Tài liệu Using Functions phần 2 doc

You use the POWER() function to get the value of a number raised to a specified power. The following example returns 8: SELECT POWER(2, 3); You use the ROUND() function to get the value of a number rounded or truncated to a specified length. The following example returns 1.23500, which is 1.23456 rounded to three decimal places: SELECT ROUND(1.23456, 3); The next example passes a non-zero number as the third parameter to ROUND(), which indicates that the number is to be truncated rather than rounded, as was done in the previous example: SELECT ROUND(1.23456, 3, 1); This example returns 1.23400, which is 1.23456 truncated to three decimal places. You use the SQUARE() function to get the square of a number. The following example returns 16.0: SELECT SQUARE(4); You use the SQRT() function to get the square root of a number. The following example returns 4.0: SELECT SQRT(16); Using String Functions The string functions allow you to manipulate strings. For example, you can replace specified characters in a string. Table 4.5 lists the string functions available in SQL Server. Table 4.5: STRING FUNCTIONS FUNCTION DESCRIPTION ASCII(charExpression) Returns the ASCII code for the leftmost character of charExpression. CHAR(intExpression) Returns the character that corresponds to the ASCII code specified by intExpression. CHARINDEX (charExpression1, Returns the position of the characters specified by Table 4.5: STRING FUNCTIONS FUNCTION DESCRIPTION charExpression2 [, start ]) charExpression1 in charExpression2, starting at the optional position specified by start. DIFFERENCE (charExpression1, charExpression2) Returns the difference between the SOUNDEX values of the two character expressions. You use the SOUNDEX code to evaluate the phonetic similarity of two strings. The returned value is between 0 and 4; 4 indicates that the two expressions are phonetically identical. LEFT(charExpression, intExpression) Returns the leftmost characters specified by intExprssion from charExpression. LEN(charExpression) Returns the number of characters in charExpression. LOWER(charExpression) Converts the characters in charExpression to lowercase and returns those characters. LTRIM(charExpression) Removes any leading spaces from the start of charExpression and returns the remaining characters. NCHAR(intExpression) Returns the Unicode character with the code specified by intExpression. PATINDEX('%pattern%', charExpression) Returns the starting position of the first occurrence of pattern in charExpression. If pattern is not found then zeros are returned. REPLACE (charExpression1, charExpression2, charExpression3) Replaces all occurrences of charExpression2 in charExpression1 with charExpression3. QUOTENAME ('charString' [ , 'quoteChar' ]) Returns a Unicode string with the delimiters specified by quoteChar added to make charString a valid delimited identifier. REPLICATE (charExpression, intExpression) Repeats charExpression a total of intExpression times. REVERSE(charExpression) Reverses the characters in charExpression and returns those characters. RIGHT(charExpression, intExpression) Returns the rightmost characters specified by intExprssion from charExpression. RTRIM(charExpression) Removes any trailing spaces from the end of charExpression and returns the remaining characters. SOUNDEX(charExpression) Returns the four-character SOUNDEX code. You use this code to evaluate the phonetic similarity of Table 4.5: STRING FUNCTIONS FUNCTION DESCRIPTION two strings. SPACE(intExpression) Returns a string of repeated spaces for a total specified by intExpression. STR(floatExpression [ , length [ , decimal ] ]) Converts the number specified by floatExpression to characters; length specifies the total number of characters you want to see (including digits and spaces, plus the positive or negative sign and decimal point); decimal specifies the number of digits to the right of the decimal point. The number is rounded if necessary. STUFF (charExpression1, start, length, charExpression2) Deletes characters from charExpression1, starting at the position specified by start for a total of length characters, and then inserts the characters specified by charExpression2. SUBSTRING(expression, start, length) Returns part of a character, binary, text, or image expression. UNICODE('nCharExpression') Returns the Unicode value for the first character of the nchar or nvarchar expression nCharExpression. UPPER(charExpression) Converts the characters in charExpression to uppercase and returns those characters. Let's consider examples that use some of the string functions. You use the ASCII() function to get the ASCII code for the leftmost character of the supplied character expression. The following example returns 65 and 97: SELECT ASCII('A'), ASCII('a'); You use the CHAR() function to get the character that corresponds to the ASCII code of the supplied integer expression. The following example returns A and a: SELECT