def POST(self, userurlkey=None):
url=web.data()
if len(url) > config.MAX_LEN_URL:
web.ctx.status="400 Bad request"
return "url too long"
if userurlkey is not None:
print userurlkey
if len(userurlkey) < config.MIN_LEN_USERURLKEY:
web.ctx.status="400 Bad request"
return "key too short"
if len(userurlkey) > config.MAX_LEN_URLKEY:
web.ctx.status="400 Bad request"
return "key too long"
userurlkey = str.lower(utils.encode_string(userurlkey))
url=utils.encode_string(url)
if False == utils.check_url(url):
web.ctx.status="400 Bad request"
return "bad url"
ret, n_affected, urlkey = model.url_new(url, userurlkey)
if ret != 0:
return web.internalerror("db error")
retval = {
"is_created": (n_affected==1 or True and False),
"key": urlkey
}
web.ctx.status="200 OK"
return json.dumps(retval)
def object_create(self, res_type, obj_id, obj_dict):
obj_type = res_type.replace('-', '_')
obj_class = self._get_resource_class(obj_type)
# Gather column values for obj and updates to backrefs
# in a batch and write it at the end
bch = self._obj_uuid_cf.batch()
obj_cols = {}
obj_cols['fq_name'] = json.dumps(obj_dict['fq_name'])
obj_cols['type'] = json.dumps(obj_type)
if 'parent_type' in obj_dict:
# non config-root child
parent_type = obj_dict['parent_type']
parent_method_type = parent_type.replace('-', '_')
parent_fq_name = obj_dict['fq_name'][:-1]
obj_cols['parent_type'] = json.dumps(parent_type)
parent_uuid = self.fq_name_to_uuid(parent_method_type,
parent_fq_name)
self._create_child(bch, parent_method_type, parent_uuid, obj_type,
obj_id)
# Properties
for prop_field in obj_class.prop_fields:
field = obj_dict.get(prop_field)
if field is None:
continue
if prop_field == 'id_perms':
field['created'] = datetime.datetime.utcnow().isoformat()
field['last_modified'] = field['created']
self._create_prop(bch, obj_id, prop_field, field)
# References
# e.g. ref_field = 'network_ipam_refs'
# ref_type = 'network-ipam'
# ref_link_type = 'VnSubnetsType'
# is_weakref = False
for ref_field in obj_class.ref_fields:
ref_type, ref_link_type, _ = obj_class.ref_field_types[ref_field]
refs = obj_dict.get(ref_field, [])
for ref in refs:
ref_uuid = self.fq_name_to_uuid(ref_type, ref['to'])
ref_attr = ref.get('attr')
ref_data = {'attr': ref_attr, 'is_weakref': False}
self._create_ref(bch, obj_type, obj_id,
ref_type.replace('-', '_'), ref_uuid,
ref_data)
bch.insert(obj_id, obj_cols)
bch.send()
# Update fqname table
fq_name_str = ':'.join(obj_dict['fq_name'])
fq_name_cols = {
utils.encode_string(fq_name_str) + ':' + obj_id: json.dumps(None)
}
self._obj_fq_name_cf.insert(obj_type, fq_name_cols)
return (True, '')
def read_character_information(html):
reponse_character_data = {}
pattern = re.compile(r'Character Information')
character_data = html.find(text=pattern).parent.parent.parent.parent
for data in character_data:
text = encode_string(data.get_text())
two_points_index = text.find(':')
reponse_character_data[camelize(
text[:two_points_index])] = text[two_points_index + 1:]
return reponse_character_data
def fq_name_to_uuid(self, obj_type, fq_name):
method_name = obj_type.replace("-", "_")
fq_name_str = ":".join(fq_name)
col_start = "%s:" % (utils.encode_string(fq_name_str))
col_fin = "%s;" % (utils.encode_string(fq_name_str))
try:
col_info_iter = self._obj_fq_name_cf.xget(method_name, column_start=col_start, column_finish=col_fin)
except pycassa.NotFoundException:
raise NoIdError("%s %s" % (obj_type, fq_name))
col_infos = list(col_info_iter)
if len(col_infos) == 0:
raise NoIdError("%s %s" % (obj_type, fq_name))
for (col_name, col_val) in col_infos:
obj_uuid = col_name.split(":")[-1]
return obj_uuid
def _object_create(self, res_type, obj_ids, obj_dict):
obj_type = res_type.replace('-', '_')
