def get_sphinx_data(sphinx_id):
query = Search(indexes=['sourcecode'], config=BaseSearchConfig)
query = query.filter(id__eq=sphinx_id)
results = query.ask()
if len(results['result']['items']) == 0:
flask.abort(404)
filename = results['result']['items'][0]['path']
if not os.path.isfile(filename):
return filename, 'File not found. Please resphinx_id.'
code = ''
with open(filename) as f:
code = f.read()
try:
# This is the line that throws the exception.
lexer = get_lexer_for_filename(filename)
formatter = HtmlFormatter(noclasses=True)
result = highlight(code, lexer, formatter)
except ClassNotFound:
# Syntax highlighting not supported.'
result = '<pre>{}</pre>'.format(code)
url = flask.url_for('display', sphinx_id=sphinx_id)
return {'body': result, 'path': filename, 'url': url}
def test_with_options(self):
search = Search(['company'], config=SearchConfig)
search = (
search
.match('Yandex')
.select('id', 'name')
.options(
ranker='proximity',
max_matches=100,
field_weights={'name': 100},
)
.order_by('name', 'desc')
)
correct_options_qls = [
"max_matches=100, ranker=proximity, field_weights=(name=100)",
"max_matches=100, field_weights=(name=100), ranker=proximity",
"ranker=proximity, field_weights=(name=100), max_matches=100",
"ranker=proximity, max_matches=100, field_weights=(name=100)",
"field_weights=(name=100), max_matches=100, ranker=proximity",
"field_weights=(name=100), ranker=proximity, max_matches=100",
]
correct_qls = [
" ".join((
"SELECT id, name FROM company WHERE MATCH('Yandex') ORDER BY name DESC OPTION",
opt
))
for opt in correct_options_qls
]
self.assertIn(search.lex(), correct_qls)
def test_with_raw_attr(self):
search = Search(['company'], config=SearchConfig)
search = search.match('Yandex').select(RawAttr('@weight*10', 'skey'))
self.assertEqual(
search.lex(),
"SELECT @weight*10 AS skey FROM company WHERE MATCH('Yandex')"
)
def json_search(request):
search_query = Search(indexes=['rt_main'], config=SphinxitConfig)
keyword = request.GET['keyword']
start = request.GET.get('start', 0)
count = request.GET.get('count', 10)
sort = request.GET.get('sort', '')
category = request.GET.get('category', '')
if request.GET.get('base64') == '1':
keyword = keyword.decode('base64').decode('utf8')
mckey = str(binascii.crc32((u'%s%s%s%s%s' % (keyword,start,count,sort,category)).encode('utf8')) & 0xFFFFFFFFL)
cache = mc.get(mckey)
if cache:
print 'bingo', keyword.encode('utf8'), mckey
return HttpResponse(cache)
q = search_query.match(keyword)
if category: q = q.filter(category__eq=binascii.crc32(category)&0xFFFFFFFFL)
if sort == 'create_time': q = q.order_by('create_time', 'desc')
if sort == 'length': q = q.order_by('length', 'desc')
q = q.limit(start, count)
q2 = search_query.match(keyword).select('category', Count()).group_by('category').named('cats')
res = q.ask(subqueries=[q2])
jsp = JsonResponse(res)
mc.set(mckey, jsp.content)
return jsp
def test_update_syntax(self):
search = Search(['company'], config=SearchConfig)
search = search.match('Yandex').update(products=(5,2)).filter(id__gt=1)
self.assertEqual(
search.lex(),
"UPDATE company SET products=(5,2) WHERE MATCH('Yandex') AND id>1"
)
def json_search(request):
search_query = Search(indexes=['rt_main'], config=SphinxitConfig)
keyword = request.GET['keyword']
start = request.GET.get('start', 0)
count = request.GET.get('count', 10)
sort = request.GET.get('sort', '')
category = request.GET.get('category', '')
if request.GET.get('base64') == '1':
keyword = keyword.decode('base64').decode('utf8')
mckey = str(
binascii.crc32(
(u'%s%s%s%s%s' %
(keyword, start, count, sort, category)).encode('utf8'))
& 0xFFFFFFFFL)
cache = mc.get(mckey)
if cache:
print 'bingo', keyword.encode('utf8'), mckey
return HttpResponse(cache)
q = search_query.match(keyword)
if category:
q = q.filter(category__eq=binascii.crc32(category) & 0xFFFFFFFFL)