CHAR(65), CHAR(97); You use the CHARINDEX() function to get the position of characters. The following example returns 16, which is the position where the word ten starts: SELECT CHARINDEX('ten', 'Four-score and ten years'); You use the DIFFERENCE() function to obtain the difference between the SOUNDEX values of two character expressions. The following example returns 4, indicating that Brown and Browne are phonetically identical: SELECT DIFFERENCE('Brown', 'Browne'); You use the LEFT() function to obtain the leftmost characters of a character expression. The following example returns Four-score, which are the 10 leftmost characters of Four- score and ten years: SELECT LEFT('Four-score and ten years', 10); You use the RIGHT() function to obtain the rightmost characters of a character expression. The following example returns years, which are the five rightmost characters of Four-score and ten years: SELECT RIGHT('Four-score and ten years', 5); You use the LEN() function to obtain the digits in a character expression. The following example returns 24: SELECT LEN('Four-score and ten years'); You use the LOWER() function to obtain the lowercase version of a character expression. The following example returns four-score and ten years: SELECT LOWER('FOUR-SCORE AND TEN YEARS'); You use the UPPER() function to obtain the uppercase version of a character expression. The following example returns FOUR-SCORE AND TEN YEARS: SELECT UPPER('four-score and ten years'); You use the LTRIM() and RTRIM() functions to remove any spaces from the left and right of a character expression. The following example returns FOUR-SCORE and AND TEN YEARS (spaces removed): SELECT LTRIM(' FOUR-SCORE'), RTRIM('AND TEN YEARS '); You use the STR() function to convert a numeric value to a string consisting of numbers. The first parameter is the number to convert, the second is the total number of characters you want in your string, and the third is the number of digits after the decimal point. The following example returns 123.46: SELECT STR(123.456, 6, 2); The number 123.456 is converted to a string of six characters, with two digits after the decimal point, and rounded. You use the STUFF() function to replace characters. The first parameter is the string you want to replace characters in, the second is the starting position, the third is the total number of characters, and the fourth is the set of characters to insert. The following example returns Five-score and ten: SELECT STUFF('Four-score and ten', 1, 4, 'Five'); Four is replaced with Five. You use the SUBSTRING() function to obtain part of a string. The first parameter is the string, the second is the starting position, and the third is the total number of characters. The following example returns Four: SELECT SUBSTRING('Four-score and ten', 1, 4); You use the UNICODE() function to obtain the Unicode value for the first character. The following example returns 65 and 97: SELECT UNICODE('A'), UNICODE('a'); Using Date and Time Functions The date and time functions allow you to manipulate dates and times. For example, you can add a number of days to a given date. Table 4.6 lists the date and time functions available in SQL Server. Table 4.6: DATE AND TIME FUNCTIONS FUNCTION DESCRIPTION DATEADD(interval, number, date) Returns a datetime that is the result of adding the specified number of interval units to date. Valid intervals include year, quarter, month, dayofyear, day, week, hour, minute, second, and millisecond. DATEDIFF (interval, startDate, endDate) Returns the difference between startDate and endDate, with the difference calculated in interval units (year, quarter, and so on). DATENAME(interval, date) Returns a character string that represents the name of interval part of date. Table 4.6: DATE AND TIME FUNCTIONS FUNCTION DESCRIPTION DATEPART(interval, date) Returns an integer that represents the interval part of date. DAY(date) Returns an integer that represents the day part of date. GETDATE() Returns a datetime containing the current system date. GETUTCDATE() Returns a datetime containing the current system date as UTC time (Universal Time Coordinate or Greenwich Mean Time). The UTC time is derived from the current local time and the system time-zone setting. MONTH(date) Returns an integer that represents the month part of date. YEAR(date) Returns an integer that represents the year part of date. Let's consider examples that use some of the date and time functions. You use the DATEADD() function to add a number of intervals to a date. The following example adds two days to the date 12/20/2003 and returns 2003-12-22 00:00:00.000: SELECT DATEADD(day, 2, '12/20/2003'); You use the DATEDIFF() function to obtain the difference between two dates. The following example obtains the difference between 12/20/2003 and 12/22/2003 in days and returns 2 days: SELECT DATEDIFF(day, '12/20/2003', '12/22/2003'); You use the DATENAME() method to obtain a character string that represents the interval part of a