def add_node(e1, e2, rel, c1, c2):
message = "添加成功"
a = [e1, e2, rel, c1, c2]
num1 = graph.run("MATCH (m:Concept {name:'%s' })return count(m)" %
e1).data()[0]['count(m)']
num2 = graph.run("MATCH (m:Concept {name:'%s' })return count(m)" %
e2).data()[0]['count(m)']
if num1 == 0 and num2 == 0:
graph.run(
"CREATE(n:Concept {name:'%s', cate:'%s'})-[r:%s{relation: '%s'}]->(m:Concept {name:'%s', cate:'%s'})"
% (e1, c1, rel, rel, e2, c2))
if num1 != 0 and num2 == 0:
graph.run(
"MATCH(m: Concept {name: '%s', cate:'%s'}) CREATE(m) - [r:%s{relation: '%s'}]->(n:Concept {name:'%s', cate:'%s'})"
% (e1, c1, rel, rel, e2, c2))
if num1 == 0 and num2 != 0:
graph.run(
"MATCH(m: Concept {name: '%s', cate:'%s'}) CREATE(n:Concept {name:'%s', cate:'%s'})-[r:%s{relation: '%s'}]->(m)"
% (e2, c2, e1, c1, rel, rel))
# num = data
if num1 != 0 and num2 != 0:
graph.run(
"MATCH(m: Concept {name: '%s'}),(n:Concept {name:'%s'}) CREATE (m)-[r:%s{relation: '%s'}]->(n)"
% (e2, e1, rel, rel))
# message = "此条关系已存在！"
print(num1, num2, message)
# MATCH(m: Concept {name: 'C语言'}) CREATE(m) - [: 包括]->(n:Concept {name:'函数'}) return m
# graph.run("CREATE(n:Concept {name:'%s', cate:'%s'}) return n") %(e1,e2)
print(a)
return message
def query_all():
nodes = map(buildNodes,
graph.run('MATCH (p) RETURN p.Name, p.Clan').data())
edges = map(
buildEdges,
graph.run(
'MATCH (p)-[r]->(n) RETURN p.Name, r.relation,n.Name').data())
return nodes, edges
def query_paper_info(id):
json_data = {'data': [], "links": []}
data_item1 = {}
data_item1['name'] = id
data_item1['category'] = "AUTHOR"
json_data['data'].append(data_item1)
count = 0
name_dict = {}
name_dict[data_item1['name']] = count
count += 1
data1 = graph.run("match(p:PAPER{paperID:'%s'} )-[r]->(n:PAPER) return p, r, n" % (id))
data1 = list(data1)
for i in data1:
data_item2 = {}
data_item2['name'] = i['n']['paperID']
data_item2['category'] = "PAPER"
name_dict[data_item2['name']] = count
count += 1
link_item = {}
link_item['source'] = name_dict[i['p']['paperID']]
link_item['target'] = name_dict[i['n']['paperID']]
link_item['value'] = 'CITATION'
json_data['data'].append(data_item2)
json_data['links'].append(link_item)
data2 = graph.run("match(p:AUTHOR )-[r:AUTHOR2PAPER]->(n:PAPER{paperID:'%s'}) return p, r, n"%(id))
data2 = list(data2)
for i in data2:
data_item2 = {}
data_item2['name'] = i['p']['authorName']
data_item2['category'] = "AUTHOR"
name_dict[data_item2['name']] = count
count += 1
link_item = {}
link_item['source'] = name_dict[i['p']['authorName']]
link_item['target'] = name_dict[i['n']['paperID']]
link_item['value'] = 'AUTHOR2PAPER'
json_data['data'].append(data_item2)
json_data['links'].append(link_item)
data3 = graph.run(
"match(p:PAPER)-[r]->(n:PAPER{paperID:'%s'} ) return p, r, n " % (id))
data3 = list(data3)
for i in data3:
data_item2 = {}
data_item2['name'] = i['p']['paperID']
data_item2['category'] = "PAPER"
name_dict[data_item2['name']] = count
count += 1
link_item = {}
link_item['source'] = name_dict[i['p']['paperID']]
link_item['target'] = name_dict[i['n']['paperID']]
link_item['value'] = 'CITATION'
json_data['data'].append(data_item2)
json_data['links'].append(link_item)
return json_data
def get_data_num():
data1 = graph.run("match (n) return count(*)").data()
entity = data1[0]['count(*)']
data2 = graph.run("MATCH (n)-[r]->() RETURN COUNT(r)").data()
relation = data2[0]['COUNT(r)']
# getpath = os.path.abspath(os.path.dirname(os.path.dirname(__file__)))
# getpath = ('/').join(getpath.split('\\'))
#
# with open(getpath + '/raw_data/relation.csv', 'r') as f:
# relation = len(f.readlines())
# with open(getpath + '/KGQA/my_dict.txt', 'r', encoding='utf-8') as f:
# entity = len(f.readlines())
print(relation, entity)
return relation, entity
def query_branch(name, deep):
if deep == '二度查询':
graph.run(
'CALL apoc.export.json.query("match data=(n1{name:\'%s\'})-[r1]->(n2) return n1 {.*}, r1 {.*}, n2 {.*}", '
'"knows-with-node-properties.json", {writeNodeProperties:true})' %
(name))
elif deep == '三度查询':
graph.run(
'CALL apoc.export.json.query("match data=(n1{name:\'%s\'})-[r1]->(n2)-[r2]->(n3) return n1 {.*}, r1 {.*}, n2 {.*}, r2 {.*}, n3 {.*}", '
'"knows-with-node-properties.json", {writeNodeProperties:true})' %
(name))
elif deep == '四度查询':
graph.run(
'CALL apoc.export.json.query("match data=(n1{name:\'%s\'})-[r1]->(n2)-[r2]->(n3)-[r3]->(n4) return n1 {.*}, r1 {.*}, n2 {.*}, r2 {.*}, n3 {.*}, r3 {.*}, n4 {.*}", '
'"knows-with-node-properties.json", {writeNodeProperties:true})' %
(name))
else:
graph.run(
'CALL apoc.export.json.query("match data=(n1{name:\'%s\'})-[r1]->(n2)-[r2]->(n3) return n1 {.*}, r1 {.*}, n2 {.*}, r2 {.*}, n3 {.*}", '
'"knows-with-node-properties.json", {writeNodeProperties:true})' %
(name))
# print(data)
json_data = {'data': [], "links": []} # echarts关系图所需要的数据格式
d = []
with open(
"D:\\Program Files (x86)\\.Neo4jDesktop\\neo4jDatabases\\database-3c0c8037-2a1b-4d51-baf0-9de130846239\\installation-3.5.12\\import\\knows-with-node-properties.json",
encoding='utf-8') as f:
for item in f:
item = json.loads(item)
for i in ['n1', 'n2', 'n3', 'n4', 'n5']:
if i in item:
d.append(item[i]['name'] + "_" + item[i]['cate'])
d = list(set(d))
name_dict = {}
count = 0
for j in d:
j_array = j.split("_")
data_item = {}
name_dict[j_array[0]] = count
count += 1
data_item['name'] = j_array[0]
data_item['category'] = CA_LIST[j_array[1]]
json_data['data'].append(data_item)
# print(name_dict)
with open(
"D:\\Program Files (x86)\\.Neo4jDesktop\\neo4jDatabases\\database-3c0c8037-2a1b-4d51-baf0-9de130846239\\installation-3.5.12\\import\\knows-with-node-properties.json",
encoding='utf-8') as f:
for item in f:
item = json.loads(item)
for i, key in enumerate(['r1', 'r2', 'r3', 'r4']):
if key in item:
link_item = {}
link_item['source'] = name_dict[item["n" +
str(i + 1)]['name']]
link_item['target'] = name_dict[item["n" +
str(i + 2)]['name']]
link_item['value'] = item[key]['relation']
json_data['links'].append(link_item)
# print(json_data)
return json_data
def get_KGQA_answer(array):
data_array = []
for i in range(len(array) - 2):
if i == 0:
name = array[0]
else:
name = data_array[-1]['p.Name']
data = graph.run(
"match(p)-[r:%s{relation: '%s'}]->(n:Person{Name:'%s'}) return p.Name,n.Name,r.relation,p.cate,n.cate"
% (similar_words[array[i + 1]], similar_words[array[i + 1]], name))
data = list(data)
print(data)
data_array.extend(data)
print("===" * 36)
with open("./spider/images/" + "%s.jpg" % (str(data_array[-1]['p.Name'])),
"rb") as image:
base64_data = base64.b64encode(image.read())
b = str(base64_data)
return [
get_json_data(data_array),
get_profile(str(data_array[-1]['p.Name'])),
b.split("'")[1]
]
def get_KGQA_answer2(array):
data_array = []
print(array)
k = len(array)
for i in range(5):
if i == 0:
