Saturday 7 July 2018

PARSE VS TRY_PARSE


What are the differences between PARSE and TRY_PARSE  [PARSE VS TRY_PARSE] ?

PARSE
TRY_PARSE
1
PARSE is a new feature in SQL Server 2012.

Returns the result of an expression, translated to the requested data type in SQL Server.
TRY_PARSE is a new feature in SQL Server 2012.

Returns the result of an expression, translated to the requested data type, or null if the cast fails in SQL Server.

Use TRY_PARSE only for converting from string to date/time and number types
2
SYNTAX:

PARSE (string_value AS data_type [ USING culture ] )

string_value
nvarchar(4000) value representing the formatted value to parse into the specified data type.

string_value must be a valid representation of the requested data type, or PARSE raises an error.

data_type
Literal value representing the data type requested for the result.

culture
Optional string that identifies the culture in which string_value is formatted.

SELECT name FROM sys.syslanguages
-    Provides the culture string_value

If the culture argument is not provided, then the language of the current session is used. This language is set either implicitly, or explicitly by using the SET LANGUAGE statement. culture accepts any culture supported by the .NET Framework; it is not limited to the languages explicitly supported by SQL Server. If the culture argument is not valid, PARSE raises an error
SYNTAX:

TRY_PARSE ( string_value AS data_type [ USING culture ] ) 

string_value
nvarchar(4000) value representing the formatted value to parse into the specified data type.

string_value must be a valid representation of the requested data type, or TRY PARSE returns NULL.

data_type
Literal value representing the data type requested for the result.

culture
Optional string that identifies the culture in which string_value is formatted.

SELECT name FROM sys.syslanguages
-    Provides the culture string_value


3
Return Types
Returns the result of the expression, translated to the requested data type.
Return Types
Returns the result of the expression, translated to the requested data type, or null if the cast fails.
4
EXAMPLES

SELECT PARSE('1' AS INT) AS Result;  --> 1
OUTPUT:
1

SELECT PARSE('1' AS INT USING 'en-US') AS Result;  --> 1

OUTPUT:
1

SELECT 
    CASE WHEN PARSE('Govind' AS decimal USING 'sr-Latn-CS') IS NULL 
        THEN 'True' 
        ELSE 'False' 
END 
AS Result; --> error

OUTPUT:
Error converting string value 'Govind' into data type numeric using culture 'sr-Latn-CS'

SELECT 
    CASE WHEN PARSE('1' AS decimal USING 'sr-Latn-CS') IS NULL 
        THEN 'True' 
        ELSE 'False' 
END 
AS Result; --> False

OUTPUT:
False

SELECT IIF(PARSE('255' AS TINYINT) IS NULL, 'True', 'False') AS Result;  --> False

OUTPUT:
False

SELECT IIF(PARSE('256' AS TINYINT) IS NULL, 'True', 'False') AS Result;  --> True

OUTPUT:
Error converting string value '256' into data type tinyint using culture

SELECT PARSE('Govindaraj Kanniappan' AS datetime2 USING 'en-US') AS Result;  --> NULL

OUTPUT:
Error converting string value 'Govindaraj Kanniappan' into data type datetime2 using culture 'en-US'.

SELECT PARSE('Govindaraj Kanniappan' AS VARCHAR USING 'en-US') AS Result;  --> error, invalid data type varchar in function PARSE.

OUTPUT:
Error, invalid data type varchar in function PARSE.


SELECT IIF(PARSE('01/01/2011' AS INT) IS NULL, 'True', 'False') AS Result;  --> error

OUTPUT:
Error converting string value '01/01/2011' into data type int using culture ''.

SELECT IIF(PARSE('01/01/2011' AS datetime2) IS NULL, 'True', 'False') AS Result;  --> False

OUTPUT:
False


SELECT PARSE('Monday, 13 December 2010' AS date USING 'en-US') AS Result      --> 2010-12-13

OUTPUT:
2010-12-13

SELECT PARSE('Monday, 13 December' AS date USING 'en-US') AS Result           --> error, due to invalid input

OUTPUT:
Error converting string value 'Monday, 13 December' into data type date using culture 'en-US'.


SELECT PARSE('$345,98' AS money) AS Result --> 34598.00                  

OUTPUT:
34598.00

SELECT PARSE('$345,98' AS money USING 'en-US') AS Result --> 34598.00                

OUTPUT:
34598.00

SELECT PARSE('€345,98' AS money USING 'en-US') AS Result --> error, invalid type money using culture 'en-US'

OUTPUT:
error, invalid type money using culture 'en-US


SELECT PARSE('€345,98' AS money USING 'de-DE') AS Result --> 345.98 

OUTPUT:

345.98

EXAMPLES

SELECT TRY_PARSE('1' AS INT) AS Result;  --> 1
OUTPUT:
1

SELECT TRY_PARSE('1' AS INT USING 'en-US') AS Result;  --> 1

OUTPUT:
1

SELECT 
    CASE WHEN TRY_PARSE('Govind' AS decimal USING 'sr-Latn-CS') IS NULL 
        THEN 'True' 
        ELSE 'False' 
END 
AS Result; --> True

OUTPUT:
True


SELECT 
    CASE WHEN TRY_PARSE('1' AS decimal USING 'sr-Latn-CS') IS NULL 
        THEN 'True' 
        ELSE 'False' 
END 
AS Result; --> False

OUTPUT:
False

SELECT IIF(TRY_PARSE('255' AS TINYINT) IS NULL, 'True', 'False') AS Result;  --> False

OUTPUT:
False

SELECT IIF(TRY_PARSE('256' AS TINYINT) IS NULL, 'True', 'False') AS Result;  --> True


OUTPUT:
True

SELECT TRY_PARSE('Govindaraj Kanniappan' AS datetime2 USING 'en-US') AS Result;  --> NULL

OUTPUT:
NULL


SELECT TRY_PARSE('Govindaraj Kanniappan' AS VARCHAR USING 'en-US') AS Result;  --> error, invalid data type varchar in function TRY_PARSE.

OUTPUT:
error, invalid data type varchar in function TRY_PARSE.


SELECT IIF(TRY_PARSE('01/01/2011' AS INT) IS NULL, 'True', 'False') AS Result;  --> True

OUTPUT:
True



SELECT IIF(TRY_PARSE('01/01/2011' AS datetime2) IS NULL, 'True', 'False') AS Result;  --> False

OUTPUT:
False



No comments:

Post a Comment