Question

hadoop dfs输入数据cat：

[ituser1@genome-dev3 ~]$ hadoop fs -cat FOR_COPY/COMPETITOR_BROKERING/part-r-00000 | head -1

返回：

836646827,1000.0,2016-02-20,34,CAPITAL BOOK,POS/CAPITAL BOOK/NEW DELHI/200216/14:18,BOOKS AND STATIONERY,5497519004453567/41043516,MARRIED,M,SALARIED,D,5942,1

我的猪代码：

DATA = LOAD 'FOR_COPY/COMPETITOR_BROKERING' USING PigStorage(',') AS (CUST_ID:chararray,TXN_AMT:chararray,TXN_DATE:chararray,AGE_CASA:chararray,MERCH_NAME:chararray,TXN_PARTICULARS:chararray,MCC_CATEGORY:chararray,TXN_REMARKS:chararray,MARITAL_STATUS_CASA:chararray,GENDER_CASA:chararray,OCCUPATION_CAT_V2_NEW:chararray,DR_CR:chararray,MCC_CODE:chararray,OCCURANCE:int);

DATA_FIL = FOREACH DATA GENERATE                
                (chararray)CUST_ID AS CUST_ID,
                (chararray)TXN_AMT AS TXN_AMT,
                (chararray)TXN_DATE AS TXN_DATE,
                (chararray)AGE_CASA AS AGE_CASA,
                (chararray)MERCH_NAME AS MERCH_NAME,
                (chararray)TXN_PARTICULARS AS TXN_PARTICULARS,
                (chararray)MCC_CATEGORY AS MCC_CATEGORY,
                (chararray)TXN_REMARKS AS TXN_REMARKS,
                (chararray)MARITAL_STATUS_CASA AS MARITAL_STATUS_CASA,
                (chararray)GENDER_CASA AS GENDER_CASA,
                (chararray)OCCUPATION_CAT_V2_NEW AS OCCUPATION_CAT_V2_NEW,
                (chararray)DR_CR AS DR_CR,
                (chararray)MCC_CODE AS MCC_CODE;

STORE DATA_FIL INTO 'hbase://TXN_EVENTS' USING org.apache.pig.backend.hadoop.hbase.HBaseStorage ('DETAILS:CUST_ID DETAILS:TXN_AMT DETAILS:TXN_DATE DETAILS:AGE_CASA DETAILS:MERCH_NAME DETAILS:TXN_PARTICULARS DETAILS:MCC_CATEGORY DETAILS:TXN_REMARKS DETAILS:MARITAL_STATUS_CASA DETAILS:GENDER_CASA DETAILS:OCCUPATION_CAT_V2_NEW DETAILS:DR_CR DETAILS:MCC_CODE');

但是给出了错误：

ERROR org.apache.pig.tools.grunt.GruntParser - ERROR 2244: Job job_1457792710587_0100 failed, hadoop does not return any error message

但我的负载工作正常：

HDATA = LOAD 'hbase://TXN_EVENTS'
       USING org.apache.pig.backend.hadoop.hbase.HBaseStorage(
       'DETAILS:CUST_ID DETAILS:TXN_AMT DETAILS:TXN_DATE DETAILS:AGE_CASA DETAILS:MERCH_NAME DETAILS:TXN_PARTICULARS DETAILS:MCC_CATEGORY DETAILS:TXN_REMARKS DETAILS:MARITAL_STATUS_CASA DETAILS:GENDER_CASA DETAILS:OCCUPATION_CAT_V2_NEW DETAILS:DR_CR DETAILS:MCC_CODE','-loadKey true' )
       AS (ROWKEY:chararray,CUST_ID:chararray,TXN_AMT:chararray,TXN_DATE:chararray,AGE_CASA:chararray,MERCH_NAME:chararray,TXN_PARTICULARS:chararray,MCC_CATEGORY:chararray,TXN_REMARKS:chararray,MARITAL_STATUS_CASA:chararray,GENDER_CASA:chararray,OCCUPATION_CAT_V2_NEW:chararray,DR_CR:chararray,MCC_CODE:chararray);

DUMP HDATA; （这给出了完美的结果）：

2016-03-01,1,20.0,2016-03-22,27,test_merch,test/particulars,test_category,test/remarks,married,M,service,D,1234

感谢帮助

我在分布式模式下使用Horton堆栈：

HDP2.3 Apache Pig版本0.15.0 HBase 1.1.1

所有罐子都已到位，因为我已经通过Ambari安装了它们。

Answer 1

解决了数据上传问题：

因为我缺少排名关系，因此hbase rowkey成为排名。\

DATA_FIL_1 = RANK DATA_FIL_2;

注意：这将生成任意rowkey。

但是，如果你想定义你的行键，请使用like：

你必须给另一种关系，只有STORE功能不起作用。这将把第一个元组作为rowkey（你已定义）

storage_data = STORE DATA_FIL INTO'hbase：// genome：event_sink'使用org.apache.pig.backend.hadoop.hbase.HBaseStorage（'event_data：CUST_ID event_data：EVENT transaction_data：TXN_AMT transaction_data ：TXN_DATE transaction_data：AGE_CASA transaction_data：MERCH_NAME transaction_data：TXN_PARTICULARS transaction_data：MCC_CATEGORY transaction_data：TXN_REMARKS transaction_data：MARITAL_STATUS_CASA transaction_data：GENDER_CASA transaction_data：OCCUPATION_CAT_V2_NEW transaction_data：DR_CR transaction_data：MCC_CODE'）;

使用Pig

1 个答案: