.upper_fuzzy), test_case[3])
elif len(test_case) == 5:
self.assertEqual(str(f.lower), test_case[1])
self.assertEqual(str(f.upper), test_case[2])
self.assertEqual(str(f.lower_fuzzy), test_case[3])
self.assertEqual(str(f.upper_fuzzy), test_case[4])
elif len(test_case) == 1:
self.assertEqual(f.is_valid(), False)
else:
raise Exception(test_case)
def add_date_tier(date_tier,
low_date,
high_date,
previous_period_agg=None):
interval = date_tier["interval"]
name = date_tier["name"]
within_range = True
if "root" in date_tier:
high_date = int(high_date) + interval
for period in range(int(low_date), int(high_date), interval):
min_period = period
max_period = period + interval
if "range" in date_tier:
within_range = min_period >= date_tier["range"][
"min"] and max_period <= date_tier["range"]["max"]
period_name = "{0} ({1} - {2})".format(
name, min_period, max_period)
nodegroups = self.get_permitted_nodegroups(
user) if "root" in date_tier else None
period_boolquery = gen_range_agg(
gte=ExtendedDateFormat(min_period).lower,
lte=ExtendedDateFormat(max_period).lower,
permitted_nodegroups=nodegroups)
period_agg = FiltersAgg(name=period_name)
period_agg.add_filter(period_boolquery)
if "root" not in date_tier:
if within_range is True:
previous_period_agg.add_aggregation(period_agg)
range_lookup[period_name] = [min_period, max_period]
if "child" in date_tier:
add_date_tier(date_tier["child"], min_period,
max_period, period_agg)
if "root" in date_tier:
query.add_aggregation(period_agg)
def append_to_document(self, document, nodevalue, nodeid, tile, provisional=False):
try:
assert isinstance(nodevalue, (list, tuple)) #assert nodevalue is an array
except AssertionError:
nodevalue = [nodevalue]
for valueid in nodevalue:
value = self.get_value(valueid)
date_range = self.get_concept_dates(value.concept)
if date_range is not None:
min_date = ExtendedDateFormat(date_range['min_year']).lower
max_date = ExtendedDateFormat(date_range['max_year']).upper
if {'gte': min_date, 'lte': max_date} not in document['date_ranges']:
document['date_ranges'].append({'date_range': {'gte': min_date, 'lte': max_date}, 'nodegroup_id': tile.nodegroup_id, 'provisional': provisional})
document['domains'].append({'label': value.value, 'conceptid': value.concept_id, 'valueid': valueid, 'nodegroup_id': tile.nodegroup_id, 'provisional': provisional})
document['strings'].append({'string': value.value, 'nodegroup_id': tile.nodegroup_id, 'provisional': provisional})
def time_wheel_config(request):
se = SearchEngineFactory().create()
query = Query(se, limit=0)
nested_agg = NestedAgg(path='dates', name='min_max_agg')
nested_agg.add_aggregation(MinAgg(field='dates.date'))
nested_agg.add_aggregation(MaxAgg(field='dates.date'))
query.add_aggregation(nested_agg)
results = query.search(index='resource')
if results is not None and results['aggregations']['min_max_agg'][
'min_dates.date']['value'] is not None and results['aggregations'][
'min_max_agg']['max_dates.date']['value'] is not None:
min_date = int(results['aggregations']['min_max_agg']['min_dates.date']
['value']) / 10000
max_date = int(results['aggregations']['min_max_agg']['max_dates.date']
['value']) / 10000
# round min and max date to the nearest 1000 years
min_date = math.ceil(math.fabs(min_date) /
1000) * -1000 if min_date < 0 else math.floor(
min_date / 1000) * 1000
max_date = math.floor(math.fabs(max_date) /
1000) * -1000 if max_date < 0 else math.ceil(
max_date / 1000) * 1000
query = Query(se, limit=0)
range_lookup = {}
def gen_range_agg(gte=None, lte=None, permitted_nodegroups=None):
date_query = Bool()
date_query.filter(
Range(field='dates.date',
gte=gte,
lte=lte,
relation='intersects'))
if permitted_nodegroups:
date_query.filter(
Terms(field='dates.nodegroup_id',
terms=permitted_nodegroups))
date_ranges_query = Bool()
date_ranges_query.filter(
Range(field='date_ranges.date_range',
gte=gte,
lte=lte,
relation='intersects'))
if permitted_nodegroups:
date_ranges_query.filter(
Terms(field='date_ranges.nodegroup_id',
terms=permitted_nodegroups))
wrapper_query = Bool()
