Transact-SQL 数据类型转换

When expression is a date or time data type, style can be one of the values shown in the following table. Other values are processed as 0. . Beginning with SQL Server 2012, the only styles that are supported when converting from date and time types to datetimeoffset are 0 or 1. All other conversion styles return error 9809.

SQL Server supports the date format in Arabic style by using the Kuwaiti algorithm.

Without century (yy) (1)	With century (yyyy)	Standard	Input/Output (3)
-	0 or100(1, 2)	Default for datetime and smalldatetime	mon dd yyyy hh:miAM (or PM)
1	101	U.S.	1 = mm/dd/yy 101 = mm/dd/yyyy
2	102	ANSI	2 = yy.mm.dd 102 = yyyy.mm.dd
3	103	British/French	3 = dd/mm/yy 103 = dd/mm/yyyy
4	104	German	4 = dd.mm.yy 104 = dd.mm.yyyy
5	105	Italian	5 = dd-mm-yy 105 = dd-mm-yyyy
6	106 (1)	-	6 = dd mon yy 106 = dd mon yyyy
7	107 (1)	-	7 = Mon dd, yy 107 = Mon dd, yyyy
8	108	-	hh:mi:ss
-	9 or109(1, 2)	Default + milliseconds	mon dd yyyy hh:mi:ss:mmmAM (or PM)
10	110	USA	10 = mm-dd-yy 110 = mm-dd-yyyy
11	111	JAPAN	11 = yy/mm/dd 111 = yyyy/mm/dd
12	112	ISO	12 = yymmdd 112 = yyyymmdd
-	13 or113(1, 2)	Europe default + milliseconds	dd mon yyyy hh:mi:ss:mmm(24h)
14	114	-	hh:mi:ss:mmm(24h)
-	20 or120 (2)	ODBC canonical	yyyy-mm-dd hh:mi:ss(24h)
-	21 or121 (2)	ODBC canonical (with milliseconds) default for time, date, datetime2, and datetimeoffset	yyyy-mm-dd hh:mi:ss.mmm(24h)
-	126 (4)	ISO8601	yyyy-mm-ddThh:mi:ss.mmm (no spaces) Note When the value for milliseconds (mmm) is 0, the millisecond value is not displayed. For example, the value ‘2012-11-07T18:26:20.000 is displayed as ‘2012-11-07T18:26:20‘.
-	127(6, 7)	ISO8601 with time zone Z.	yyyy-mm-ddThh:mi:ss.mmmZ (no spaces) Note When the value for milliseconds (mmm) is 0, the milliseconds value is not displayed. For example, the value ‘2012-11-07T18:26:20.000 is displayed as ‘2012-11-07T18:26:20‘.
-	130(1, 2)	Hijri (5)	dd mon yyyy hh:mi:ss:mmmAM In this style, mon represents a multi-token Hijri unicode representation of the full month‘s name. This value will not render correctly on a default US installation of SSMS.
-	131 (2)	Hijri (5)	dd/mm/yyyy hh:mi:ss:mmmAM

1 These style values return nondeterministic results. Includes all (yy) (without century) styles and a subset of (yyyy) (with century) styles.

2 The default values (style 0 or 100, 9 or 109, 13 or 113, 20 or 120, and 21 or 121) always return the century (yyyy).

3 Input when you convert to datetime; output when you convert to character data.

4 Designed for XML use. For conversion from datetime or smalldatetime to character data, the output format is as described in the previous table.

5 Hijri is a calendar system with several variations. SQL Server uses the Kuwaiti algorithm.

Important

By default, SQL Server interprets two-digit years based on a cutoff year of 2049. That is, the two-digit year 49 is interpreted as 2049 and the two-digit year 50 is interpreted as 1950. Many client applications, such as those based on Automation objects, use a cutoff year of 2030. SQL Server provides the two digit year cutoff configuration option that changes the cutoff year used by SQL Server and allows for the consistent treatment of dates. We recommend specifying four-digit years.

6 Only supported when casting from character data to datetime or smalldatetime. When character data that represents only date or only time components is cast to the datetime or smalldatetime data types, the unspecified time component is set to 00:00:00.000, and the unspecified date component is set to 1900-01-01.

