基于Flask的简易搜索引擎，倒排索引存放在mongodb服务器

from flask import Flask, request from jpype import * from mySegment import * import pymongo from bson import ObjectId,json_util app = Flask(__name__, static_folder='static/', static_url_path='') segment_test = mySegment() mongodbAddr = 'mongodb://xxx.xxx.xxx.xxx:27017/' myclient = pymongo.MongoClient(mongodbAddr) guangmingNewsDB = myclient['guangmingNews'] wordDoc = guangmingNewsDB['word'] urlDoc = guangmingNewsDB['url'] @app.route('/') def root(): return app.send_static_file('index.html') @app.route('/query', methods=['GET', 'POST']) def query(): add_info = "empty" res = "我不明白你的意思" a = request print(request) if request.method == 'GET': str_in = request.args.get("queryFromUser") else: str_in = request.form.get("queryFromUser") #这里名字改了之后，前段部分也要改，如果名字一致还是报错，就需要清理浏览器缓存 print("this is query:") print(str_in) #对查询语句进行分词 segment_list = segment_test.segmentFilter(str_in) if len(segment_list) != 1: #按多个关键词搜索 test_words = segment_list search_array = [] for each in test_words: search_array.append(dict(_id=each)) print(search_array) test_pipline = [ { '$lookup': { 'from': 'word', 'localField': '_id', 'foreignField': '_id', 'as': 'aggregate' }}, { '$match': { '$or': search_array } }, { '$limit': 10 } ] test_aggregate_result = wordDoc.aggregate(test_pipline) #mongodb聚合查询 intersection_result = [] i = 0 for each in test_aggregate_result: if i == 0: intersection_result = list(each['count'].keys()) print(each['count']) intersection_result = list(set(intersection_result).intersection(set(each['count'].keys()))) i = i + 1 print("交集: ") print(intersection_result) url_result = [] for each in intersection_result: urlId = each tmp = urlDoc.find_one({"_id": ObjectId(urlId)}) tmp["_id"] = tmp["_id"].__str__() url_result.append(tmp) print(json_util.dumps(url_result)) myclient.close() return json_util.dumps(url_result) else: #按单个关键词查询 word = "".join(segment_list) query_result = wordDoc.find_one({"_id": word}) if not query_result: return 0 print(query_result["count"]) sort_result = segment_test.sort_by_value(query_result["count"]) url_result = [] for each in sort_result: #print(each) urlId = each tmp = urlDoc.find_one({"_id": ObjectId(urlId)}) tmp["_id"] = tmp["_id"].__str__() url_result.append(tmp) #print(url_result) print(json_util.dumps(url_result)) myclient.close() return json_util.dumps(url_result)