如何将CSV文件导出到带有蜂记录值的Hive表中？

Question

Input File

   11/24/2013,bank of nyc,withdrawl,deposit,in progress
   11/16/2014,bank of dc,opeanig,closing,resolved    

I want them in the table

  Date           Bank name         issue                status 
  11/24/2013     bank of nyc     withdrawl,deposit     in progress
  11/16/2014     bank of dc      opeanig,closing       resolved

Answer 1

Well, the problem is that the comma is not escaped...how is hive supposed to know if a comma is part of a string, or a separator ?

It is possible if you only can have extra commas in one known column, like in this case, the third. You can then write a regular expression that takes anything between the second comma and the last and use it with the Regexp serde. This works for your example, given that only 'issue' may have commas.

CREATE TABLE csvsample(
  date STRING,
  bank_name STRING,
  issue STRING,
  status STRING
) ROW FORMAT SERDE 'org.apache.hadoop.hive.contrib.serde2.RegexSerDe' 
  WITH SERDEPROPERTIES (
   "input.regex" = "^([^,]+),([^,]+),(.+),([^,]+)$") ;
hive> select * from csvsample;
OK
11/24/2013  bank of nyc withdrawl,deposit   in progress
11/16/2014  bank of dc  opeanig,closing resolved