wrapper_query.should(
Nested(path='date_ranges', query=date_ranges_query))
wrapper_query.should(Nested(path='dates', query=date_query))
return wrapper_query
for millennium in range(int(min_date), int(max_date) + 1000, 1000):
min_millenium = millennium
max_millenium = millennium + 1000
millenium_name = "Millennium (%s - %s)" % (min_millenium,
max_millenium)
mill_boolquery = gen_range_agg(
gte=ExtendedDateFormat(min_millenium).lower,
lte=ExtendedDateFormat(max_millenium).lower,
permitted_nodegroups=get_permitted_nodegroups(request.user))
millenium_agg = FiltersAgg(name=millenium_name)
millenium_agg.add_filter(mill_boolquery)
range_lookup[millenium_name] = [min_millenium, max_millenium]
for century in range(min_millenium, max_millenium, 100):
min_century = century
max_century = century + 100
century_name = "Century (%s - %s)" % (min_century, max_century)
cent_boolquery = gen_range_agg(
gte=ExtendedDateFormat(min_century).lower,
lte=ExtendedDateFormat(max_century).lower)
century_agg = FiltersAgg(name=century_name)
century_agg.add_filter(cent_boolquery)
millenium_agg.add_aggregation(century_agg)
range_lookup[century_name] = [min_century, max_century]
for decade in range(min_century, max_century, 10):
min_decade = decade
max_decade = decade + 10
decade_name = "Decade (%s - %s)" % (min_decade, max_decade)
dec_boolquery = gen_range_agg(
gte=ExtendedDateFormat(min_decade).lower,
lte=ExtendedDateFormat(max_decade).lower)
decade_agg = FiltersAgg(name=decade_name)
decade_agg.add_filter(dec_boolquery)
century_agg.add_aggregation(decade_agg)
range_lookup[decade_name] = [min_decade, max_decade]
query.add_aggregation(millenium_agg)
root = d3Item(name='root')
results = {'buckets': [query.search(index='resource')['aggregations']]}
results_with_ranges = appendDateRanges(results, range_lookup)
transformESAggToD3Hierarchy(results_with_ranges, root)
return JSONResponse(root, indent=4)
else:
return HttpResponseNotFound(
_('Error retrieving the time wheel config'))
def build_search_results_dsl(request):
term_filter = request.GET.get('termFilter', '')
spatial_filter = JSONDeserializer().deserialize(
request.GET.get('mapFilter', '{}'))
include_provisional = get_provisional_type(request)
export = request.GET.get('export', None)
mobile_download = request.GET.get('mobiledownload', None)
page = 1 if request.GET.get('page') == '' else int(
request.GET.get('page', 1))
temporal_filter = JSONDeserializer().deserialize(
request.GET.get('temporalFilter', '{}'))
advanced_filters = JSONDeserializer().deserialize(
request.GET.get('advanced', '[]'))
search_buffer = None
se = SearchEngineFactory().create()
if export != None:
limit = settings.SEARCH_EXPORT_ITEMS_PER_PAGE
elif mobile_download != None:
limit = settings.MOBILE_DOWNLOAD_RESOURCE_LIMIT
else:
limit = settings.SEARCH_ITEMS_PER_PAGE
query = Query(se, start=limit * int(page - 1), limit=limit)
search_query = Bool()
nested_agg = NestedAgg(path='points', name='geo_aggs')
nested_agg_filter = FiltersAgg(name='inner')
if include_provisional == True:
nested_agg_filter.add_filter(
Terms(field='points.provisional', terms=['false', 'true']))
else:
provisional_resource_filter = Bool()
if include_provisional == False:
provisional_resource_filter.filter(
Terms(field='provisional', terms=['false', 'partial']))
nested_agg_filter.add_filter(
Terms(field='points.provisional', terms=['false']))
elif include_provisional == 'only provisional':
provisional_resource_filter.filter(
Terms(field='provisional', terms=['true', 'partial']))
nested_agg_filter.add_filter(
Terms(field='points.provisional', terms=['true']))
search_query.must(provisional_resource_filter)
nested_agg_filter.add_aggregation(
GeoHashGridAgg(field='points.point',
name='grid',
precision=settings.HEX_BIN_PRECISION))
nested_agg_filter.add_aggregation(
GeoBoundsAgg(field='points.point', name='bounds'))
nested_agg.add_aggregation(nested_agg_filter)
query.add_aggregation(nested_agg)
permitted_nodegroups = get_permitted_nodegroups(request.user)
if term_filter != '':
for term in JSONDeserializer().deserialize(term_filter):