7The optional time zone indicator, Z, is used to make it easier to map XML datetime values that have time zone information to SQL Server datetime values that have no time zone. Z is the indicator for time zone UTC-0. Other time zones are indicated with HH:MM offset in the + or - direction. For example: 2006-12-12T23:45:12-08:00.

When you convert to character data from smalldatetime, the styles that include seconds or milliseconds show zeros in these positions. You can truncate unwanted date parts when you convert from datetime or smalldatetime values by using an appropriate char or varchar data type length.

When you convert to datetimeoffset from character data with a style that includes a time, a time zone offset is appended to the result.

When expression is float or real, style can be one of the values shown in the following table. Other values are processed as 0.

When expression is money or smallmoney, style can be one of the values shown in the following table. Other values are processed as 0.

When expression is xml, style can be one of the values shown in the following table. Other values are processed as 0.

Value	Output
0(default)	Use default parsing behavior that discards insignificant white space and does not allow for an internal DTD subset. Note When you convert to the xml data type, SQL Server insignificant white space is handled differently than in XML 1.0. For more information, see Create Instances of XML Data.
1	Preserve insignificant white space. This style setting sets the default xml:space handling to behave the same as ifxml:space="preserve" has been specified instead.
2	Enable limited internal DTD subset processing. If enabled, the server can use the following information that is provided in an internal DTD subset to perform nonvalidating parse operations. Defaults for attributes are applied. Internal entity references are resolved and expanded. The DTD content model will be checked for syntactical correctness. The parser will ignore external DTD subsets. It also does not evaluate the XML declaration to see whether the standalone attribute is set yes or no, but instead parses the XML instance as if it is a stand-alone document.
3	Preserve insignificant white space and enable limited internal DTD subset processing.

When expression is binary(n), varbinary(n), char(n), or varchar(n), style can be one of the values shown in the following table. Style values that are not listed in the table return an error.

Implicit conversions are those conversions that occur without specifying either the CAST or CONVERT function. Explicit conversions are those conversions that require the CAST or CONVERT function to be specified. The following illustration shows all explicit and implicit data type conversions that are allowed for SQL Server system-supplied data types. These include xml, bigint, and sql_variant. There is no implicit conversion on assignment from the sql_variant data type, but there is implicit conversion to sql_variant.

Tip
This chart is available as a downloadable PDF file at the Microsoft Download Center.

When you convert between datetimeoffset and the character types char, varchar, nchar, and nvarchar the converted time zone offset part should always be double digits for both HH and MM for example, -08:00.

Note
Because Unicode data always uses an even number of bytes, use caution when you convert binary or varbinary to or from Unicode supported data types. For example, the following conversion does not return a hexadecimal value of 41; it returns 4100: SELECT CAST(CAST(0x41 AS nvarchar) AS varbinary).

Large-value data types exhibit the same implicit and explicit conversion behavior as their smaller counterparts, specifically the varchar, nvarcharand varbinary data types. However, you should consider the following guidelines:

• Conversion from image to varbinary(max) and vice-versa is an implicit conversion, and so are conversions between text andvarchar(max), and ntext and nvarchar(max).

• Conversion from large-value data types, such as varchar(max), to a smaller counterpart data type, such as varchar, is an implicit conversion, but truncation will occur if the large value is too big for the specified length of the smaller data type.

• Conversion from varchar, nvarchar, or varbinary to their corresponding large-value data types is performed implicitly.

• Conversion from the sql_variant data type to the large-value data types is an explicit conversion.

• Large-value data types cannot be converted to the sql_variant data type.

For more information about how to convert from the xml data type, see Create Instances of XML Data.

When you explicitly or implicitly cast the xml data type to a string or binary data type, the content of the xml data type is serialized based on a set of rules. For information about these rules, see Define the Serialization of XML Data. For information about how to convert from other data types to the xml data type, see Create Instances of XML Data.