date. The following example gets the month name of 12/20/2003 and returns December: SELECT DATENAME(month, '12/20/2003'); You use the DATEPART() method to obtain an integer that represents the interval part of a date. The following example gets the month number of 12/20/2003 and returns 12: SELECT DATEPART(month, '12/20/2003'); You use the DAY() function to obtain an integer that represents the day part of a date. The following example gets the day number of 12/20/2003 and returns 20: SELECT DAY('12/20/2003'); You use the MONTH() function to obtain an integer that represents the month part of a date. The following example gets the month number of 12/20/2003 and returns 12: SELECT MONTH('12/20/2003'); You use the YEAR() function to obtain an integer that represents the year part of a date. The following example gets the year number of 12/20/2003 and returns 2003: SELECT YEAR('12/20/2003'); You use the GETDATE() function to obtain the current system date. The following example returns 2002-07-16 12:59:50.823: SELECT GETDATE(); You use the GETUTCDATE() function to obtain the current system date as UTC time. The following example returns 2002-07-16 20:02:18.123: SELECT GETUTCDATE(); Using System Functions The system functions allow you to manipulate and obtain information about values, objects, and settings in SQL Server. For example, you can convert a value in one type to another type. Table 4.7 lists some of the system functions available in SQL Server. Table 4.7: SYSTEM FUNCTIONS FUNCTION DESCRIPTION CONVERT(dataType expression [, style [(length )], ]) Converts the value in expression to the type specified by dataType. If you are converting to an nchar, nvarchar, char, varchar, binary, or varbinary type, you can also specify an optional length, which specifies the length of the new value. You can use the optional style when • Converting datetime or smalldatetime data to character data; style is the format for the date and time. • Converting float, real, money, or smallmoney data to character data; style is the string format for the number. You can look up the details for style option in the SQL Server Books Online Table 4.7: SYSTEM FUNCTIONS FUNCTION DESCRIPTION documentation. COALESCE(expression1 [ , . expressionN]) Returns the first non-null expression in the list of expressions. DATALENGTH(expression). Returns the number of bytes used to represent expression. @@ERROR Returns the error number for the last T-SQL statement that was executed. @@IDENTITY Returns the last inserted identity value. ISDATE(expression) Returns 1 when expression is a valid date, otherwise 0 is returned. ISNULL(expression, replacementValue) If expression is null, then replacementValue is returned, otherwise expression is returned. ISNUMERIC(expression). Returns 1 when expression is a valid number, otherwise 0 is returned. NEWID() Returns a unique value of the uniqueidentifier type. NULLIF(expression1, expression2) Returns a null if expression1 equals expression2. @@ROWCOUNT Returns the number of rows affected by the last T- SQL statement that was executed. @@TRANCOUNT Returns the number of active transactions for the currentconnection to the database. Let's consider examples that use some of the system functions. You use the CONVERT() function to convert a value from one type to another. The following example converts the number 123.456 to an nvarchar and returns 123.456: SELECT CONVERT(nvarchar, 123.456); You use the COALESCE() function to obtain the first non-null expression in a list. The following example returns 123.456: SELECT COALESCE(null, null, 123.456, null); You use the DATALENGTH() function to obtain the number of bytes used to represent an expression. The following example displays the number of bytes used to represent the value stored in the CompanyName column of the Customers table for the row where CustomerID equals ALFKI: SELECT DATALENGTH(CompanyName), CompanyName FROM Customers WHERE CustomerID = 'ALFKI'; This example returns 38 and Alfreds Futterkiste, which contains 19 letters. Each letter is stored in 2 bytes, and the 19-letters string therefore takes up 38 bytes (2 * 19). You use the ISDATE() function to determine if an expression is a valid date. ISDATE() returns 1 when the expression is a valid date, otherwise it returns 0. The following example returns 1 and 0: SELECT ISDATE('12/20/2004'), ISDATE(1234); You use the ISNUMERIC() function to determine if an expression is a valid number. The following example returns 1 and 0: SELECT ISNUMERIC(1234), ISNUMERIC('abc'); You use the ISNULL() function to replace a null value with another value. The following example returns 10 and 20: SELECT ISNULL(null, 10), ISNULL(20, 10); . difference between 12/ 20 /20 03 and 12/ 22/ 2003 in days and returns 2 days: SELECT DATEDIFF(day, ' 12/ 20 /20 03', ' 12/ 22/ 2003'); You use. adds two days to the date 12/ 20 /20 03 and returns 20 03- 12- 22 00:00:00.000: SELECT DATEADD(day, 2, ' 12/ 20 /20 03'); You use the DATEDIFF() function to