term_query = Bool()
provisional_term_filter = Bool()
if term['type'] == 'term' or term['type'] == 'string':
string_filter = Bool()
if term['type'] == 'term':
string_filter.must(
Match(field='strings.string',
query=term['value'],
type='phrase'))
elif term['type'] == 'string':
string_filter.should(
Match(field='strings.string',
query=term['value'],
type='phrase_prefix'))
string_filter.should(
Match(field='strings.string.folded',
query=term['value'],
type='phrase_prefix'))
if include_provisional == False:
string_filter.must_not(
Match(field='strings.provisional',
query='true',
type='phrase'))
elif include_provisional == 'only provisional':
string_filter.must_not(
Match(field='strings.provisional',
query='false',
type='phrase'))
string_filter.filter(
Terms(field='strings.nodegroup_id',
terms=permitted_nodegroups))
nested_string_filter = Nested(path='strings',
query=string_filter)
if term['inverted']:
search_query.must_not(nested_string_filter)
else:
search_query.must(nested_string_filter)
# need to set min_score because the query returns results with score 0 and those have to be removed, which I don't think it should be doing
query.min_score('0.01')
elif term['type'] == 'concept':
concept_ids = _get_child_concepts(term['value'])
conceptid_filter = Bool()
conceptid_filter.filter(
Terms(field='domains.conceptid', terms=concept_ids))
conceptid_filter.filter(
Terms(field='domains.nodegroup_id',
terms=permitted_nodegroups))
if include_provisional == False:
conceptid_filter.must_not(
Match(field='domains.provisional',
query='true',
type='phrase'))
elif include_provisional == 'only provisional':
conceptid_filter.must_not(
Match(field='domains.provisional',
query='false',
type='phrase'))
nested_conceptid_filter = Nested(path='domains',
query=conceptid_filter)
if term['inverted']:
search_query.must_not(nested_conceptid_filter)
else:
search_query.filter(nested_conceptid_filter)
if 'features' in spatial_filter:
if len(spatial_filter['features']) > 0:
feature_geom = spatial_filter['features'][0]['geometry']
feature_properties = {}
if 'properties' in spatial_filter['features'][0]:
feature_properties = spatial_filter['features'][0][
'properties']
buffer = {'width': 0, 'unit': 'ft'}
if 'buffer' in feature_properties:
buffer = feature_properties['buffer']
search_buffer = _buffer(feature_geom, buffer['width'],
buffer['unit'])
feature_geom = JSONDeserializer().deserialize(search_buffer.json)
geoshape = GeoShape(field='geometries.geom.features.geometry',
type=feature_geom['type'],
coordinates=feature_geom['coordinates'])
invert_spatial_search = False
if 'inverted' in feature_properties:
invert_spatial_search = feature_properties['inverted']
spatial_query = Bool()
if invert_spatial_search == True:
spatial_query.must_not(geoshape)
else:
spatial_query.filter(geoshape)
# get the nodegroup_ids that the user has permission to search
spatial_query.filter(
Terms(field='geometries.nodegroup_id',
terms=permitted_nodegroups))
if include_provisional == False:
spatial_query.filter(
Terms(field='geometries.provisional', terms=['false']))
elif include_provisional == 'only provisional':
spatial_query.filter(
Terms(field='geometries.provisional', terms=['true']))
search_query.filter(Nested(path='geometries', query=spatial_query))
if 'fromDate' in temporal_filter and 'toDate' in temporal_filter:
now = str(datetime.utcnow())
start_date = ExtendedDateFormat(temporal_filter['fromDate'])
end_date = ExtendedDateFormat(temporal_filter['toDate'])
date_nodeid = str(
temporal_filter['dateNodeId']
) if 'dateNodeId' in temporal_filter and temporal_filter[
'dateNodeId'] != '' else None
query_inverted = False if 'inverted' not in temporal_filter else temporal_filter[
'inverted']
temporal_query = Bool()
if query_inverted:
# inverted date searches need to use an OR clause and are generally more complicated to structure (can't use ES must_not)
# eg: less than START_DATE OR greater than END_DATE
inverted_date_query = Bool()
inverted_date_ranges_query = Bool()
if start_date.is_valid():
inverted_date_query.should(
