对于Spark SQL:

DESCRIBE [db_name.]table_name column_name

在oracle SQL中，你可以这样做:

SELECT
    DATA_TYPE
FROM
    all_tab_columns 
WHERE
    table_name = 'TABLE NAME' -- in uppercase
AND column_name = 'COLUMN NAME' -- in uppercase

对于Apache Derby，如下面的答案所示:

select columndatatype from sys.syscolumns
  where referenceid = (
    select tableid from sys.systables
    where tablename = 'YOUR_TABEL_NAME'
    and columnname= 'YOUR_COLUMN_NAME')

下面的SQL脚本在我的结束完美地工作。

SELECT column_name AS 'Column Name',data_type AS 'Data Type' 
FROM information_schema.columns 
WHERE table_name = '[Table Name]'

使用此查询获取Schema, Table, Column,Type, max_length, is_nullable

SELECT QUOTENAME(SCHEMA_NAME(tb.[schema_id])) AS 'Schema'
    ,QUOTENAME(OBJECT_NAME(tb.[OBJECT_ID])) AS 'Table'
    ,C.NAME as 'Column'
    ,T.name AS 'Type'
    ,C.max_length
    ,C.is_nullable
FROM SYS.COLUMNS C INNER JOIN SYS.TABLES tb ON tb.[object_id] = C.[object_id]
    INNER JOIN SYS.TYPES T ON C.system_type_id = T.user_type_id
WHERE tb.[is_ms_shipped] = 0
ORDER BY tb.[Name]

MS SQL的另一个选项是用你想要的类型的查询替换这里的选择查询:

declare @sql varchar(4000);

set @sql = 'select ''hi'' as greeting';

select * from master.sys.dm_exec_describe_first_result_set (@sql, Null, 0);