if sort == 'create_time': q = q.order_by('create_time', 'desc')
if sort == 'length': q = q.order_by('length', 'desc')
q = q.limit(start, count)
q2 = search_query.match(keyword).select(
'category', Count()).group_by('category').named('cats')
res = q.ask(subqueries=[q2])
jsp = JsonResponse(res)
mc.set(mckey, jsp.content)
return jsp
def test_with_grouping(self):
search = Search(['company'], config=SearchConfig)
search = search.match('Yandex').select(Count()).group_by('date_created')
self.assertEqual(
search.lex(),
"SELECT COUNT(*) AS num FROM company WHERE MATCH('Yandex') GROUP BY date_created"
)
def test_simple(self):
search = Search(indexes=['company'], config=SearchConfig)
search = search.match('Yandex')
self.assertEqual(
search.lex(),
"SELECT * FROM company WHERE MATCH('Yandex')"
)
def test_with_params(self):
search = Search(['company'], config=SearchConfig)
search = search.match('Yandex').limit(0, 100).order_by('name', 'desc')
self.assertEqual(
search.lex(),
"SELECT * FROM company WHERE MATCH('Yandex') ORDER BY name DESC LIMIT 0,100"
)
def test_with_double_match(self):
search = Search(['company'], config=SearchConfig)
search = search.match('ОАО').match('ТНК')
self.assertEqual(
search.lex(),
"SELECT * FROM company WHERE MATCH('ОАО ТНК')"
)
def search():
term = flask.request.args.get('q')
query = Search(indexes=['sourcecode'], config=BaseSearchConfig)
query = query.match(term)
query = query.select('id', 'path')
results = query.ask()
results = [get_row(result) for result in results['result']['items']]
return uncached(json.dumps(results), mimetype='application/json')
def search(query):
search = Search(indexes=['events'], config=SearchConfig)
search = search.match(query)
# sphinxit uses set - an unordered data structure - for storing query order params
search = search.limit(0, 50).order_by('@weight DESC, event_date',
'DESC')
result = search.ask()
return result
def test_with_or_filters_and_fields(self):
correct_qls = [
"SELECT id, (id>=100 OR id=1) AS cnd FROM company WHERE MATCH('Yandex') AND cnd>0",
"SELECT id, (id=1 OR id>=0) AS cnd FROM company WHERE MATCH('Yandex') AND cnd>0",
]
search = Search(['company'], config=SearchConfig).select('id')
search = search.match('Yandex').filter(OR(id__gte=100, id__eq=1))
self.assertIn(search.lex(), correct_qls)
def test_with_select(self):
search = Search(indexes=['company'], config=SearchConfig)
search = search.select('id', 'date_created')
search = search.match('Yandex')
self.assertEqual(
search.lex(),
"SELECT id, date_created FROM company WHERE MATCH('Yandex')"
)
def test_with_time_filter(self):
search = Search(['company'], config=SearchConfig)
search = search.match('Yandex').filter(date_created__lte=datetime.date.today())
today = datetime.date.today()
sxql = (
"SELECT * FROM company WHERE MATCH('Yandex') "
"AND date_created<=%s" % unix_timestamp(today)
)
self.assertEqual(search.lex(), sxql)
def test_with_multiindex(self):
search = Search(
indexes=['company', 'company_delta'],
config=SearchConfig
)
search = search.filter(id__gte=100)
self.assertEqual(
search.lex(),
"SELECT * FROM company, company_delta WHERE id>=100"
)
def search(self, keyword, start=0, count=10, category=None, sort=None):
search_query = Search(indexes=['rt_main'], config=SphinxitConfig)
q = search_query.match(keyword)
if category: q = q.filter(category__eq=binascii.crc32(category)&0xFFFFFFFFL)
if sort == 'create_time': q = q.order_by('create_time', 'desc')
if sort == 'length': q = q.order_by('length', 'desc')
q = q.limit(start, count)
q2 = search_query.match(keyword).select('category', Count()).group_by('category').named('cats')
res = q.ask(subqueries=[q2])
return res
def search(query, eventType_id=None, contract_id=None, speciality_id=None):
search = Search(indexes=['event_service'], config=SearchConfig)
