hive 面试题 转载

转自：http://blog.csdn.net/ningguixin/article/details/12852051

有一张很大的表：TRLOG
该表大概有2T左右
TRLOG：
CREATE TABLE TRLOG
(PLATFORM string,
USER_ID int,
CLICK_TIME string,
CLICK_URL string)
row format delimited
fields terminated by '\t';

数据：
PLATFORM USER_ID CLICK_TIME CLICK_URL
WEB 12332321 2013-03-21 13:48:31.324 /home/
WEB 12332321 2013-03-21 13:48:32.954 /selectcat/er/
WEB 12332321 2013-03-21 13:48:46.365 /er/viewad/12.html
WEB 12332321 2013-03-21 13:48:53.651 /er/viewad/13.html
WEB 12332321 2013-03-21 13:49:13.435 /er/viewad/24.html
WEB 12332321 2013-03-21 13:49:35.876 /selectcat/che/
WEB 12332321 2013-03-21 13:49:56.398 /che/viewad/93.html
WEB 12332321 2013-03-21 13:50:03.143 /che/viewad/10.html
WEB 12332321 2013-03-21 13:50:34.265 /home/
WAP 32483923 2013-03-21 23:58:41.123 /m/home/
WAP 32483923 2013-03-21 23:59:16.123 /m/selectcat/fang/
WAP 32483923 2013-03-21 23:59:45.123 /m/fang/33.html
WAP 32483923 2013-03-22 00:00:23.984 /m/fang/54.html
WAP 32483923 2013-03-22 00:00:54.043 /m/selectcat/er/
WAP 32483923 2013-03-22 00:01:16.576 /m/er/49.html
…… …… …… ……

需要把上述数据处理为如下结构的表ALLOG：
CREATE TABLE ALLOG
(PLATFORM string,
USER_ID int,
SEQ int,
FROM_URL string,
TO_URL string)
row format delimited
fields terminated by '\t';

整理后的数据结构：
PLATFORM USER_ID SEQ FROM_URL TO_URL
WEB 12332321 1 NULL /home/
WEB 12332321 2 /home/ /selectcat/er/
WEB 12332321 3 /selectcat/er/ /er/viewad/12.html
WEB 12332321 4 /er/viewad/12.html /er/viewad/13.html
WEB 12332321 5 /er/viewad/13.html /er/viewad/24.html
WEB 12332321 6 /er/viewad/24.html /selectcat/che/
WEB 12332321 7 /selectcat/che/ /che/viewad/93.html
WEB 12332321 8 /che/viewad/93.html /che/viewad/10.html
WEB 12332321 9 /che/viewad/10.html /home/
WAP 32483923 1 NULL /m/home/
WAP 32483923 2 /m/home/ /m/selectcat/fang/
WAP 32483923 3 /m/selectcat/fang/ /m/fang/33.html
WAP 32483923 4 /m/fang/33.html /m/fang/54.html
WAP 32483923 5 /m/fang/54.html /m/selectcat/er/
WAP 32483923 6 /m/selectcat/er/ /m/er/49.html
…… …… …… ……
PLATFORM和USER_ID还是代表平台和用户ID；SEQ字段代表用户按时间排序后的访问顺序，FROM_URL和TO_URL分别代表用户从哪一页跳转到哪一页。对于某个平台上某个用户的第一条访问记录，其FROM_URL是NULL（空值）。

面试官说需要用两种办法做出来：
1、实现一个能加速上述处理过程的Hive Generic UDF，并给出使用此UDF实现ETL过程的Hive SQL

2、实现基于纯Hive SQL的ETL过程，从TRLOG表生成ALLOG表；（结果是一套SQL）

答案：

1.

