def _select(self, params):
# encode the query as utf-8 so urlencode can handle it
params['q'] = unicode_safe(params['q'])
path = '%s/select/?%s' % (self.path, urlencode(params))
conn = HTTPConnection(self.host, self.port)
conn.request('GET', path)
return conn.getresponse()
def _select(self, params):
# encode the query as utf-8 so urlencode can handle it
params['q'] = unicode_safe(params['q'])
path = '%s/select/?%s' % (self.path, urlencode(params))
conn = HTTPConnection(self.host, self.port)
conn.request('GET', path)
return conn.getresponse()
def fetch_article(article,titlefield,linkfield,niceify=False):
if isinstance(article,dict):
keys=article.keys()
if titlefield in keys and linkfield in keys:
title=unicode_safe(validate_title(article[titlefield]))
link=validate_link(article[linkfield])
if niceify:
title=niceify_title(title,niceify)
print "Niced title: %s" % title
if title and link:
return dict(title=title,link=link)
return None
def _from_python(value):
"""
Converts python values to a form suitable for insertion into the xml
we send to solr.
"""
if isinstance(value, datetime):
value = value.strftime('%Y-%m-%dT%H:%M:%S.000Z')
elif isinstance(value, date):
value = value.strftime('%Y-%m-%dT00:00:00.000Z')
elif isinstance(value, bool):
if value:
value = 'true'
else:
value = 'false'
else:
value = unicode_safe(value)
return value
def _from_python(value):
"""
Converts python values to a form suitable for insertion into the xml
we send to solr.
"""
if isinstance(value, datetime):
value = value.strftime('%Y-%m-%dT%H:%M:%S.000Z')
elif isinstance(value, date):
value = value.strftime('%Y-%m-%dT00:00:00.000Z')
elif isinstance(value, bool):
if value:
value = 'true'
else:
value = 'false'
else:
value = unicode_safe(value)
return value
def changed(types=None,since=None,commit=True,optimize=False):
"""
Run by `cron` (through `paster run`) on a schedule to update
all Things that have been created or have changed since the
last run. Things add themselves to a `thing_changes` table,
which we read, find the Things, tokenise, and re-submit them
to Solr
"""
global indexed_types
set_emptying_cache()
start_t = datetime.now()
if not types:
types = indexed_types
if not since:
since = get_last_run()
all_changed = []
for cls in types:
changed = set(x[0]
for x in thing_changes.get_changed(cls,min_date = since))
# changed =:= [(Fullname,Date) | ...]
changed = cls._by_fullname(changed,
data=True, return_dict=False)
changed = [x for x in changed if not x._spam and not x._deleted]
# note: anything marked as spam or deleted is not updated in
# the search database. Since these are filtered out in the UI,
# that's probably fine.
if len(changed) > 0:
changed = tokenize_things(changed)
print "Found %d %ss starting with %s" % (len(changed),cls.__name__,unicode_safe(changed[0]['contents']))
all_changed += changed
else:
print "No changed %ss detected" % (cls.__name__,)
with SolrConnection(commit=commit,optimize=optimize) as s:
s.add(all_changed)
save_last_run(start_t)
Thing:
(Field('fullname',
'_fullname'), Field('date', '_date', is_date=True, reverse=True),
Field('lang'), Field('ups', '_ups', is_number=True, reverse=True),
Field('downs', '_downs', is_number=True,
reverse=True), Field('spam', '_spam'), Field('deleted', '_deleted'),
Field('hot', lambda t: t._hot * 1000, is_number=True, reverse=True),
Field('controversy', '_controversy', is_number=True, reverse=True),
Field('points',
lambda t: (t._ups - t._downs),
is_number=True,
reverse=True)),
Subreddit: (
Field('contents',
lambda s: ' '.join([
unicode_safe(s.name),
unicode_safe(s.title),
unicode_safe(s.description),
unicode_safe(s.firsttext)
]),
tokenize=True),
Field('boost', '_downs'),
#Field('title'),
#Field('firsttext'),
#Field('description'),
#Field('over_18'),
#Field('sr_type','type'),
),
Link: (
Field('contents', 'title', tokenize=True),
Field(
def str_to_python(self, value):
"""
Convert an 'str' field from solr's xml format to python and return it.
"""
return unicode_safe(value)
# discussion of multi-language search. The 'boost' field is a
# solr-magic field that ends up being an attribute on the <doc>
# message (rather than a field), and is used to do an index-time boost
# (this magic is done in pysolr.dor_to_elemtree)
search_fields={Thing: (Field('fullname', '_fullname'),
Field('date', '_date', is_date = True, reverse=True),
Field('lang'),
Field('ups', '_ups', is_number=True, reverse=True),
Field('downs', '_downs', is_number=True, reverse=True),
Field('spam','_spam'),
Field('deleted','_deleted'),
Field('hot', lambda t: t._hot*1000, is_number=True, reverse=True),
Field('controversy', '_controversy', is_number=True, reverse=True),
Field('points', lambda t: (t._ups - t._downs), is_number=True, reverse=True)),
Subreddit: (Field('contents',
lambda s: ' '.join([unicode_safe(s.name),
unicode_safe(s.title),
unicode_safe(s.description),
unicode_safe(s.firsttext)]),
tokenize = True),
Field('boost', '_downs'),
#Field('title'),
#Field('firsttext'),
#Field('description'),
#Field('over_18'),
#Field('sr_type','type'),
),
Link: (Field('contents','title', tokenize = True),
Field('boost', lambda t: int(t._hot*1000),
# yes, it's a copy of 'hot'
is_number=True, reverse=True),
return ("<ThingField: (%s,%s,%s,%s)>"
% (self.name,self.cls,self.id_attr,self.lu_attr_name))
search_fields={Thing: (Field('fullname', '_fullname'),
Field('date', '_date', is_date = True, reverse=True),
Field('lang'),
Field('ups', '_ups', is_number=True, reverse=True),
Field('downs', '_downs', is_number=True, reverse=True),
Field('spam','_spam'),
Field('deleted','_deleted'),
Field('hot', lambda t: t._hot*1000, is_number=True, reverse=True),
Field('controversy', '_controversy', is_number=True, reverse=True),
Field('points', lambda t: (t._ups - t._downs), is_number=True, reverse=True)),
Subreddit: (Field('contents',
lambda s: ' '.join([unicode_safe(s.name),
unicode_safe(s.title),
unicode_safe(s.description),
unicode_safe(s.firsttext)]),
tokenize = True),
Field('boost', '_downs'),
#Field('title'),
#Field('firsttext'),
#Field('description'),
#Field('over_18'),
#Field('sr_type','type'),
),
Link: (Field('contents','title', tokenize = True),
Field('boost', lambda t: int(t._hot*1000),
# yes, it's a copy of 'hot'
is_number=True, reverse=True),
def str_to_python(self, value):
"""
Convert an 'str' field from solr's xml format to python and return it.
"""
return unicode_safe(value)
