hive 批量添加，删除分区

一.批量添加分区： use bigdata;alter table siebel_memberadd if not existspartition(dt=’20180401′) location ‘20180401’partition(dt=’20180402′) location ‘20180402’partition(dt=’20180403′) location ‘20180403’partition(dt=’20180404′) location ‘20180404’; 二.批量删除分区 alter table siebel_member drop if existspartition(dt=’20180401′),partition(dt=’20180402′),partition(dt=’20180403′),partition(dt=’20180404′); 三.大量分区的添加虽说是批量添加分区，但是一下子添加2000个分区，如果把2000个分区写在一个语句里面，hive是不允许的。所以只好一个月一个月的加，尝试了使用shell 生成 line=”hive -e “xxxxx””  ${line} ,但是解析错误，采用python + shell 解决了问题。python 代码如下： #!/usr/bin/env python#encoding=utf8 import datetimeimport osimport sys  def dateRange(beginDate, endDate):    dates = []    dt = datetime.datetime.strptime(beginDate, “%Y%m%d”)    date = beginDate[:]    while date < endDate:        dates.append(date)        dt = dt + datetime.timedelta(1)        date = dt.strftime(“%Y%m%d”)    return dates  def genSql(datelist, table):    sql = “””hive -e “use bigdata;alter table “”” + table + “””add if not exists “””    for i in datelist:        sql = sql + “””partition(dt='””” + i + “””‘) location ‘””” + i + “””‘ “””    sql = sql.strip() + “””;” “””    os.system(sql)  if __name__ == ‘__main__’:    if sys.argv.__len__() != 4:        print(“参数错误”)        sys.exit()    st = sys.argv[1]    et = sys.argv[2]    table = sys.argv[3]    datelist = dateRange(st,et)    genSql(datelist,table)  shell 用来调度 py#添加分区,按月添加分区 会添加到et所在月的最后一天function add_par(){st=$1et=$2tbname=$3while [ ${et} -ge ${st} ]dopet=$(date –date=”+1 month ${et}” +”%Y%m%d”)python a.py ${et} ${pet} ${tbname}et=$(date –date=”1 month ago ${et}” +”%Y%m%d”)done}