Automatic data type conversion is not supported for the text and image data types. You can explicitly convert text data to character data, andimage data to binary or varbinary, but the maximum length is 8000 bytes. If you try an incorrect conversion such as trying to convert a character expression that includes letters to an int, SQL Server returns an error message.

When the output of CAST or CONVERT is a character string, and the input is a character string, the output has the same collation and collation label as the input. If the input is not a character string, the output has the default collation of the database, and a collation label of coercible-default. For more information, see Collation Precedence (Transact-SQL).

To assign a different collation to the output, apply the COLLATE clause to the result expression of the CAST or CONVERT function. For example:

SELECT CAST(‘abc‘ AS varchar(5)) COLLATE French_CS_AS

When you convert character or binary expressions (char, nchar, nvarchar, varchar, binary, or varbinary) to an expression of a different data type, data can be truncated, only partially displayed, or an error is returned because the result is too short to display. Conversions to char, varchar,nchar, nvarchar, binary, and varbinary are truncated, except for the conversions shown in the following table.

* = Result length too short to display. E = Error returned because result length is too short to display.

SQL Server guarantees that only roundtrip conversions, conversions that convert a data type from its original data type and back again, will yield the same values from version to version. The following example shows such a roundtrip conversion:

DECLARE @myval decimal (5, 2);
SET @myval = 193.57;
SELECT CAST(CAST(@myval AS varbinary(20)) AS decimal(10,5));
-- Or, using CONVERT
SELECT CONVERT(decimal(10,5), CONVERT(varbinary(20), @myval));

Note
Do not try to construct binary values and then convert them to a data type of the numeric data type category. SQL Server does not guarantee that the result of a decimal or numeric data type conversion to binary will be the same between versions of SQL Server.

The following example shows a resulting expression that is too small to display.

USE AdventureWorks2012;
GO
SELECT p.FirstName, p.LastName, SUBSTRING(p.Title, 1, 25) AS Title, CAST(e.SickLeaveHours AS char(1)) AS [Sick Leave]
FROM HumanResources.Employee e JOIN Person.Person p ON e.BusinessEntityID = p.BusinessEntityID
WHERE NOT e.BusinessEntityID >5;

Here is the result set.

FirstName   LastName      Title   Sick Leave

---------   ------------- ------- --------

Ken         Sanchez       NULL   *

Terri       Duffy         NULL   *

Roberto     Tamburello    NULL   *

Rob         Walters       NULL   *

Gail        Erickson      Ms.    *

(5 row(s) affected)

When you convert data types that differ in decimal places, sometimes the result value is truncated and at other times it is rounded. The following table shows the behavior.

For example, the result of the following conversion is 10:

SELECT CAST(10.6496 AS int);

When you convert data types in which the target data type has fewer decimal places than the source data type, the value is rounded. For example, the result of the following conversion is $10.3497:

SELECT CAST(10.3496847 AS money);

SQL Server returns an error message when nonnumeric char, nchar, varchar, or nvarchar data is converted to int, float, numeric, or decimal. SQL Server also returns an error when an empty string (" ") is converted to numeric or decimal.

The following table lists the styles for which the string-to-datetime conversion is nondeterministic.

1 With the exception of styles 20 and 21

Beginning in SQL Server 2012, if you use supplementary character (SC) collations, a CAST operation from nchar or nvarchar to an nchar ornvarchar type of smaller length will not truncate inside a surrogate pair; it truncates before the supplementary character. For example, the following code fragment leaves @x holding just ‘ab‘. There is not enough space to hold the supplementary character.

DECLARE @x NVARCHAR(10) = ‘ab‘ + NCHAR(0x10000);
SELECT CAST (@x AS NVARCHAR(3));

When using SC collations the behavior of CONVERT, is analogous to that of CAST.

In earlier versions of SQL Server, the default style for CAST and CONVERT operations on time and datetime2 data types is 121 except when either type is used in a computed column expression. For computed columns, the default style is 0. This behavior impacts computed columns when they are created, used in queries involving auto-parameterization, or used in constraint definitions.

Under compatibility level 110 and higher, the default style for CAST and CONVERT operations on time and datetime2 data types is always 121. If your query relies on the old behavior, use a compatibility level less than 110, or explicitly specify the 0 style in the affected query.

