def setUp(self):
self.parser = l2cs.make_parser(int_fields=["count", "number"],
yesno_fields=["active", "ready"],
aliases={"alias": ["alias1", "alias2"]})
self.schema = l2cs.make_schema(["foo", "bar", "baz", "count", "number",
"active", "text", "ready", "active",
"alias", "alias1", "alias2"],
["timestamp", "date"])
self.schema_parser = l2cs.make_parser(int_fields=["count", "number"],
yesno_fields=["active", "ready"],
aliases={"alias": ["alias1",
"alias2"]},
schema=self.schema)
def setUp(self):
self.parser = l2cs.make_parser(int_fields=["count", "number"],
yesno_fields=["active", "ready"],
aliases={"alias": ["alias1", "alias2"]})
self.schema = l2cs.make_schema([
"foo", "bar", "baz", "count", "number", "active", "text", "ready",
"active", "alias", "alias1", "alias2"
], ["timestamp", "date"])
self.schema_parser = l2cs.make_parser(
int_fields=["count", "number"],
yesno_fields=["active", "ready"],
aliases={"alias": ["alias1", "alias2"]},
schema=self.schema)
class LinkSearchQuery(CloudSearchQuery):
search_api = g.CLOUDSEARCH_SEARCH_API
sorts = {
'relevance': '-relevance',
'hot': '-hot2',
'top': '-top',
'new': '-timestamp',
'comments': '-num_comments',
}
sorts_menu_mapping = {
'relevance': 1,
'hot': 2,
'new': 3,
'top': 4,
'comments': 5,
}
recents = {
'hour': timedelta(hours=1),
'day': timedelta(days=1),
'week': timedelta(days=7),
'month': timedelta(days=31),
'year': timedelta(days=366),
'all': None,
None: None,
}
schema = l2cs.make_schema(LinkFields.lucene_fieldnames())
lucene_parser = l2cs.make_parser(
int_fields=LinkFields.lucene_fieldnames(type_=int),
yesno_fields=LinkFields.lucene_fieldnames(type_="yesno"),
schema=schema)
known_syntaxes = ("cloudsearch", "lucene", "plain")
default_syntax = "lucene"
def customize_query(self, bq=u''):
queries = []
if bq:
queries = [bq]
subreddit_query = self._get_sr_restriction(self.sr)
if subreddit_query:
queries.append(subreddit_query)
if self.recent:
recent_query = self._restrict_recent(self.recent)
queries.append(recent_query)
if not self.include_over18:
queries.append('over18:0')
return self.create_boolean_query(queries)
@staticmethod
def _restrict_recent(recent):
now = datetime.now(g.tz)
since = epoch_seconds(now - recent)
return 'timestamp:%i..' % since
@staticmethod
def _get_sr_restriction(sr):
'''Return a cloudsearch appropriate query string that restricts
results to only contain results from self.sr
'''
bq = []
if (not sr) or sr == All or isinstance(sr, DefaultSR):
return None
elif isinstance(sr, MultiReddit):
bq = ["(or"]
for sr_id in sr.sr_ids:
bq.append("sr_id:%s" % sr_id)
bq.append(")")
elif isinstance(sr, DomainSR):
bq = ["site:'%s'" % sr.domain]
elif sr == Friends:
if not c.user_is_loggedin or not c.user.friends:
return None
bq = ["(or"]
# The query limit is roughly 8k bytes. Limit to 200 friends to
# avoid getting too close to that limit
friend_ids = c.user.friends[:200]
friends = [
"author_fullname:'%s'" %
Account._fullname_from_id36(r2utils.to36(id_))
for id_ in friend_ids
]
bq.extend(friends)
bq.append(")")
elif isinstance(sr, ModContribSR):
bq = ["(or"]
for sr_id in sr.sr_ids:
bq.append("sr_id:%s" % sr_id)
bq.append(")")
elif not isinstance(sr, FakeSubreddit):
bq = ["sr_id:%s" % sr._id]
return ' '.join(bq)
class LinkSearchQuery(CloudSearchQuery):
search_api = g.CLOUDSEARCH_SEARCH_API
sorts = {
'relevance': '-relevance',
'relevance2': '-relevance2',
'hot': '-hot2',
'top': '-top',
'new': '-timestamp',
'comments': '-num_comments',
}
recents = {
'hour': timedelta(hours=1),
'day': timedelta(days=1),
'week': timedelta(days=7),
'month': timedelta(days=31),
'year': timedelta(days=366),
'all': None,
None: None,
}
schema = l2cs.make_schema(LinkFields.lucene_fieldnames())
lucene_parser = l2cs.make_parser(
int_fields=LinkFields.lucene_fieldnames(type_=int),
yesno_fields=LinkFields.lucene_fieldnames(type_="yesno"),
schema=schema)
known_syntaxes = g.search_syntaxes
default_syntax = "lucene"