Range(field='dates.date', lt=start_date.lower))
inverted_date_ranges_query.should(
Range(field='date_ranges.date_range', lt=start_date.lower))
if end_date.is_valid():
inverted_date_query.should(
Range(field='dates.date', gt=end_date.upper))
inverted_date_ranges_query.should(
Range(field='date_ranges.date_range', gt=end_date.upper))
date_query = Bool()
date_query.filter(inverted_date_query)
date_query.filter(
Terms(field='dates.nodegroup_id', terms=permitted_nodegroups))
if include_provisional == False:
date_query.filter(
Terms(field='dates.provisional', terms=['false']))
elif include_provisional == 'only provisional':
date_query.filter(
Terms(field='dates.provisional', terms=['true']))
if date_nodeid:
date_query.filter(Term(field='dates.nodeid', term=date_nodeid))
else:
date_ranges_query = Bool()
date_ranges_query.filter(inverted_date_ranges_query)
date_ranges_query.filter(
Terms(field='date_ranges.nodegroup_id',
terms=permitted_nodegroups))
if include_provisional == False:
date_ranges_query.filter(
Terms(field='date_ranges.provisional',
terms=['false']))
elif include_provisional == 'only provisional':
date_ranges_query.filter(
Terms(field='date_ranges.provisional', terms=['true']))
temporal_query.should(
Nested(path='date_ranges', query=date_ranges_query))
temporal_query.should(Nested(path='dates', query=date_query))
else:
date_query = Bool()
date_query.filter(
Range(field='dates.date',
gte=start_date.lower,
lte=end_date.upper))
date_query.filter(
Terms(field='dates.nodegroup_id', terms=permitted_nodegroups))
if include_provisional == False:
date_query.filter(
Terms(field='dates.provisional', terms=['false']))
elif include_provisional == 'only provisional':
date_query.filter(
Terms(field='dates.provisional', terms=['true']))
if date_nodeid:
date_query.filter(Term(field='dates.nodeid', term=date_nodeid))
else:
date_ranges_query = Bool()
date_ranges_query.filter(
Range(field='date_ranges.date_range',
gte=start_date.lower,
lte=end_date.upper,
relation='intersects'))
date_ranges_query.filter(
Terms(field='date_ranges.nodegroup_id',
terms=permitted_nodegroups))
if include_provisional == False:
date_ranges_query.filter(
Terms(field='date_ranges.provisional',
terms=['false']))
if include_provisional == 'only provisional':
date_ranges_query.filter(
Terms(field='date_ranges.provisional', terms=['true']))
temporal_query.should(
Nested(path='date_ranges', query=date_ranges_query))
temporal_query.should(Nested(path='dates', query=date_query))
search_query.filter(temporal_query)
datatype_factory = DataTypeFactory()
if len(advanced_filters) > 0:
advanced_query = Bool()
grouped_query = Bool()
grouped_queries = [grouped_query]
for index, advanced_filter in enumerate(advanced_filters):
tile_query = Bool()
for key, val in advanced_filter.iteritems():
if key != 'op':
node = models.Node.objects.get(pk=key)
if request.user.has_perm('read_nodegroup', node.nodegroup):
datatype = datatype_factory.get_instance(node.datatype)
datatype.append_search_filters(val, node, tile_query,
request)
nested_query = Nested(path='tiles', query=tile_query)
if advanced_filter['op'] == 'or' and index != 0:
grouped_query = Bool()
grouped_queries.append(grouped_query)
grouped_query.must(nested_query)
for grouped_query in grouped_queries:
advanced_query.should(grouped_query)
search_query.must(advanced_query)
query.add_query(search_query)
if search_buffer != None:
search_buffer = search_buffer.geojson
return {'query': query, 'search_buffer': search_buffer}
def append_dsl(self, search_results_object, permitted_nodegroups,
include_provisional):
search_query = Bool()
querysting_params = self.request.GET.get(details["componentname"], "")
temporal_filter = JSONDeserializer().deserialize(querysting_params)
if "fromDate" in temporal_filter and "toDate" in temporal_filter:
# now = str(datetime.utcnow())
start_date = ExtendedDateFormat(temporal_filter["fromDate"])
end_date = ExtendedDateFormat(temporal_filter["toDate"])
date_nodeid = (str(temporal_filter["dateNodeId"])
if "dateNodeId" in temporal_filter
