我目前可以提供2个选择，没有一个是完美的。
顺便说一句， "output.format.string" 已过时且无效。

1

create external table mytable
(
    q1          string    
   ,field1      string
   ,q2          string
   ,field2      string
   ,q3          string
   ,field3      string
)
row format serde 'org.apache.hadoop.hive.serde2.RegexSerDe'
with serdeproperties ('input.regex' = '.*?=(?<q1>"?)(.*?)(?:\\k<q1>)\\|.*?=(?<q2>"?)(.*?)(?:\\k<q2>)\\|.*?=(?<q3>"?)(.*?)(?:\\k<q3>)')
stored as textfile
;

select * from mytable
;

+----+--------+----+--------+----+-----------+
| q1 | field1 | q2 | field2 | q3 |  field3   |
+----+--------+----+--------+----+-----------+
|    | value2 |    | value2 | "  | va , lues |
+----+--------+----+--------+----+-----------+

2

create external table mytable
(
    field1 string
   ,field2 string
   ,field3 string
)
row format serde 'org.apache.hadoop.hive.serde2.RegexSerDe'
with serdeproperties ('input.regex' = '.*?=(".*?"|.*?)\\|.*?=(".*?"|.*?)\\|.*?=(".*?"|.*?)')
stored as textfile
;

select * from mytable
;

+--------+--------+-------------+
| field1 | field2 |   field3    |
+--------+--------+-------------+
| value2 | value2 | "va , lues" |
+--------+--------+-------------+