name = array[0]
else:
if i < k:
name = array[i]
else:
break
# name=data_array[i]['p.name']
data = graph.run(
"match(p{name:'%s'})-[r]->(n) return p.name,n.name,r.property1,p.label,n.label"
% (name))
# array.append(data[i].name)
# data2 = data['p.name']
data = list(data)
data_array.extend(data)
if data:
for j in range(len(data)):
array.append(data[j]['n.name'])
else:
break
k = len(array)
print("===" * 36)
# with open("./spider/images/"+"%s.jpg" % (str(data_array[-1]['p.name'])), "rb") as image:
# base64_data = base64.b64encode(image.read())
# b=str(base64_data)
return get_json_data(data_array)
def query(name):
data = graph.run( # 匹配所有与查询人有关系的完整数据（无论是正向关系还是反向关系），包括实体及其各种属性
"match(p)-[r]->(n:Person{Name:'%s'}) return p.Name,r.relation,n.Name,p.cate,n.cate\
Union all\
match(p:Person {Name:'%s'}) -[r]->(n) return p.Name, r.relation, n.Name, p.cate, n.cate" % (name,name)
)
data = list(data)
return get_json_data(data) #
def query(name, method):
if method == 'true':
data = graph.run(
"match(p)-[r]->(n) where n.Name =~ '.*%s.*' return p.Name,r.relation,n.Name,p.cate,n.cate\
Union all\
match(p) -[r]->(n) where p.Name =~ '.*%s.*' return p.Name, r.relation, n.Name, p.cate, n.cate\
" % (name, name))
data = list(data)
return get_json_data(data)
elif method == 'false':
data = graph.run(
"match(p)-[r]->(n) where n.Name =~ '%s' return p.Name,r.relation,n.Name,p.cate,n.cate\
Union all\
match(p) -[r]->(n) where p.Name =~ '%s' return p.Name, r.relation, n.Name, p.cate, n.cate\
" % (name, name))
data = list(data)
return get_json_data(data)
def query(name):
data = graph.run(
"match(p )-[r]->(n:Person{Name:'%s'}) return p.Name,r.relation,n.Name,p.cate,n.cate\
Union all\
match(p:Person {Name:'%s'}) -[r]->(n) return p.Name, r.relation, n.Name, p.cate, n.cate" % (name, name)
)
data = list(data)
return get_json_data(data)
def query_coauthor(name,level):
if level == '1':
data = graph.run(
"match(p )-[r:COAUTHOR]->(n:AUTHOR{authorName:'%s'}) return p.authorName, r, n.authorName\
Union all\
match(p:AUTHOR {authorName:'%s'}) -[r:COAUTHOR]->(n) return p.authorName, r, n.authorName" % (name,name)
)
elif level == '2':
data = graph.run(
"match(p:AUTHOR{authorName:'%s'} )-[r:COAUTHOR*1..2]->(n) return p.authorName, r, n.authorName" % (name)
)
elif level == '3':
data = graph.run(
"match(p:AUTHOR{authorName:'%s'} )-[r:COAUTHOR*1..3]->(n) return p.authorName, r, n.authorName" % (name)
)
data = list(data)
return get_json_data(data)
def query(name):
data = graph.run(
# match(p{name: '朱文生'})-[r]->(n)return p.name, r.property1, n.name, p.label, n.label
"match(p)-[r]->(n{name:'%s'}) return p.name,r.property1,n.name,p.label,n.label\
Union all\
match(p{Name:'%s'}) -[r]->(n) return p.name, r.property1, n.name, p.label, n.label"
% (name, name))
data = list(data)
return get_json_data(data)
def query_name(name):
nodes = map(
buildNodes,
graph.run(
"match(p)-[r]->(n:Person{Name:'%s'}) return p.Name,n.Name,p.Clan,n.Clan\
Union all\
match(p:Person {Name:'%s'}) -[r]->(n) return p.Name,n.Name,p.Clan,n.Clan" %
(name, name)).data())
edges = map(
buildEdges,
graph.run(
"match(p)-[r]->(n:Person{Name:'%s'}) return p.Name, r.relation,n.Name\
Union all\
match(p:Person {Name:'%s'}) -[r]->(n) return p.Name, r.relation,n.Name" %
(name, name)).data())
return nodes, edges
def query(name):
# print(name)
name_ = name.capitalize()
data = graph.run(
