T-SQL :: information_type 未在 SQL Server 2019 中解析

Question

我的查询似乎没有解析 information_type

STEP TO REPRODUCE:

采用 AdventureWork 或任何其他测试数据库
右键点击你的测试数据库>Tasks > Data Discovery and Classification > Classify Data...然后对一堆数据进行分类
现在使用此查询来检查您的 information_type 列（这里我使用的是 AdventureWork2019）：

粘贴此select并执行

SELECT 
    schema_name(O.schema_id) AS schema_name,
    O.NAME AS table_name,
    C.NAME AS column_name,
         [Type] = 
    CASE 
      WHEN ct.[name] IN ('varchar', 'char') THEN ct.[name] + '(' + IIF(c.max_length = -1, 'max', CAST(c.max_length AS VARCHAR(25))) + ')' 
      WHEN ct.[name] IN ('nvarchar','nchar') THEN ct.[name] + '(' + IIF(c.max_length = -1, 'max', CAST(c.max_length / 2 AS VARCHAR(25)))+ ')'      
      WHEN ct.[name] IN ('decimal', 'numeric') THEN ct.[name] + '(' + CAST(c.precision AS VARCHAR(25)) + ', ' + CAST(c.scale AS VARCHAR(25)) + ')'
      WHEN ct.[name] IN ('datetime2') THEN ct.[name] + '(' + CAST(c.scale AS VARCHAR(25)) + ')'
      ELSE ct.[name]
    END,
    information_type,
    label,
    rank,
    rank_desc,
    CAST(
             CASE
                  WHEN information_type = 'Contact Info' 
                     THEN 'contact info'
                  ELSE 'not contact info'
             END AS varchar(max)) as Checking_Content
FROM sys.sensitivity_classifications sc
    JOIN sys.objects O
    ON  sc.major_id = O.object_id
    JOIN sys.columns C 
    ON  sc.major_id = C.object_id  AND sc.minor_id = C.column_id
    JOIN sys.types ct ON C.user_type_id = ct.user_type_id
    --AND EP.minor_id = C.column_id
    order by information_type

如您所见，即使列 information_type 的值为 Contact Info

，查询也会返回 not contact info

schema_name	table_name	column_name	Type	information_type	label	rank	rank_desc	Checking_Content
Person	PersonPhone	PhoneNumber	Phone	Contact Info	Confidential	20	MEDIUM	not contact info
Person	PersonPhone	PhoneNumberTypeID	int	Contact Info	Confidential	20	MEDIUM	not contact info
Person	PhoneNumberType	PhoneNumberTypeID	int	Contact Info	Confidential	20	MEDIUM	not contact info
Person	Address	AddressLine1	nvarchar(60)	Contact Info	Confidential	20	MEDIUM	not contact info
Person	Address	AddressLine2	nvarchar(60)	Contact Info	Confidential	20	MEDIUM	not contact info
Person	Address	City	nvarchar(30)	Contact Info	Confidential	20	MEDIUM	not contact info
Person	Address	PostalCode	nvarchar(15)	Contact Info	Confidential	20	MEDIUM	not contact info
Production	ProductReview	EmailAddress	nvarchar(50)	Contact Info	Confidential	20	MEDIUM	not contact info
Person	EmailAddress	EmailAddress	nvarchar(50)	Contact Info	Confidential	20	MEDIUM	not contact info
dbo	ErrorLog	UserName	sysname	Credentials	Confidential	20	MEDIUM	not contact info
Person	Password	PasswordHash	varchar(128)	Credentials	Confidential	20	MEDIUM	not contact info
Person	Password	PasswordSalt	varchar(10)	Credentials	Confidential	20	MEDIUM	not contact info
Sales	CreditCard	CreditCardID	int	Credit Card	Confidential	20	MEDIUM	not contact info
Sales	CreditCard	CardType	nvarchar(50)	Credit Card	Confidential	20	MEDIUM	not contact info
Sales	CreditCard	CardNumber	nvarchar(25)	Credit Card	Confidential	20	MEDIUM	not contact info
Sales	CreditCard	ExpYear	smallint	Credit Card	Confidential	20	MEDIUM	not contact info
Sales	PersonCreditCard	CreditCardID	int	Credit Card	Confidential	20	MEDIUM	not contact info
Sales	SalesOrderHeader	CreditCardID	int	Credit Card	Confidential	20	MEDIUM	not contact info
Sales	SalesOrderHeader	CreditCardApprovalCode	varchar(15)	Credit Card	Confidential	20	MEDIUM	not contact info
HumanResources	Employee	BirthDate	date	Date Of Birth	Confidential - GDPR	20	MEDIUM	not contact info

我很困惑：

你怎么看？

Answer 1

根据

exec sp_describe_first_result_set N'select * from sys.sensitivity_classifications sc'

information_type 是一个 sql_variant。这不符合 docs 但不应该成为问题，因为兼容类型已转换为 sql_variant 以进行比较而不会出现问题。乙

select case when  cast(N'Contact Info' as sql_variant) = 'Contact Info' then 1 else 0 end

returns

1

所以这可能是一个错误。您可以通过显式转换为 nvarchar 来变通。例如

SELECT 
    schema_name(O.schema_id) AS schema_name,
    O.NAME AS table_name,
    C.NAME AS column_name,
         [Type] = 
    CASE 
      WHEN ct.[name] IN ('varchar', 'char') THEN ct.[name] + '(' + IIF(c.max_length = -1, 'max', CAST(c.max_length AS VARCHAR(25))) + ')' 
      WHEN ct.[name] IN ('nvarchar','nchar') THEN ct.[name] + '(' + IIF(c.max_length = -1, 'max', CAST(c.max_length / 2 AS VARCHAR(25)))+ ')'      
      WHEN ct.[name] IN ('decimal', 'numeric') THEN ct.[name] + '(' + CAST(c.precision AS VARCHAR(25)) + ', ' + CAST(c.scale AS VARCHAR(25)) + ')'
      WHEN ct.[name] IN ('datetime2') THEN ct.[name] + '(' + CAST(c.scale AS VARCHAR(25)) + ')'
      ELSE ct.[name]
    END,
    information_type,
    label,
    rank,
    rank_desc,
    CAST(
             CASE
                  WHEN cast(information_type as nvarchar(200)) = 'Contact Info' 
                     THEN 'contact info'
                  ELSE 'not contact info'
             END AS varchar(max)) as Checking_Content
FROM sys.sensitivity_classifications sc
    JOIN sys.objects O
    ON  sc.major_id = O.object_id
    JOIN sys.columns C 
    ON  sc.major_id = C.object_id  AND sc.minor_id = C.column_id
    JOIN sys.types ct ON C.user_type_id = ct.user_type_id
    --AND EP.minor_id = C.column_id
    order by information_type

T-SQL :: information_type 未在 SQL Server 2019 中解析

T-SQL :: information_type is not parsed in SQL Server 2019

sql

sql-server

tsql

data-masking

dynamic-data-masking