UDF

package org.apache.hadoop.hive.udf;

public class RowNumber extends org.apache.hadoop.hive.ql.exec.UDF {

    private static int MAX_VALUE = 50;
    private static String comparedColumn[] = new String[MAX_VALUE];
    private static int rowNum = 1;

    public int evaluate(Object... args) {
        String columnValue[] = new String[args.length];
        for (int i = 0; i < args.length; i++)
            columnValue[i] = args[i].toString();
        if (rowNum == 1)
        {

            for (int i = 0; i < columnValue.length; i++)
                comparedColumn[i] = columnValue[i];
        }

        for (int i = 0; i < columnValue.length; i++)
        {

            if (!comparedColumn[i].equals(columnValue[i]))
            {
                for (int j = 0; j < columnValue.length; j++)
                {
                    comparedColumn[j] = columnValue[j];
                }
                rowNum = 1;
                return rowNum++;
            }
        }
        return rowNum++;
    }

    public static void main(String[] args) {
    	RowNumber aRowNumber = new RowNumber();
    	System.out.println(aRowNumber.evaluate("12332321"));
    	System.out.println(aRowNumber.evaluate("12332321"));
    	System.out.println(aRowNumber.evaluate("12332321"));
    	System.out.println(aRowNumber.evaluate("12332321"));
    	System.out.println(aRowNumber.evaluate("12332321"));
	}

}

　　

INSERT OVERWRITE TABLE ALLOG
SELECT t1.platform,t1.user_id,RowNumber(t1.user_id)seq,t2.click_url FROM_URL,t1.click_url TO_URL FROM
(select *,RowNumber(user_id)seq from trlog)t1
LEFT OUTER JOIN
(select *,RowNumber(user_id)seq from trlog)t2
on t1.user_id = t2.user_id and t1.seq=t2.seq+1;

2.

INSERT OVERWRITE TABLE ALLOG
SELECT t1.platform,t1.user_id,t1.seq,t2.click_url FROM_URL,t1.click_url TO_URL FROM
(SELECT platform,user_id,click_time,click_url,count(1) seq FROM (SELECT a.*,b.click_time click_time1,b.click_url click_url2  FROM trlog a left outer join trlog b on a.user_id = b.user_id)t WHERE click_time>=click_time1 GROUP BY platform,user_id,click_time,click_url)t1
LEFT OUTER JOIN
(SELECT platform,user_id,click_time,click_url,count(1) seq FROM (SELECT a.*,b.click_time click_time1,b.click_url click_url2  FROM trlog a left outer join trlog b on a.user_id = b.user_id)t WHERE click_time>=click_time1 GROUP BY platform,user_id,click_time,click_url )t2 
on t1.user_id = t2.user_id and t1.seq = t2.seq + 1;

使用到的知识点为：

left outer join  左表全部显示，右表只显示满足条件的

3、对于以上的文本处理 我们可以很快的联想到shell中awk的处理

利用awk 中数组的相关操作，方法如下

cat url.txt |awk -F\\t 'BEGIN{OFS="\t"}{a[$1]++;b[a[$1]]=$4;print a[$1],$1,$2,$3,b[a[$1]-1],$4}'

　　其中OFS为输出的字段的定界符，这里利用了2个数组，a和b

输出为：

1	WEB	12332321	2013-03-21 13:48:31.324		/home
2	WEB	12332321	2013-03-21 13:48:32.954	/home	/selectcat/er
3	WEB	12332321	2013-03-21 13:48:46.365	/selectcat/er	/er/viewad/12.html
4	WEB	12332321	2013-03-21 13:48:53.651	/er/viewad/12.html	/er/viewad/13.html
5	WEB	12332321	2013-03-21 13:49:13.435	/er/viewad/13.html	/er/viewad/24.html
6	WEB	12332321	2013-03-21 13:49:35.876	/er/viewad/24.html	/selectcat/che/
7	WEB	12332321	2013-03-21 13:49:56.398	/selectcat/che/	/che/viewad/93.html
8	WEB	12332321	2013-03-21 13:50:03.143	/che/viewad/93.html	/che/viewad/10.html
9	WEB	12332321	2013-03-21 13:50:34.265	/che/viewad/10.html	/home/
1	WAP	32483923	2013-03-21 23:58:41.123		/m/home/
2	WAP	32483923	2013-03-21 23:59:16.123	/m/home/	/m/selectcat/fang/
3	WAP	32483923	2013-03-21 23:59:45.123	/m/selectcat/fang/	/m/fang/33.html
4	WAP	32483923	2013-03-22 00:00:23.984	/m/fang/33.html	/m/fang/54.html
5	WAP	32483923	2013-03-22 00:00:54.043	/m/fang/54.html	/m/selectcat/er/
6	WAP	32483923	2013-03-22 00:01:16.576	/m/selectcat/er/	/m/er/49.html