obj_class = self._get_resource_class(obj_type)
# Gather column values for obj and updates to backrefs
# in a batch and write it at the end
bch = self._obj_uuid_cf.batch()
obj_cols = {}
obj_cols['fq_name'] = json.dumps(obj_dict['fq_name'])
obj_cols['type'] = json.dumps(obj_type)
if 'parent_type' in obj_dict:
# non config-root child
parent_type = obj_dict['parent_type']
parent_method_type = parent_type.replace('-', '_')
parent_fq_name = obj_dict['fq_name'][:-1]
obj_cols['parent_type'] = json.dumps(parent_type)
parent_uuid = self.fq_name_to_uuid(parent_method_type, parent_fq_name)
self._create_child(bch, parent_method_type, parent_uuid, obj_type, obj_ids['uuid'])
# Properties
for prop_field in obj_class.prop_fields:
field = obj_dict.get(prop_field)
if field is None:
continue
if prop_field == 'id_perms':
field['created'] = datetime.datetime.utcnow().isoformat()
field['last_modified'] = field['created']
self._create_prop(bch, obj_ids['uuid'], prop_field, field)
# References
# e.g. ref_field = 'network_ipam_refs'
# ref_type = 'network-ipam'
# ref_link_type = 'VnSubnetsType'
# is_weakref = False
for ref_field in obj_class.ref_fields:
ref_type, ref_link_type, _ = obj_class.ref_field_types[ref_field]
refs = obj_dict.get(ref_field, [])
for ref in refs:
ref_uuid = self.fq_name_to_uuid(ref_type, ref['to'])
ref_attr = ref.get('attr')
ref_data = {'attr': ref_attr, 'is_weakref': False}
self._create_ref(bch, obj_type, obj_ids['uuid'],
ref_type.replace('-', '_'), ref_uuid, ref_data)
bch.insert(obj_ids['uuid'], obj_cols)
bch.send()
# Update fqname table
fq_name_str = ':'.join(obj_dict['fq_name'])
fq_name_cols = {utils.encode_string(fq_name_str) + ':' + obj_ids['uuid']: json.dumps(None)}
self._obj_fq_name_cf.insert(obj_type, fq_name_cols)
return (True, '')
def fq_name_to_uuid(self, obj_type, fq_name):
method_name = obj_type.replace('-', '_')
fq_name_str = ':'.join(fq_name)
col_start = '%s:' % (utils.encode_string(fq_name_str))
col_fin = '%s;' % (utils.encode_string(fq_name_str))
try:
col_info_iter = self._obj_fq_name_cf.xget(
method_name, column_start=col_start, column_finish=col_fin)
except pycassa.NotFoundException:
raise NoIdError('%s %s' % (obj_type, fq_name))
col_infos = list(col_info_iter)
if len(col_infos) == 0:
raise NoIdError('%s %s' % (obj_type, fq_name))
for (col_name, col_val) in col_infos:
obj_uuid = col_name.split(':')[-1]
return obj_uuid
def POST(self, urlkey):
url=web.data()
if len(url) > config.MAX_LEN_URL:
web.ctx.status="400 Bad request"
return "url too long"
if urlkey is not None:
if len(urlkey) > config.MAX_LEN_URLKEY:
web.ctx.status="400 Bad request"
return "key too long"
urlkey = str.lower(utils.encode_string(urlkey))
url=utils.encode_string(url)
if False == utils.check_url(url):
web.ctx.status="400 Bad request"
return "bad url"
ret, n_affected = model.url_modify(urlkey, url)
if ret != 0:
return web.internalerror("db error")
ret_val={ 'n_affected': n_affected}
return json.dumps(ret_val)
def read_character_death_information(html):
response_character_data = []
pattern = re.compile(r'Character Deaths')
character_data = html.find(text=pattern).parent.parent.parent.parent
for data in character_data:
text = encode_string(data.get_text())
cet_index = text.find('CET')
if cet_index > 1:
response_character_data.append({
'date':
text[:cet_index + 3],
'killed_by_message':
text[cet_index + 3:]
})
return response_character_data
def object_delete(self, res_type, obj_uuid):
obj_type = res_type.replace('-', '_')
obj_class = self._get_resource_class(obj_type)
obj_uuid_cf = self._obj_uuid_cf
try:
fq_name = json.loads(
obj_uuid_cf.get(obj_uuid, columns=['fq_name'])['fq_name'])
except pycassa.NotFoundException:
raise NoIdError(obj_uuid)
bch = obj_uuid_cf.batch()
# unlink from parent
col_start = 'parent:'
col_fin = 'parent;'
col_name_iter = obj_uuid_cf.xget(
obj_uuid, column_start=col_start, column_finish=col_fin)
for (col_name, col_val) in col_name_iter:
(_, parent_type, parent_uuid) = col_name.split(':')
self._delete_child(
bch, parent_type, parent_uuid, obj_type, obj_uuid)
# remove refs
col_start = 'ref:'
col_fin = 'ref;'
col_name_iter = obj_uuid_cf.xget(
obj_uuid, column_start=col_start, column_finish=col_fin)
for (col_name, col_val) in col_name_iter:
(_, ref_type, ref_uuid) = col_name.split(':')
self._delete_ref(bch, obj_type, obj_uuid, ref_type, ref_uuid)
# remove link from relaxed back refs
col_start = 'relaxbackref:'
col_fin = 'relaxbackref;'
col_name_iter = obj_uuid_cf.xget(
obj_uuid, column_start=col_start, column_finish=col_fin)
for (col_name, col_val) in col_name_iter:
(_, backref_uuid) = col_name.split(':')
self._delete_ref(bch, None, backref_uuid, obj_type, obj_uuid)
bch.remove(obj_uuid)
bch.send()
# Update fqname table
fq_name_str = ':'.join(fq_name)
fq_name_col = utils.encode_string(fq_name_str) + ':' + obj_uuid
self._obj_fq_name_cf.remove(obj_type, columns = [fq_name_col])
return (True, '')
def read_online_players(html):
response_character_online = []
reference_content = html.findAll("tr", {"class": "LabelH"})
online_players_table = reference_content[0].parent.find_all('tr')
for playerRow in online_players_table:
character_data_dict = {}
count = 0
cols = playerRow.find_all('td')
for characterData in cols:
text = encode_string(characterData.get_text())
if count == 0:
character_data_dict['name'] = text
elif count == 1:
character_data_dict['level'] = text
elif count == 2:
character_data_dict['vocation'] = text
response_character_online.append(character_data_dict)
count += 1
return response_character_online
def object_delete(self, res_type, obj_uuid):
obj_type = res_type.replace('-', '_')
obj_class = self._get_resource_class(obj_type)
obj_uuid_cf = self._obj_uuid_cf
try:
fq_name = json.loads(
obj_uuid_cf.get(obj_uuid, columns=['fq_name'])['fq_name'])
except pycassa.NotFoundException:
raise NoIdError(obj_uuid)
bch = obj_uuid_cf.batch()
# unlink from parent
col_start = 'parent:'
col_fin = 'parent;'
col_name_iter = obj_uuid_cf.xget(obj_uuid,
column_start=col_start,
column_finish=col_fin)
for (col_name, col_val) in col_name_iter:
(_, parent_type, parent_uuid) = col_name.split(':')
self._delete_child(bch, parent_type, parent_uuid, obj_type,
obj_uuid)
# remove refs
col_start = 'ref:'
col_fin = 'ref;'
col_name_iter = obj_uuid_cf.xget(obj_uuid,
column_start=col_start,
column_finish=col_fin)
for (col_name, col_val) in col_name_iter:
(_, ref_type, ref_uuid) = col_name.split(':')
self._delete_ref(bch, obj_type, obj_uuid, ref_type, ref_uuid)
bch.remove(obj_uuid)
bch.send()
# Update fqname table
fq_name_str = ':'.join(fq_name)
fq_name_col = utils.encode_string(fq_name_str) + ':' + obj_uuid
self._obj_fq_name_cf.remove(obj_type, columns=[fq_name_col])
return (True, '')
def object_create(self, res_type, obj_id, obj_dict):
obj_type = res_type.replace('-', '_')
obj_class = self._get_resource_class(obj_type)
# Gather column values for obj and updates to backrefs
# in a batch and write it at the end
bch = self._obj_uuid_cf.batch()
obj_cols = {}
obj_cols['fq_name'] = json.dumps(obj_dict['fq_name'])
obj_cols['type'] = json.dumps(obj_type)
if 'parent_type' in obj_dict:
# non config-root child
parent_type = obj_dict['parent_type']
parent_method_type = parent_type.replace('-', '_')
parent_fq_name = obj_dict['fq_name'][:-1]
obj_cols['parent_type'] = json.dumps(parent_type)
parent_uuid = self.fq_name_to_uuid(parent_method_type, parent_fq_name)
