def start(): connectDB() calc.init_similarity() createSourceFile() callBGLL() getAllResult() clearHierarchyRelation() # group: the id of each tree for (group, pid_set) in result_paper_cate[total_level].items(): print "======================level:%s====================" % group is_selected = set() updateHierarchyRelation(group, pid_set, total_level, -1, is_selected) closeDB()
def __init__(self): self.BGLL = BGLL(self.output_path, self.is_weighted) self.database = Database(self.db_host, self.db_user, self.db_pass, self.db_name) calc.init_similarity()
import MySQLdb import os import CalcFather as calc # database config db_host = "localhost" db_user = "******" db_pass = "******" db_name = "papernet" # global database connection and cursor conn = MySQLdb.connect(db_host, db_user, db_pass, db_name) cursor = conn.cursor() calc.init_similarity() # read the database paper_paper_relation and write into paper.txt map1 = dict() # from old to new map2 = dict() # from new to old file_path = '/home/cowx/workspace/BGLL/paper/paper.txt' f = open(file_path, 'w+') sql = "select pid1, pid2 from paper_paper_relation order by pid1, pid2" cursor.execute(sql) result = cursor.fetchall() pid_set = set() # the set of all nodes for row in result: pid_set.add(row[0]) pid_set.add(row[1])