Upgrading the database to compatibility level 110 and higher will not change user data that has been stored to disk. You must manually correct this data as appropriate. For example, if you used SELECT INTO to create a table from a source that contained a computed column expression described above, the data (using style 0) would be stored rather than the computed column definition itself. You would need to manually update this data to match style 121.

Each example retrieves the name of the product for those products that have a 3 in the first digit of their list price and converts their ListPrice toint.

-- Use CAST
USE AdventureWorks2012;
GO
SELECT SUBSTRING(Name, 1, 30) AS ProductName, ListPrice
FROM Production.Product
WHERE CAST(ListPrice AS int) LIKE ‘3%‘;
GO

-- Use CONVERT.
USE AdventureWorks2012;
GO
SELECT SUBSTRING(Name, 1, 30) AS ProductName, ListPrice
FROM Production.Product
WHERE CONVERT(int, ListPrice) LIKE ‘3%‘;
GO

The following example calculates a single column computation (Computed) by dividing the total year-to-date sales (SalesYTD) by the commission percentage (CommissionPCT). This result is converted to an int data type after being rounded to the nearest whole number.

USE AdventureWorks2012;
GO
SELECT CAST(ROUND(SalesYTD/CommissionPCT, 0) AS int) AS Computed
FROM Sales.SalesPerson 
WHERE CommissionPCT != 0;
GO

Here is the result set.

Computed

------

379753754

346698349

257144242

176493899

281101272

0

301872549

212623750

298948202

250784119

239246890

101664220

124511336

97688107

(14 row(s) affected)

The following example concatenates noncharacter, nonbinary expressions by using CAST.

USE AdventureWorks2012;
GO
SELECT ‘The list price is ‘ + CAST(ListPrice AS varchar(12)) AS ListPrice
FROM Production.Product
WHERE ListPrice BETWEEN 350.00 AND 400.00;
GO

Here is the result set.

ListPrice

------------------

The list price is 357.06

The list price is 364.09

The list price is 364.09

The list price is 364.09

The list price is 364.09

(5 row(s) affected)

The following example uses CAST in the select list to convert the Name column to a char(10) column.

USE AdventureWorks2012;
GO
SELECT DISTINCT CAST(p.Name AS char(10)) AS Name, s.UnitPrice
FROM Sales.SalesOrderDetail AS s 
JOIN Production.Product AS p 
    ON s.ProductID = p.ProductID
WHERE Name LIKE ‘Long-Sleeve Logo Jersey, M‘;
GO

Here is the result set.

Name       UnitPrice

---------- ---------------------

Long-Sleev 31.2437

Long-Sleev 32.4935

Long-Sleev 49.99

(3 row(s) affected)

The following example converts the money column SalesYTD to an int and then to a char(20) column so that it can be used with the LIKEclause.

USE AdventureWorks2012;
GO
SELECT p.FirstName, p.LastName, s.SalesYTD, s.BusinessEntityID
FROM Person.Person AS p 
JOIN Sales.SalesPerson AS s 
    ON p.BusinessEntityID = s.BusinessEntityID
WHERE CAST(CAST(s.SalesYTD AS int) AS char(20)) LIKE ‘2%‘;
GO

Here is the result set.

FirstName        LastName            SalesYTD         SalesPersonID

---------------- ------------------- ---------------- -------------

Tsvi             Reiter              2811012.7151      279

Syed             Abbas               219088.8836       288

Rachel           Valdez              2241204.0424      289

(3 row(s) affected)

The following are several examples that show using CONVERT to convert to typed XML by using the XML Data Type and Columns (SQL Server).

This example converts a string with white space, text and markup into typed XML and removes all insignificant white space (boundary white space between nodes):

CONVERT(XML, ‘<root><child/></root>‘)

This example converts a similar string with white space, text and markup into typed XML and preserves insignificant white space (boundary white space between nodes):

CONVERT(XML, ‘<root>          <child/>         </root>‘, 1)

This example casts a string with white space, text, and markup into typed XML:

CAST(‘<Name><FName>Carol</FName><LName>Elliot</LName></Name>‘  AS XML)

For more examples, see Create Instances of XML Data.