self._create_child(bch, parent_method_type, parent_uuid, obj_type, obj_id)
# Properties
for prop_field in obj_class.prop_fields:
field = obj_dict.get(prop_field)
# Specifically checking for None
if field is None:
continue
if prop_field == 'id_perms':
field['created'] = datetime.datetime.utcnow().isoformat()
field['last_modified'] = field['created']
if prop_field in obj_class.prop_list_fields:
# store list elements in list order
# iterate on wrapped element or directly or prop field
if obj_class.prop_list_field_has_wrappers[prop_field]:
wrapper_field = field.keys()[0]
list_coll = field[wrapper_field]
else:
list_coll = field
for i in range(len(list_coll)):
self._add_to_prop_list(
bch, obj_id, prop_field, list_coll[i], str(i))
elif prop_field in obj_class.prop_map_fields:
# iterate on wrapped element or directly or prop field
if obj_class.prop_map_field_has_wrappers[prop_field]:
wrapper_field = field.keys()[0]
map_coll = field[wrapper_field]
else:
map_coll = field
map_key_name = obj_class.prop_map_field_key_names[prop_field]
for map_elem in map_coll:
map_key = map_elem[map_key_name]
self._set_in_prop_map(
bch, obj_id, prop_field, map_elem, map_key)
else:
self._create_prop(bch, obj_id, prop_field, field)
# References
# e.g. ref_field = 'network_ipam_refs'
# ref_type = 'network-ipam'
# ref_link_type = 'VnSubnetsType'
# is_weakref = False
for ref_field in obj_class.ref_fields:
ref_type, ref_link_type, _ = obj_class.ref_field_types[ref_field]
refs = obj_dict.get(ref_field, [])
for ref in refs:
ref_uuid = self.fq_name_to_uuid(ref_type, ref['to'])
ref_attr = ref.get('attr')
ref_data = {'attr': ref_attr, 'is_weakref': False}
self._create_ref(bch, obj_type, obj_id,
ref_type.replace('-', '_'), ref_uuid, ref_data)
bch.insert(obj_id, obj_cols)
bch.send()
# Update fqname table
fq_name_str = ':'.join(obj_dict['fq_name'])
fq_name_cols = {utils.encode_string(fq_name_str) + ':' + obj_id:
json.dumps(None)}
self._obj_fq_name_cf.insert(obj_type, fq_name_cols)
return (True, '')
for i, row in enumerate(x):
row = list(row)
target = row.pop()
d = {4: 0, 5: 0, 6: 0}
for key, t_row in t_rows.items():
if i not in t_row:
d[predict(roots[key], row, t_columns[key])] += 1
if max(d.values()) != 0:
total += 1
if max(d.items(), key=operator.itemgetter(1))[0] == target:
cnt += 1
print("accuracy ", cnt / total)
"""
Predictions are made on the testing set using all the trees
"""
f = open('prediction.csv', 'w')
f.write("id,class\n")
y = load_data('testing.csv')
for row in y:
idx = row[0]
seq = encode_string(row[1])
d = {4: 0, 5: 0, 6: 0}
for i, root in enumerate(roots):
d[predict(root, seq, t_columns[i])] += 1
prediction = decode(max(d.items(), key=operator.itemgetter(1))[0])
f.write(str(idx) + "," + prediction + "\n")
f.close()
def read_guild_information(html):
guild_data = {}
guild_information = html.find('div', {'id': 'GuildInformationContainer'})
guild_data['guild_information'] = encode_string(
guild_information.get_text())
return guild_data
ax2.set_xlabel("Word Offset")
plt.savefig('key_words_fig.pdf', format='pdf')
if __name__ == '__main__':
if len(sys.argv) > 1:
text = utils.get_text(sys.argv[1])
if len(sys.argv) > 2:
print summarize(text, int(sys.argv[2]))
else:
print summarize(text)
while True:
is_drawing = raw_input("Want to draw key words dispersion plot?[y/n]:")
if is_drawing == "y" or is_drawing == "Y":
keywords_ranking = word_ranking(text).most_common(15)
words, scores = list(zip(*keywords_ranking))
x, y = dispersion(text, words)
words = [utils.encode_string(w) for w in words]
draw(words, scores, x, y)
break
elif is_drawing == "n" or is_drawing == "N":
break
else:
print("Incorrect command.")
sys.exit(0)
else:
print('There is no text to summarize')
sys.exit(1)
