class MediaSearch(BaseFacetedSearch):
doc_types = [MediaDocument]
fields = [
'tags',
'name',
]
facets = [
('tags', TermsFacet(field='tags', size=100)),
('type', TermsFacet(field='type', size=20, order={'_key': 'asc'})),
('version', TermsFacet(field='version',
size=100,
order={'_key': 'asc'})),
('num_emissions',
RangeFacet(field='num_emissions',
ranges=[
('0', (0, 1)),
('1 - 10', (1, 10)),
('11 - 50', (11, 50)),
('More than 50', (51, None)),
])),
('last_emission',
RangeFacet(
field='last_emission',
ranges=[
('Last 7 days', (str(
(timezone.now() - datetime.timedelta(days=7)).date()),
None)),
('Last month', (str(
(timezone.now() - datetime.timedelta(days=30)).date()),
None)),
('More than a month ago',
(None,
str((timezone.now() -
datetime.timedelta(days=30)).date()))),
])),
('bitrate',
RangeFacet(field='bitrate',
ranges=[('Low', (0, 100)), ('Medium', (100, 200)),
('High', (200, None))])),
('samplerate',
RangeFacet(field='samplerate',
ranges=[('Low', (0, 44100)), ('Medium', (44100, 48000)),
('High', (48000, None))])),
('encoding', TermsFacet(field='encoding', size=100)),
('license', TermsFacet(field='license')),
('lyrics_language', TermsFacet(field='lyrics_language')),
#('key', __paste__),
]
class LabelSearch(BaseFacetedSearch):
doc_types = [LabelDocument]
fields = [
'tags',
'name',
]
facets = [
('tags', TermsFacet(field='tags', size=100)),
('country', TermsFacet(field='country',
size=500,
order={'_key': 'asc'})),
('type', TermsFacet(field='type', size=20, order={'_key': 'asc'})),
('established',
RangeFacet(field='year_start',
ranges=[
('Before 1940\'s', (0, 1940)),
('40\'s', (1940, 1950)),
('50\'s', (1950, 1960)),
('60\'s', (1960, 1970)),
('70\'s', (1970, 1980)),
('80\'s', (1980, 1990)),
('90\'s', (1990, 2000)),
('2000\'s', (2000, 2010)),
('2010\'s', (2010, 2020)),
('This Year', (2018, 2019)),
])),
]
def get_configured_facets():
facets = None
configuration = get_configuration()
if configuration and hasattr(configuration, 'facets'):
facets = configuration.facets.split()
if facets:
configured_facets = {}
for facet in facets:
if ',' in facet:
field, interval = facet.split(',', 1)
if ',' in interval:
intervals = interval.split(',')
ranges = []
for interval in intervals:
name, numbers = interval.split(':')
numbers = numbers.split('-')
irange = []
for number in numbers:
if number.lower() == 'none':
irange.append(None)
else:
try:
irange.append(int(number))
except ValueError:
continue
irange = tuple(irange)
ranges.append((name, irange))
configured_facets[field] = RangeFacet(field=field,
ranges=ranges)
elif interval in DATE_INTERVALS:
configured_facets[field] = DateHistogramFacet(
field=field, interval=interval)
else:
try:
interval = int(interval)
configured_facets[field] = HistogramFacet(
field=field, interval=interval)
except ValueError:
pass
else:
configured_facets[facet] = TermsFacet(field=facet + '.keyword')
else:
configured_facets = DEFAULT_FACETS
return configured_facets
class SouSearch(FacetedSearch):
index = ES_INDEX_NAME # Index to search
fields = ['id_year_number^2', 'title^3', 'full_text'] # Fields to search
facets = {
'type': TermsFacet(field='type'),
'year': RangeFacet(field='year', ranges=[
('1922-1929', (1922, 1930)),
('1930-1939', (1930, 1940)),
('1940-1949', (1940, 1950)),
('1950-1959', (1950, 1960)),
('1960-1969', (1960, 1970)),
('1970-1979', (1970, 1980)),
('1980-1989', (1980, 1990)),
('1990-1999', (1990, 2000)),
('2000-2009', (2000, 2010)),
('2010-2019', (2010, 2020)),
('2020-', (2020, 2030)),
]),
}
def highlight(self, search):
return search.highlight('title', 'full_text', fragment_size=150, number_of_fragments=4)
def search(self):
s = super(SouSearch, self).search()
# Don't include the actual fulltext in result; we only need the highlighted extract
return s.source(excludes=["full_text"])