"match(p )-[r]->(n:Keyword{Name:'%s'}) return p.Name,r.relation,n.Name,p.level,n.level\
Union all\
match(p:Keyword{Name:'%s'}) -[r]->(n) return p.Name, r.relation, n.Name, p.level, n.level"
% (name_, name_))
data = list(data)
# print(data)
# json_data = get_json_data(data)
return get_json_data(data, name)
def get_KGQA_answer(sqls):
json_data = {'data': [], "links": []}
for sql_ in sqls:
question_type = sql_['question_type']
queries = sql_['sql']
for query in queries:
data = graph.run(query)
data = list(data)
res = get_json_data(data)
json_data['data'].extend(res['data'])
json_data['links'].extend(res['links'])
return json_data
def get_KGQA_answer(array):
data_array = []
result = '默认'
tags_list = [
'小说', '编程', 'web', '算法', '神经网络', '科技', '名著', '推理', '悬疑', '青春', '言情',
'校园', '经济', '漫画', '散文', '其他'
]
for i in range(len(array) - 2):
if i == 0:
name = array[0]
else:
# 下标为-1代表输出最后一个数
name = data_array[-1]['p.Name']
# 在数据库中查找以name为头实体，array[i+1]为关系的尾实体
data = graph.run(
"match(p)-[r]->(n) where r.relation =~ '.*%s.*' and (n.Name =~ '.*%s.*' or p.Name =~ '.*%s.*') return p.Name,n.Name,r.relation,p.cate,n.cate\
" % (similar_words[array[i + 1]], name, name))
# data是查询出来的结果，包含两个实体、实体的属性以及关系
# list()将元组转化为列表
data = list(data)
# data_array储存查出来的所有路径上的实体
data_array.extend(data)
if str(data_array[-1]['p.cate']) in tags_list:
result = str(data_array[-1]['p.Name'])
else:
result = '默认'
# 打开json查找图片地址
with open('./spider/json/data.json', encoding='utf-8') as f:
data = json.load(f)
img_url = 'https://ss1.bdstatic.com/70cFvXSh_Q1YnxGkpoWK1HF6hhy/it/u=1802553443,2497346274&fm=26&gp=0.jpg'
for i in data[result]:
if str(i) == "图片链接":
img_url = str(data[result][i])
break
# 将图片保存到本地
request.urlretrieve(img_url, './spider/images/' + '%s.jpg' % (result))
# with expression [as target]:
# expression是一个需要执行的表达式
# target是一个变量或元组，储存的是expression表达式执行返回的结果
# 打开对应的图片
with open("./spider/images/" + "%s.jpg" % (result), "rb") as image:
# print(str(data_array[-1]['p.Name']))
# 读取图片存入base64_data变量
base64_data = base64.b64encode(image.read())
# 转化为字符串
b = str(base64_data)
# get_json_data() 返回json格式的data_array
# get_profile() 返回查出答案的详细信息
# 返回一个数组，包含查出来的所有路径上的实体，查出实体的详细信息，查出实体的图片
# str(data_array[-1]['p.Name'])是查出来的目标结果对于的名字，如"贾宝玉"
return [get_json_data(data_array), get_profile(result), b.split("'")[1]]
def get_all_profile(name):
s = ''
data = graph.run("MATCH (n) WHERE n.name = '%s' RETURN properties(n) as props" % name).data()
# print(data)
for i in data[0]['props']:
# st = "<dt class = \"basicInfo-item name\" >" + str(i) + " \
# <dd class = \"basicInfo-item value\" >" + str(data[0]['props'][i]) + "</dd >"
st = "<dt class = \"basicInfo-item name\" >" + str(i) + " \
<input type=\"text\" value=\""+str(data[0]['props'][i])+"\" class=\"form-control\" class=\"prop\">"
# <dd class = \"basicInfo-item value\" >" + str(data[0]['props'][i]) + "</dd >"
s += st
return s
# get_all_profile("数据结构")
def fuzzy_search(array):
ss = ''
for name in array:
if name == "?" or name == "的":
continue
name_lst = name[::1]
sql_str = '.*'.join(name_lst)
data = graph.run(
"match (n:Concept) where n.name =~ '.*%s.*' return n.name" %
sql_str).data()
s = ''
if str(data) != "[]":
for i in data:
st = '<button class="btn btn-default" data=' + i[
"n.name"] + '>' + i["n.name"] + '</button> '
s += st
ss += s
return ss
def get_KGQA_answer(array):
data_array = []
for i in range(len(array) - 1):
if i == 0:
name = array[0]
else:
name = data_array[-1]['n.name'] # n
data = graph.run(
# "match(p)-[r:%s{relation: '%s'}]->(n:Person{Name:'%s'}) return p.Name,n.Name,r.relation,p.cate,n.cate" % (
# similar_words[array[i+1]], similar_words[array[i+1]], name)
"match(p:Concept {name:'%s'})-[r:%s{relation: '%s'}]->(n) return p.name,n.name,r.relation,p.cate,n.cate"
% (name, array[i + 1], array[i + 1])) # n, p
data = list(data)
# print(data)
data_array.extend(data)
# print(data_array)
# b = ''
# if len(array) >= 1:
# last_name = str(data_array[-1]['n.name'])
# image_path = getpath + "/spider/images/%s.jpg" % (str(data_array[-1]['n.name']))
# if not os.path.exists(image_path):
# image_path = getpath + "/spider/images/数据结构.jpg"
# with open(image_path, "rb") as image: # n
# base64_data = base64.b64encode(image.read())
# b = str(base64_data)
# else:
# last_name = '数据结构'
# image_path = getpath + "/spider/images/数据结构.jpg"
# with open(image_path, "rb") as image: # n
# base64_data = base64.b64encode(image.read())
# b = str(base64_data)
# with open(getpath + "/spider/images/%s.jpg" % (str(data_array[-1]['n.name'])), "rb") as image: # n
# base64_data = base64.b64encode(image.read())
# b = str(base64_data)
# return [get_json_data(data_array), get_profile(str(data_array[-1]['n.name'])), b.split("'")[1]] # n
return [get_json_data(data_array)] # n
def all():
data = graph.run(
"match (p)-[r]->(n) return p.name, r.relation, n.name, p.cate, n.cate")
data = list(data)
d = []
for i in data:
d.append(i['p.name'] + "_" + i['p.cate'])
d.append(i['n.name'] + "_" + i['n.cate'])
d = list(set(d))
name_dict = {}
count = 0
for j in d:
j_array = j.split("_")
name_dict[j_array[0]] = count
count += 1
name_dict_f = open(getpath + '/neo_db/name_dict.txt', 'w+')
name_dict_f.write(str(name_dict))
name_dict_f.close()
json_data = get_json_data(data)
return json_data, name_dict
def all_graph():
data = graph.run(
"match(p)-[r]->(n) return p.Name,r.relation,n.Name,p.cate,n.cate"
)
data = list(data)
return get_json_data(data)
def delete_node(name):
graph.run("match (n {name:'%s'} ) detach delete n" % name)
return
def concept():
data = list(graph.run("MATCH (n:CONCEPT) RETURN n LIMIT 10000"))
f = open('../raw_data/concept.txt', 'w', encoding='utf-8')
for d in data:
f.write(d['n']['conceptName'] + '\n')
def author():
data = list(graph.run("MATCH (n:AUTHOR) RETURN n LIMIT 10000"))
f = open('../raw_data/author.txt', 'w', encoding='utf-8')
for d in data:
f.write(d['n']['authorName'] + '\n')
def query_author_info(name):
data = graph.run("match(p :AUTHOR{authorName:'%s'})-[r]->(n) return p, r, n "%(name))
data = list(data)
json_data = {'data': [], "links": []}
data_item1 = {}
data_item1['name'] = name
data_item1['category'] = "AUTHOR"
json_data['data'].append(data_item1)
count = 0
name_dict = {}
name_dict[data_item1['name']] = count
count+=1
for i in data:
if i['r']['type'] == 'interest':
data_item2 = {}
data_item2['name'] = i['n']['conceptName']
data_item2['category'] = "CONCEPT"
name_dict[data_item2['name']] = count
count += 1
link_item = {}
link_item['source'] = name_dict[i['p']['authorName']]
link_item['target'] = name_dict[i['n']['conceptName']]
link_item['value'] = 'AUTHOR2CONCEPT'
json_data['data'].append(data_item2)
json_data['links'].append(link_item)
if i['r']['type'] == 'belong2':
data_item2 = {}
data_item2['name'] = i['n']['affiliationName']