and temporal_filter["dateNodeId"] != "" else None)
query_inverted = False if "inverted" not in temporal_filter else temporal_filter[
"inverted"]
temporal_query = Bool()
if query_inverted:
# inverted date searches need to use an OR clause and are generally more complicated to structure (can't use ES must_not)
# eg: less than START_DATE OR greater than END_DATE
inverted_date_query = Bool()
inverted_date_ranges_query = Bool()
if start_date.is_valid():
inverted_date_query.should(
Range(field="dates.date", lt=start_date.lower))
inverted_date_ranges_query.should(
Range(field="date_ranges.date_range",
lt=start_date.lower))
if end_date.is_valid():
inverted_date_query.should(
Range(field="dates.date", gt=end_date.upper))
inverted_date_ranges_query.should(
Range(field="date_ranges.date_range",
gt=end_date.upper))
date_query = Bool()
date_query.filter(inverted_date_query)
date_query.filter(
Terms(field="dates.nodegroup_id",
terms=permitted_nodegroups))
if include_provisional is False:
date_query.filter(
Terms(field="dates.provisional", terms=["false"]))
elif include_provisional == "only provisional":
date_query.filter(
Terms(field="dates.provisional", terms=["true"]))
if date_nodeid:
date_query.filter(
Term(field="dates.nodeid", term=date_nodeid))
else:
date_ranges_query = Bool()
date_ranges_query.filter(inverted_date_ranges_query)
date_ranges_query.filter(
Terms(field="date_ranges.nodegroup_id",
terms=permitted_nodegroups))
if include_provisional is False:
date_ranges_query.filter(
Terms(field="date_ranges.provisional",
terms=["false"]))
elif include_provisional == "only provisional":
date_ranges_query.filter(
Terms(field="date_ranges.provisional",
terms=["true"]))
temporal_query.should(
Nested(path="date_ranges", query=date_ranges_query))
temporal_query.should(Nested(path="dates", query=date_query))
else:
date_query = Bool()
date_query.filter(
Range(field="dates.date",
gte=start_date.lower,
lte=end_date.upper))
date_query.filter(
Terms(field="dates.nodegroup_id",
terms=permitted_nodegroups))
if include_provisional is False:
date_query.filter(
Terms(field="dates.provisional", terms=["false"]))
elif include_provisional == "only provisional":
date_query.filter(
Terms(field="dates.provisional", terms=["true"]))
if date_nodeid:
date_query.filter(
Term(field="dates.nodeid", term=date_nodeid))
else:
date_ranges_query = Bool()
date_ranges_query.filter(
Range(field="date_ranges.date_range",
gte=start_date.lower,
lte=end_date.upper,
relation="intersects"))
date_ranges_query.filter(
Terms(field="date_ranges.nodegroup_id",
terms=permitted_nodegroups))
if include_provisional is False:
date_ranges_query.filter(
Terms(field="date_ranges.provisional",
terms=["false"]))
if include_provisional == "only provisional":
date_ranges_query.filter(
Terms(field="date_ranges.provisional",
terms=["true"]))
temporal_query.should(
Nested(path="date_ranges", query=date_ranges_query))
temporal_query.should(Nested(path="dates", query=date_query))
search_query.filter(temporal_query)
search_results_object["query"].add_query(search_query)
def append_dsl(self, search_results_object, permitted_nodegroups,
include_provisional):
search_query = Bool()
querysting_params = self.request.GET.get(details['componentname'], '')
temporal_filter = JSONDeserializer().deserialize(querysting_params)
if 'fromDate' in temporal_filter and 'toDate' in temporal_filter:
#now = str(datetime.utcnow())
start_date = ExtendedDateFormat(temporal_filter['fromDate'])
end_date = ExtendedDateFormat(temporal_filter['toDate'])
date_nodeid = str(
temporal_filter['dateNodeId']
) if 'dateNodeId' in temporal_filter and temporal_filter[
'dateNodeId'] != '' else None
query_inverted = False if 'inverted' not in temporal_filter else temporal_filter[
'inverted']
temporal_query = Bool()
if query_inverted:
# inverted date searches need to use an OR clause and are generally more complicated to structure (can't use ES must_not)
# eg: less than START_DATE OR greater than END_DATE
inverted_date_query = Bool()
inverted_date_ranges_query = Bool()
if start_date.is_valid():
inverted_date_query.should(
Range(field='dates.date', lt=start_date.lower))
inverted_date_ranges_query.should(
Range(field='date_ranges.date_range',
lt=start_date.lower))
if end_date.is_valid():
inverted_date_query.should(
Range(field='dates.date', gt=end_date.upper))
inverted_date_ranges_query.should(
Range(field='date_ranges.date_range',
gt=end_date.upper))
date_query = Bool()
date_query.filter(inverted_date_query)
date_query.filter(
Terms(field='dates.nodegroup_id',
terms=permitted_nodegroups))
if include_provisional is False:
date_query.filter(
Terms(field='dates.provisional', terms=['false']))
elif include_provisional == 'only provisional':
date_query.filter(
Terms(field='dates.provisional', terms=['true']))
if date_nodeid:
date_query.filter(
Term(field='dates.nodeid', term=date_nodeid))
else:
date_ranges_query = Bool()
date_ranges_query.filter(inverted_date_ranges_query)
date_ranges_query.filter(
Terms(field='date_ranges.nodegroup_id',
terms=permitted_nodegroups))
if include_provisional is False:
date_ranges_query.filter(
Terms(field='date_ranges.provisional',
terms=['false']))
elif include_provisional == 'only provisional':
date_ranges_query.filter(
Terms(field='date_ranges.provisional',
terms=['true']))
temporal_query.should(
Nested(path='date_ranges', query=date_ranges_query))
temporal_query.should(Nested(path='dates', query=date_query))
else:
date_query = Bool()
date_query.filter(
Range(field='dates.date',
gte=start_date.lower,
lte=end_date.upper))
date_query.filter(
Terms(field='dates.nodegroup_id',
terms=permitted_nodegroups))
if include_provisional is False:
date_query.filter(
Terms(field='dates.provisional', terms=['false']))
elif include_provisional == 'only provisional':
date_query.filter(
Terms(field='dates.provisional', terms=['true']))
if date_nodeid:
date_query.filter(
Term(field='dates.nodeid', term=date_nodeid))
else:
date_ranges_query = Bool()
date_ranges_query.filter(
Range(field='date_ranges.date_range',
gte=start_date.lower,
lte=end_date.upper,
relation='intersects'))
date_ranges_query.filter(
Terms(field='date_ranges.nodegroup_id',
terms=permitted_nodegroups))
if include_provisional is False:
date_ranges_query.filter(
Terms(field='date_ranges.provisional',
terms=['false']))
if include_provisional == 'only provisional':
date_ranges_query.filter(
Terms(field='date_ranges.provisional',
terms=['true']))
temporal_query.should(
Nested(path='date_ranges', query=date_ranges_query))
temporal_query.should(Nested(path='dates', query=date_query))
search_query.filter(temporal_query)
search_results_object['query'].add_query(search_query)
def build_search_results_dsl(request):
term_filter = request.GET.get('termFilter', '')
spatial_filter = JSONDeserializer().deserialize(request.GET.get('mapFilter', '{}'))
include_provisional = get_provisional_type(request)
export = request.GET.get('export', None)
mobile_download = request.GET.get('mobiledownload', None)
page = 1 if request.GET.get('page') == '' else int(request.GET.get('page', 1))
temporal_filter = JSONDeserializer().deserialize(request.GET.get('temporalFilter', '{}'))
advanced_filters = JSONDeserializer().deserialize(request.GET.get('advanced', '[]'))
search_buffer = None
se = SearchEngineFactory().create()
if export != None:
limit = settings.SEARCH_EXPORT_ITEMS_PER_PAGE
elif mobile_download != None:
limit = settings.MOBILE_DOWNLOAD_RESOURCE_LIMIT
else:
limit = settings.SEARCH_ITEMS_PER_PAGE
query = Query(se, start=limit*int(page-1), limit=limit)
search_query = Bool()
nested_agg = NestedAgg(path='points', name='geo_aggs')
nested_agg_filter = FiltersAgg(name='inner')
if include_provisional == True:
nested_agg_filter.add_filter(Terms(field='points.provisional', terms=['false','true']))
else:
provisional_resource_filter = Bool()
if include_provisional == False:
provisional_resource_filter.filter(Terms(field='provisional', terms=['false', 'partial']))
nested_agg_filter.add_filter(Terms(field='points.provisional', terms=['false']))
elif include_provisional == 'only provisional':
