日志中的时间为 17/Jul/2013:22:00:06 +0800]
a = load '/user/grid/full/201311{23,24,25}/*' using logloader() AS (remoteAddr:chararray, remoteLogname, user, time:chararray, method, uri:chararray, proto, status, bytes, referer:chararray, userAgent);
b = foreach a generate SUBSTRING(time,0,20) as d1:chararray,SUBSTRING(time,0,11) as date1:chararray,SUBSTRING(time,12,14) as d_hour:chararray,remoteAddr,LOWER(uri) as path:chararray;
c = filter b by d1 <= '25/Nov/2013:18:00:00'; --取小于这个时间的数据