def customize_query(self, bq=u''):
queries = []
if bq:
queries = [bq]
if self.sr:
subverbify_query = self._restrict_sr(self.sr)
if subverbify_query:
queries.append(subverbify_query)
if self.recent:
recent_query = self._restrict_recent(self.recent)
queries.append(recent_query)
if not self.include_over18:
queries.append('over18:0')
return self.create_boolean_query(queries)
@staticmethod
def _restrict_recent(recent):
now = datetime.now(g.tz)
since = epoch_seconds(now - recent)
return 'timestamp:%i..' % since
@staticmethod
def _restrict_sr(sr):
'''Return a cloudsearch appropriate query string that restricts
results to only contain results from sr
'''
if isinstance(sr, MultiVerbify):
if not sr.sr_ids:
raise InvalidQuery
srs = ["sr_id:%s" % sr_id for sr_id in sr.sr_ids]
return "(or %s)" % ' '.join(srs)
elif isinstance(sr, DomainSR):
return "site:'\"%s\"'" % sr.domain
elif isinstance(sr, FriendsSR):
if not c.user_is_loggedin or not c.user.friends:
raise InvalidQuery
# The query limit is roughly 8k bytes. Limit to 200 friends to
# avoid getting too close to that limit
friend_ids = c.user.friends[:200]
friends = [
"author_fullname:'%s'" %
Account._fullname_from_id36(v1utils.to36(id_))
for id_ in friend_ids
]
return "(or %s)" % ' '.join(friends)
elif isinstance(sr, AllMinus):
if not sr.exclude_sr_ids:
raise InvalidQuery
exclude_srs = ["sr_id:%s" % sr_id for sr_id in sr.exclude_sr_ids]
return "(not (or %s))" % ' '.join(exclude_srs)
elif not isinstance(sr, FakeSubverbify):
return "sr_id:%s" % sr._id
return None
class LinkSearchQuery(CloudSearchQuery):
search_api = g.CLOUDSEARCH_SEARCH_API
sorts = {
'relevance': '-relevance',
'hot': '-hot2',
'top': '-top',
'new': '-timestamp',
'comments': '-num_comments',
}
sorts_menu_mapping = {
'relevance': 1,
'hot': 2,
'new': 3,
'top': 4,
'comments': 5,
}
schema = l2cs.make_schema(LinkFields.lucene_fieldnames())
lucene_parser = l2cs.make_parser(
int_fields=LinkFields.lucene_fieldnames(type_=int),
yesno_fields=LinkFields.lucene_fieldnames(type_="yesno"),
schema=schema)
known_syntaxes = ("cloudsearch", "lucene", "plain")
default_syntax = "lucene"
def customize_query(self, bq):
subreddit_query = self._get_sr_restriction(self.sr)
return self.create_boolean_query(bq, subreddit_query)
@classmethod
def create_boolean_query(cls, query, subreddit_query):
'''Join a (user-entered) text query with the generated subreddit query
Input:
base_query: user input from the search textbox
subreddit_query: output from _get_sr_restriction(sr)
Test cases:
base_query: simple, simple with quotes, boolean, boolean w/ parens
subreddit_query: None, in parens '(or sr_id:1 sr_id:2 ...)',
without parens "author:'foo'"
'''
if subreddit_query:
bq = "(and %s %s)" % (query, subreddit_query)
else:
bq = query
return bq
@staticmethod
def _get_sr_restriction(sr):
'''Return a cloudsearch appropriate query string that restricts
results to only contain results from self.sr
'''
bq = []
if (not sr) or sr == All or isinstance(sr, DefaultSR):
return None
elif isinstance(sr, MultiReddit):
bq = ["(or"]
for sr_id in sr.sr_ids:
bq.append("sr_id:%s" % sr_id)
bq.append(")")
elif isinstance(sr, DomainSR):
bq = ["site:'%s'" % sr.domain]
elif sr == Friends:
if not c.user_is_loggedin or not c.user.friends:
return None
bq = ["(or"]
# The query limit is roughly 8k bytes. Limit to 200 friends to
# avoid getting too close to that limit
friend_ids = c.user.friends[:200]
friends = [
"author_fullname:'%s'" %
Account._fullname_from_id36(r2utils.to36(id_))
for id_ in friend_ids
]
bq.extend(friends)
bq.append(")")
elif isinstance(sr, ModContribSR):
bq = ["(or"]
for sr_id in sr.sr_ids:
bq.append("sr_id:%s" % sr_id)
bq.append(")")
elif not isinstance(sr, FakeSubreddit):
bq = ["sr_id:%s" % sr._id]
return ' '.join(bq)
def setUp(self):
self.parser = l2cs.make_parser(int_fields=["count", "number"],
yesno_fields=["active", "ready"],
aliases={"alias": ["alias1", "alias2"]})