将多行(具有相同 ID)合并为一行?

Merge multiple rows (with same ID) into one?

我希望将多行合并为具有匹配 ID 的一行。我正在使用 Teradata SQL。 示例:

  SELECT
  id_nbr AS ID,
  contact_type AS contype,
  contact_first_name AS firstName,
  contact_last_name AS lastName,
  contact_phone_number AS phoneNumber,
  contact_address AS address,
  contact_email AS email
  FROM database.account_info
  WHERE (contact_type = 'AAA' OR contact_type = 'BBB' OR contact_type = 'CCC');

结果:

ID          contype     firstName     lastName    phoneNumber     address        email
111111111   AAA         bob           smith       (999)999-9999   1 Main St      null
111111111   BBB         bob           smith       (888)888-8888   1 random rd    bob@random.com
111111111   CCC         bob           smith       null            null           bob@anotherrandom.com

我想要的:

ID          contypeAAA     firstNameAAA     lastNameAAA    phoneNumberAAA     addressAAA      emailAAA   contypeBBB  firstNameBBB     lastNameBBB    phoneNumberBBB     addressBBB      emailBBB  contypeCCC     firstNameCCC     lastNameCCC    phoneNumberCCC     addressCCC      emailCCC
111111111   AAA            bob              smith          (999)999-9999      1 Main St       null ......etc

这可能吗?这将极大地帮助我..!提前致谢!

编辑这个,ID 的contype 并不总是具有所有3 个contype。这使它变得更加棘手。例如,ID:99999999 可以有 contype:'AAA','BBB' 但是应该 return 所有 'CCC' 值的空值,因为它不存在。

我相信这会奏效(我在 Oracle 11g 而不是 Teradata 上测试过):

SELECT *
  FROM (SELECT id_nbr AS ID,
               contact_type AS contype,
               contact_first_name AS firstName,
               contact_last_name AS lastName,
               contact_phone_number AS phoneNumber,
               contact_address AS address,
               contact_email AS email
          FROM database.account_info
         WHERE contact_type in ('AAA', 'BBB', 'CCC')
        )
 PIVOT (max(firstName) AS firstName,
        max(lastName) AS lastName,
        max(phoneNumber) AS phone,
        max(email) AS email,
        max(address) AS address
   FOR contype IN ('AAA', 'BBB', 'CCC')
        )

最后一行可能需要一个别名,其中之一:

        ) AS derived_pivot
        ) derived_pivot