def query(self, search, query):
if query:
if self.fields:
return search.query('query_string', fields=self.fields, query=query, default_operator='and')
else:
return search.query('query_string', query=query, default_operator='and')
return search
class PlaylistSearch(BaseFacetedSearch):
doc_types = [PlaylistDocument]
fields = [
'tags',
'name',
]
facets = [
('tags', TermsFacet(field='tags', size=240)),
('type', TermsFacet(field='type', size=100, order={'_key': 'asc'})),
('status', TermsFacet(field='status', size=100, order={'_key':
'asc'})),
('weather', TermsFacet(field='weather',
size=100,
order={'_key': 'asc'})),
('seasons', TermsFacet(field='seasons',
size=100,
order={'_key': 'asc'})),
('daypart_days',
TermsFacet(field='daypart_days', size=100, order={'_key': 'asc'})),
('daypart_slots',
TermsFacet(field='daypart_slots', size=100, order={'_key': 'asc'})),
('target_duration', TermsFacet(field='target_duration', size=100)),
('num_emissions',
RangeFacet(field='num_emissions',
ranges=[
('0', (0, 1)),
('1 - 10', (1, 10)),
('11 - 50', (11, 50)),
('More than 50', (50, None)),
])),
(
'last_emission',
RangeFacet(
field='last_emission',
ranges=[
# ('old', (None, '2016-01-01')),
# ('2016', ('2016-01-01', '2016-12-31')),
# ('2018', ('2018-01-01', '2018-12-31')),
# ('recent', ('2019-01-01', None)),
('Last 7 days', (str(
(timezone.now() - datetime.timedelta(days=7)).date()),
None)),
('Last month', (str(
(timezone.now() - datetime.timedelta(days=30)).date()),
None)),
('More than a month ago',
(None,
str((timezone.now() -
datetime.timedelta(days=30)).date()))),
('More than year month ago',
(None,
str((timezone.now() -
datetime.timedelta(days=365)).date()))),
# ('Exclude 48 hours', (
# None,
# str((timezone.now() - datetime.timedelta(hours=48)).date())
# )),
])),
(
'next_emission',
RangeFacet(
field='next_emission',
ranges=[
# ('old', (None, '2016-01-01')),
# ('2016', ('2016-01-01', '2016-12-31')),
# ('2018', ('2018-01-01', '2018-12-31')),
# ('recent', ('2019-01-01', None)),
('Next 7 days', (None,
str((timezone.now() +
datetime.timedelta(days=7)).date()))
),
# ('Exclude 48 hours', (
# str((timezone.now() + datetime.timedelta(hours=48)).date()),
# None
# )),
])),
#('last_emission', DateHistogramFacet(field='last_emission', interval='month')),
('flags', TermsFacet(field='state_flags', order={'_key': 'asc'})),
]
class TestFacetedFilterField(object):
test_field_name = "faceted_filter_field"
@pytest.mark.parametrize(
', '.join(['facets', 'context', 'aggs_query']),
[(None, ['date'], {}),
({
'status':
TermsFacet(field='status'),
'date':
DateHistogramFacet(field='date', interval='year'),
'range':
RangeFacet(field='height',
ranges=[("few", (None, 2)), ("lots", (2, None))])
}, ['unknown'], {}),
({
'status': TermsFacet(field='status')
}, ['status'], {
'aggs': {
'_filter_status': {
'aggs': {
'status': {
'terms': {
'field': 'status'
}
}
},
'filter': {
'match_all': {}
}
}
}
}),
({
'date': DateHistogramFacet(field='date', interval='year')
}, ['date'], {
'aggs': {
'_filter_date': {
'aggs': {
'date': {
'date_histogram': {
'field': 'date',
'interval': 'year',
'min_doc_count': 0
}
}
},
'filter': {
'match_all': {}
}
}
},
}),
({
'range':
RangeFacet(field='height',
ranges=[("few", (None, 2)), ("lots", (2, None))])
}, ['range'], {
'aggs': {
'_filter_range': {
'aggs': {
'range': {
'range': {
'field':
'height',
'keyed':
False,
'ranges': [{
'key': 'few',
'to': 2
}, {
'key': 'lots',
'from': 2
}]
}
}
},
'filter': {
'match_all': {}
}
}
}
})])
def test_queryset(self, facets, context, aggs_query, es_dsl_queryset):
fld = FacetedFilterField(facets=facets,
field_name=self.test_field_name)
valid_query = {}
valid_query.update(aggs_query)
qs = fld.prepare_queryset(es_dsl_queryset, context)
ret = qs.to_dict()
assert ret == valid_query