provisional_resource_filter.filter(Terms(field='provisional', terms=['true', 'partial']))
nested_agg_filter.add_filter(Terms(field='points.provisional', terms=['true']))
search_query.must(provisional_resource_filter)
nested_agg_filter.add_aggregation(GeoHashGridAgg(field='points.point', name='grid', precision=settings.HEX_BIN_PRECISION))
nested_agg_filter.add_aggregation(GeoBoundsAgg(field='points.point', name='bounds'))
nested_agg.add_aggregation(nested_agg_filter)
query.add_aggregation(nested_agg)
permitted_nodegroups = get_permitted_nodegroups(request.user)
if term_filter != '':
for term in JSONDeserializer().deserialize(term_filter):
term_query = Bool()
provisional_term_filter = Bool()
if term['type'] == 'term' or term['type'] == 'string':
string_filter = Bool()
if term['type'] == 'term':
string_filter.must(Match(field='strings.string', query=term['value'], type='phrase'))
elif term['type'] == 'string':
string_filter.should(Match(field='strings.string', query=term['value'], type='phrase_prefix'))
string_filter.should(Match(field='strings.string.folded', query=term['value'], type='phrase_prefix'))
if include_provisional == False:
string_filter.must_not(Match(field='strings.provisional', query='true', type='phrase'))
elif include_provisional == 'only provisional':
string_filter.must_not(Match(field='strings.provisional', query='false', type='phrase'))
string_filter.filter(Terms(field='strings.nodegroup_id', terms=permitted_nodegroups))
nested_string_filter = Nested(path='strings', query=string_filter)
if term['inverted']:
search_query.must_not(nested_string_filter)
else:
search_query.must(nested_string_filter)
# need to set min_score because the query returns results with score 0 and those have to be removed, which I don't think it should be doing
query.min_score('0.01')
elif term['type'] == 'concept':
concept_ids = _get_child_concepts(term['value'])
conceptid_filter = Bool()
conceptid_filter.filter(Terms(field='domains.conceptid', terms=concept_ids))
conceptid_filter.filter(Terms(field='domains.nodegroup_id', terms=permitted_nodegroups))
if include_provisional == False:
conceptid_filter.must_not(Match(field='domains.provisional', query='true', type='phrase'))
elif include_provisional == 'only provisional':
conceptid_filter.must_not(Match(field='domains.provisional', query='false', type='phrase'))
nested_conceptid_filter = Nested(path='domains', query=conceptid_filter)
if term['inverted']:
search_query.must_not(nested_conceptid_filter)
else:
search_query.filter(nested_conceptid_filter)
if 'features' in spatial_filter:
if len(spatial_filter['features']) > 0:
feature_geom = spatial_filter['features'][0]['geometry']
feature_properties = {}
if 'properties' in spatial_filter['features'][0]:
feature_properties = spatial_filter['features'][0]['properties']
buffer = {'width':0,'unit':'ft'}
if 'buffer' in feature_properties:
buffer = feature_properties['buffer']
search_buffer = _buffer(feature_geom, buffer['width'], buffer['unit'])
feature_geom = JSONDeserializer().deserialize(search_buffer.json)
geoshape = GeoShape(field='geometries.geom.features.geometry', type=feature_geom['type'], coordinates=feature_geom['coordinates'] )
invert_spatial_search = False
if 'inverted' in feature_properties:
invert_spatial_search = feature_properties['inverted']
spatial_query = Bool()
if invert_spatial_search == True:
spatial_query.must_not(geoshape)
else:
spatial_query.filter(geoshape)
# get the nodegroup_ids that the user has permission to search
spatial_query.filter(Terms(field='geometries.nodegroup_id', terms=permitted_nodegroups))
if include_provisional == False:
spatial_query.filter(Terms(field='geometries.provisional', terms=['false']))
elif include_provisional == 'only provisional':
spatial_query.filter(Terms(field='geometries.provisional', terms=['true']))
search_query.filter(Nested(path='geometries', query=spatial_query))
if 'fromDate' in temporal_filter and 'toDate' in temporal_filter:
now = str(datetime.utcnow())
start_date = ExtendedDateFormat(temporal_filter['fromDate'])
end_date = ExtendedDateFormat(temporal_filter['toDate'])