The following example displays the current date and time, uses CAST to change the current date and time to a character data type, and then usesCONVERT display the date and time in the ISO 8901 format.

SELECT 
   GETDATE() AS UnconvertedDateTime,
   CAST(GETDATE() AS nvarchar(30)) AS UsingCast,
   CONVERT(nvarchar(30), GETDATE(), 126) AS UsingConvertTo_ISO8601  ;
GO

Here is the result set.

UnconvertedDateTime     UsingCast                      UsingConvertTo_ISO8601

----------------------- ------------------------------ ------------------------------

2006-04-18 09:58:04.570 Apr 18 2006 9:58AM            2006-04-18T09:58:04.570

(1 row(s) affected)

The following example is approximately the opposite of the previous example. The example displays a date and time as character data, uses CASTto change the character data to the datetime data type, and then uses CONVERT to change the character data to the datetime data type.

SELECT 
   ‘2006-04-25T15:50:59.997‘ AS UnconvertedText,
   CAST(‘2006-04-25T15:50:59.997‘ AS datetime) AS UsingCast,
   CONVERT(datetime, ‘2006-04-25T15:50:59.997‘, 126) AS UsingConvertFrom_ISO8601 ;
GO

Here is the result set.

UnconvertedText         UsingCast               UsingConvertFrom_ISO8601

----------------------- ----------------------- ------------------------

2006-04-25T15:50:59.997 2006-04-25 15:50:59.997 2006-04-25 15:50:59.997

(1 row(s) affected)

The following examples show the results of converting binary and character data by using different styles.

--Convert the binary value 0x4E616d65 to a character value.
SELECT CONVERT(char(8), 0x4E616d65, 0) AS [Style 0, binary to character];

Here is the result set.

Style 0, binary to character

----------------------------

Name

(1 row(s) affected)

--The following example shows how Style 1 can force the result
--to be truncated. The truncation is caused by
--including the characters 0x in the result.
SELECT CONVERT(char(8), 0x4E616d65, 1) AS [Style 1, binary to character];

Here is the result set.

Style 1, binary to character

------------------------------

0x4E616D

(1 row(s) affected)

--The following example shows that Style 2 does not truncate the
--result because the characters 0x are not included in
--the result.
SELECT CONVERT(char(8), 0x4E616d65, 2) AS [Style 2, binary to character];

Here is the result set.

Style 2, binary to character

------------------------------

4E616D65

(1 row(s) affected)

--Convert the character value ‘Name‘ to a binary value.
SELECT CONVERT(binary(8), ‘Name‘, 0) AS [Style 0, character to binary];

Here is the result set.

Style 0, character to binary

----------------------------------

0x4E616D6500000000

(1 row(s) affected)

SELECT CONVERT(binary(4), ‘0x4E616D65‘, 1) AS [Style 1, character to binary];

Here is the result set.

Style 1, character to binary

----------------------------------

0x4E616D65

(1 row(s) affected)

SELECT CONVERT(binary(4), ‘4E616D65‘, 2) AS [Style 2, character to binary];

Here is the result set.

Style 2, character to binary

----------------------------------

0x4E616D65

(1 row(s) affected)

The following example demonstrates the conversion of date, time, and datetime data types.

DECLARE @d1 date, @t1 time, @dt1 datetime;
SET @d1 = GETDATE();
SET @t1 = GETDATE();
SET @dt1 = GETDATE();
SET @d1 = GETDATE();
-- When converting date to datetime the minutes portion becomes zero.
SELECT @d1 AS [date], CAST (@d1 AS datetime) AS [date as datetime];
-- When converting time to datetime the date portion becomes zero 
-- which converts to January 1, 1900.
SELECT @t1 AS [time], CAST (@t1 AS datetime) AS [time as datetime];
-- When converting datetime to date or time non-applicable portion is dropped.
SELECT @dt1 AS [datetime], CAST (@dt1 AS date) AS [datetime as date], CAST (@dt1 AS time) AS [datetime as time];