search = search.match(query)
if eventType_id:
search = search.filter(eventType_id__eq=int(eventType_id))
if contract_id:
search = search.filter(contract_id__eq=int(contract_id))
if speciality_id:
search = search.filter(speciality_id__in=[0, int(speciality_id)])
search = search.limit(0, 100)
result = search.ask()
return result
def json_search(request):
search_query = Search(indexes=['rt_main'], config=SphinxitConfig)
keyword = request.GET['keyword']
start = request.GET.get('start', 0)
count = request.GET.get('count', 10)
if request.GET.get('base64') == '1':
keyword = keyword.decode('base64').decode('utf8')
mckey = str(binascii.crc32((u'%s%s%s' % (keyword,start,count)).encode('utf8')) & 0xFFFFFFFFL)
cache = mc.get(mckey)
if cache:
print 'bingo', keyword.encode('utf8'), mckey
return HttpResponse(cache)
q = search_query.match(keyword).limit(start, count).order_by('create_time', 'desc')
res = q.ask()
jsp = JsonResponse(res)
mc.set(mckey, jsp.content)
return jsp
def get_context_data(self, **kwargs):
context = super(SearchView, self).get_context_data(**kwargs)
if self.request.GET.get('by') == 'resume' and self.request.GET.get('search'):
search = self.request.GET.get('search')
search_query = Search(indexes=['resume'], config=ResumeSearch)
search_query = search_query.match(search)
search_result = search_query.ask()
search_results = search_result['result']['items']
results = []
for result in search_results:
results.append(Resume.objects.get(id=result['id']))
context['search_results'] = results
context['search'] = self.request.GET.get('search')
context['url'] = 'resume'
elif self.request.GET.get('by') == 'job' and self.request.GET.get('search'):
search = self.request.GET.get('search')
search_query = Search(indexes=['jobs'], config=JobSearch)
search_query = search_query.match(search)
search_result = search_query.ask()
search_results = search_result['result']['items']
results = []
for result in search_results:
results.append(Job.objects.get(id=result['id']))
context['search_results'] = results
context['search'] = self.request.GET.get('search')
context['url'] = 'job'
return context
def json_search(request):
search_query = Search(indexes=['rt_main'], config=SphinxitConfig)
keyword = request.GET['keyword']
start = request.GET.get('start', 0)
count = request.GET.get('count', 10)
if request.GET.get('base64') == '1':
keyword = keyword.decode('base64').decode('utf8')
mckey = str(
binascii.crc32((u'%s%s%s' % (keyword, start, count)).encode('utf8'))
& 0xFFFFFFFFL)
cache = mc.get(mckey)
if cache:
print 'bingo', keyword.encode('utf8'), mckey
return HttpResponse(cache)
q = search_query.match(keyword).limit(start, count).order_by(
'create_time', 'desc')
res = q.ask()
jsp = JsonResponse(res)
mc.set(mckey, jsp.content)
return jsp
def search(name):
search = Search(indexes=['patient'], config=SearchConfig)
search = search.match(name).limit(0, 100)
search = search.options(field_weights={'code': 100,
'lastName': 90,
'birthDate_f1': 80,
'birthDate_f2': 80,
'firstName': 70,
'patrName': 60,
'SNILS': 50,
'document': 50,
'policy': 50})
#fixme: after sphinxit merge https://github.com/semirook/sphinxit/pull/20
search = search.order_by('@weight desc, lastName asc, firstName asc, patrName', 'asc')
result = search.ask()
return result
from sphinxit.core.processor import Search
from sphinxit.core.helpers import BaseSearchConfig
class SphinxitConfig(BaseSearchConfig):
WITH_STATUS = False
search_query = Search(indexes=['cigar'], config=SphinxitConfig)
search_query = search_query.match("cohiba")
# search_query.
rs = search_query.ask()
print(rs)
s = ''
for i in rs['result']['items']:
print(i['id'], ",")
s += str(i['id']) + ","
print(s)
def search(name):
search = Search(indexes=['person'], config=SearchConfig)
search = search.match(name).limit(0, 100)
result = search.ask()
return result