data_item2['category'] = "AFFILIATION"
name_dict[data_item2['name']] = count
count += 1
link_item = {}
link_item['source'] = name_dict[i['p']['authorName']]
link_item['target'] = name_dict[i['n']['affiliationName']]
link_item['value'] = 'AUTHOR2AFFILIATION'
json_data['data'].append(data_item2)
json_data['links'].append(link_item)
if i['r']['type'] == 'own':
data_item2 = {}
data_item2['name'] = i['n']['paperID']
data_item2['category'] = "PAPER"
name_dict[data_item2['name']] = count
count += 1
link_item = {}
link_item['source'] = name_dict[i['p']['authorName']]
link_item['target'] = name_dict[i['n']['paperID']]
link_item['value'] = 'AUTHOR2PAPER'
json_data['data'].append(data_item2)
json_data['links'].append(link_item)
if i['r']['type'] == 'Collaborate':
data_item2 = {}
data_item2['name'] = i['n']['authorName']
data_item2['category'] = "AUTHOR"
name_dict[data_item2['name']] = count
count += 1
link_item = {}
link_item['source'] = name_dict[i['p']['authorName']]
link_item['target'] = name_dict[i['n']['authorName']]
link_item['value'] = 'COAUTHOR'
json_data['data'].append(data_item2)
json_data['links'].append(link_item)
return json_data
def get_hot_author():
data = graph.run(
"match(p:AUTHOR{authorName:'%s'} )-[r:COAUTHOR*1..3]->(n) return p.authorName, n.authorName"
)
data = list(data)
from py2neo import Graph, Node, Relationship
import json
from neo_db.config import graph
graph.run("match (n) detach delete n")
with open("../raw_data/relation.csv", encoding='gbk') as f:
for line in f.readlines():
rela_array = line.strip("\n").split(",")
print(rela_array)
graph.run("merge (p:Concept {cate:'%s', name: '%s'})" %
(rela_array[3], rela_array[0]))
graph.run("merge (p: Concept{cate:'%s',name: '%s'})" %
(rela_array[4], rela_array[1]))
graph.run(
"MATCH(e: Concept), (cc: Concept) WHERE e.name='%s' AND cc.name='%s' CREATE(e)-"
"[r:%s{relation: '%s'}]->(cc) RETURN r" %
(rela_array[0], rela_array[1], rela_array[2], rela_array[2]))
# 添加属性
name_lst = []
with open('../KGQA/my_dict.txt', 'r', encoding='utf-8') as f:
for line in f:
name_lst.append(line.strip('\n'))
with open('../spider/json/data.json', encoding='utf-8') as f:
data = json.load(f)
for name in name_lst:
for i in data[name]:
print(name, i, data[name][i])
j = "".join(i.split())
graph.run('match (n:Concept{name:"%s"}) set n.%s="%s"' %
(name, j, data[name][i]))
def query_by_sentence(sent_list):
if (sent_list[0] == 1):
#build nodes 焦大和尤氏
nodes1 = map(
buildNodes2,
graph.run(
"match t = allshortestPaths((p:Person) -[*]->(n:Person)) WHERE p.Name = '%s' and n.Name = '%s' return t as shortpath, NODES(t) AS nodes Union all match t = allshortestPaths((p:Person) -[*]->(n:Person)) WHERE p.Name = '%s' and n.Name = '%s' return t as shortpath, NODES(t) AS nodes"
% (sent_list[1], sent_list[2], sent_list[2],
sent_list[1])).data())
nodes = []
for i in list(nodes1):
for ii in i:
tmpdict = {}
tmpdict['data'] = ii
nodes.append(tmpdict)
#build edges
edges1 = map(
buildEdges2,
graph.run(
"match t = allshortestPaths((p:Person) -[*]->(n:Person)) WHERE p.Name = '%s' and n.Name = '%s' return t as shortpath, NODES(t) AS nodes Union all match t = allshortestPaths((p:Person) -[*]->(n:Person)) WHERE p.Name = '%s' and n.Name = '%s' return t as shortpath, NODES(t) AS nodes"
% (sent_list[1], sent_list[2], sent_list[2],
sent_list[1])).data())
edgelist = list(edges1)
edges = []
for i in edgelist:
for ii in i:
tmpdict = {}
tmpdict['data'] = ii
edges.append(tmpdict)
return nodes, edges
elif (sent_list[0] == 2):
nodes = map(
buildNodes,
graph.run(
"match(n:Person{Name:'%s'})-[r:%s{relation: '%s'}]->(p) return p.Name,n.Name,p.Clan,n.Clan \
Union all\
match(p:Person{Name:'%s'})-[r:%s{relation: '%s'}]->(n) \
return p.Name,n.Name,p.Clan,n.Clan" %
(sent_list[1], sent_list[3], sent_list[3], sent_list[1],
sent_list[3], sent_list[3])).data())
edges = map(
buildEdges,
graph.run(
"match(p:Person{Name:'%s'})-[r:%s{relation: '%s'}]->(n) return p.Name, r.relation,n.Name"
% (sent_list[1], sent_list[3], sent_list[3])).data())
elif (sent_list[0] == 3):
nodes = map(
buildNodes,
graph.run(
"match(n)-[r:%s{relation: '%s'}]->(p:Person{Name:'%s'}) return p.Name,n.Name,p.Clan,n.Clan \
Union all\
match(p)-[r:%s{relation: '%s'}]->(n:Person{Name:'%s'}) \
return p.Name,n.Name,p.Clan,n.Clan" %
(sent_list[3], sent_list[3], sent_list[2], sent_list[3],
sent_list[3], sent_list[2])).data())
edges = map(
buildEdges,
graph.run(
"match(p)-[r:%s{relation: '%s'}]->(n:Person{Name:'%s'}) return p.Name, r.relation,n.Name"
% (sent_list[3], sent_list[3], sent_list[2])).data())
else:
pass
return nodes, edges
def query_path(a, b):
data = graph.run(
"MATCH p =shortestPath((n { name: '%s' })-[*..15]-(m {name:'%s'})) RETURN p"
% (a, b))
data = data.data()
# print(data)
data = data[0]['p']
data = str(data)
# print(data)
pattern11 = re.compile("\[:([\u4e00-\u9fa5]*) {}\]->\(([\u4e00-\u9fa5]*)\)"
) # [:属于 {}]->(线性结构)
pattern22 = re.compile("\(([\u4e00-\u9fa5]*)\)<-\[:([\u4e00-\u9fa5]*) {}\]"
) # (线性结构)<-[:分类 {}]
pattern3 = re.compile(":([\u4e00-\u9fa5]*)") # :属于
pattern4 = re.compile("\(([\u4e00-\u9fa5]*)\)") # (串)
right = pattern11.findall(
data) # [('属于', '线性结构'), ('研究对象', '存储结构'), ('包括', '索引存储')]
left = pattern22.findall(data) # [('线性结构', '分类')]
# relation = pattern3.findall(data) # ['属于', '分类', '研究对象', '包括']
concept = pattern4.findall(data) # ['串', '线性结构', '数据结构', '存储结构', '索引存储']
arr_list = []
for item in right:
a = ""
for j in range(len(concept)):
if concept[j] == item[1]:
a += concept[j - 1] + "," + item[1] + "," + item[0]
arr_list.append(a)
for item in left:
a = ""
for j in range(len(concept)):
if concept[j] == item[0]:
a += concept[j + 1] + "," + item[0] + "," + item[1]
arr_list.append(a)
print(
f"list:{arr_list}"
) # list:['串,线性结构,属于', '数据结构,存储结构,研究对象', '存储结构,索引存储,包括', '数据结构,线性结构,分类']
data_dict = col_to_dic()
json_data = {'data': [], "links": []}
d = []
for line in arr_list:
rela_array = line.split(",")
for item in data_dict:
if rela_array[0] in data_dict[item]:
rela_array.append(item)
if rela_array[1] in data_dict[item]:
rela_array.append(item)
d.append(rela_array[0] + "_" + rela_array[3])
d.append(rela_array[1] + "_" + rela_array[4])
d = list(set(d))
name_dict = {}
count = 0
for j in d:
j_array = j.split("_")
data_item = {}
name_dict[j_array[0]] = count
count += 1
data_item['name'] = j_array[0]
data_item['category'] = CA_LIST[j_array[1]]
json_data['data'].append(data_item)
for line in arr_list:
rela_array = line.split(",")
link_item = {}
link_item['source'] = name_dict[rela_array[0]]
link_item['target'] = name_dict[rela_array[1]]
link_item['value'] = rela_array[2]
json_data['links'].append(link_item)
return json_data
def update_node(node, cate):
print(node, cate)
graph.run("MATCH (n) WHERE n.name = '%s' SET n.cate = '%s'" % (node, cate))
return