date_nodeid = str(temporal_filter['dateNodeId']) if 'dateNodeId' in temporal_filter and temporal_filter['dateNodeId'] != '' else None
query_inverted = False if 'inverted' not in temporal_filter else temporal_filter['inverted']
temporal_query = Bool()
if query_inverted:
# inverted date searches need to use an OR clause and are generally more complicated to structure (can't use ES must_not)
# eg: less than START_DATE OR greater than END_DATE
inverted_date_query = Bool()
inverted_date_ranges_query = Bool()
if start_date.is_valid():
inverted_date_query.should(Range(field='dates.date', lt=start_date.lower))
inverted_date_ranges_query.should(Range(field='date_ranges.date_range', lt=start_date.lower))
if end_date.is_valid():
inverted_date_query.should(Range(field='dates.date', gt=end_date.upper))
inverted_date_ranges_query.should(Range(field='date_ranges.date_range', gt=end_date.upper))
date_query = Bool()
date_query.filter(inverted_date_query)
date_query.filter(Terms(field='dates.nodegroup_id', terms=permitted_nodegroups))
if include_provisional == False:
date_query.filter(Terms(field='dates.provisional', terms=['false']))
elif include_provisional == 'only provisional':
date_query.filter(Terms(field='dates.provisional', terms=['true']))
if date_nodeid:
date_query.filter(Term(field='dates.nodeid', term=date_nodeid))
else:
date_ranges_query = Bool()
date_ranges_query.filter(inverted_date_ranges_query)
date_ranges_query.filter(Terms(field='date_ranges.nodegroup_id', terms=permitted_nodegroups))
if include_provisional == False:
date_ranges_query.filter(Terms(field='date_ranges.provisional', terms=['false']))
elif include_provisional == 'only provisional':
date_ranges_query.filter(Terms(field='date_ranges.provisional', terms=['true']))
temporal_query.should(Nested(path='date_ranges', query=date_ranges_query))
temporal_query.should(Nested(path='dates', query=date_query))
else:
date_query = Bool()
date_query.filter(Range(field='dates.date', gte=start_date.lower, lte=end_date.upper))
date_query.filter(Terms(field='dates.nodegroup_id', terms=permitted_nodegroups))
if include_provisional == False:
date_query.filter(Terms(field='dates.provisional', terms=['false']))
elif include_provisional == 'only provisional':
date_query.filter(Terms(field='dates.provisional', terms=['true']))
if date_nodeid:
date_query.filter(Term(field='dates.nodeid', term=date_nodeid))
else:
date_ranges_query = Bool()
date_ranges_query.filter(Range(field='date_ranges.date_range', gte=start_date.lower, lte=end_date.upper, relation='intersects'))
date_ranges_query.filter(Terms(field='date_ranges.nodegroup_id', terms=permitted_nodegroups))
if include_provisional == False:
date_ranges_query.filter(Terms(field='date_ranges.provisional', terms=['false']))
if include_provisional == 'only provisional':
date_ranges_query.filter(Terms(field='date_ranges.provisional', terms=['true']))
temporal_query.should(Nested(path='date_ranges', query=date_ranges_query))
temporal_query.should(Nested(path='dates', query=date_query))
search_query.filter(temporal_query)
datatype_factory = DataTypeFactory()
if len(advanced_filters) > 0:
advanced_query = Bool()
grouped_query = Bool()
grouped_queries = [grouped_query]
for index, advanced_filter in enumerate(advanced_filters):
tile_query = Bool()
for key, val in advanced_filter.iteritems():
if key != 'op':
node = models.Node.objects.get(pk=key)
if request.user.has_perm('read_nodegroup', node.nodegroup):
datatype = datatype_factory.get_instance(node.datatype)
datatype.append_search_filters(val, node, tile_query, request)
nested_query = Nested(path='tiles', query=tile_query)
if advanced_filter['op'] == 'or' and index != 0:
grouped_query = Bool()
grouped_queries.append(grouped_query)
grouped_query.must(nested_query)
for grouped_query in grouped_queries:
advanced_query.should(grouped_query)
search_query.must(advanced_query)
query.add_query(search_query)
if search_buffer != None:
search_buffer = search_buffer.geojson
return {'query': query, 'search_buffer':search_buffer}
