638 CHAPTER 28 FROM DATABASES TO DATATYPES (Web hosts)
638 CHAPTER 28 FROM DATABASES TO DATATYPES NUMERIC (P,(S)) The NUMERIC datatype can store numbers with up to 1,000 digits of precision, and will perform calculations exactly. It is normally recommended that you store monetary values in this datatype, though you should be aware that arithmetic on NUMERIC values may be slow relative to the other numeric types. The NUMERIC type can be declared with an optional precision and scale. The precision is the total number of digits on both sides of the decimal, whereas the scale is the total number of digits to the right of the decimal. For example, the value 123.45 would be represented as numeric(5,2). If you attempt to insert a number that is too large, PostgreSQL rounds the number to a legal value for INSERT; for example, inserting 123.456 into the above representation would be stored as 123.46. You should also be aware that you can declare NUMERIC without a scale, which implies 0 for the scale, or without either a scale or precision, which implies no limits, although this latter practice is not recommended, for performance reasons. The storage requirement for NUMERIC varies depending on the number being stored. The basic formula is 4 bytes for a variable-length header, 4 bytes for a numeric header, and 2 bytes for every 4 decimal digits, but even 0 requires 8 bytes. Note The DECIMAL(P,(S)) type is equivalent to the NUMERIC type. REAL The REAL datatype is an inexact, variable-precision, floating-point number. On most platforms it has a range of at least 1E-37 to 1E+37 and supports a precision of at least six decimal places. Using the NUMERIC datatype rather than REAL is usually recommended, because REAL stores numbers in an inexact, though standards-compliant, way. The storage requirement is 4 bytes. DOUBLE PRECISION The DOUBLE PRECISION datatype is a variable-precision, floating-point number, supporting a range of around 11E-307to 1E+308 and a precision of at least 15 digits. The storage requirement is 8 bytes. FLOAT [(p)] The FLOAT datatype is a SQL-standard notation for specifying inexact numeric types. FLOAT takes an optional argument, (p), which signifies the minimum acceptable precision in binary digits. PostgreSQL interprets FLOAT(1)to FLOAT(24) as the REAL datatype, and FLOAT(25) to FLOAT(53) as the DOUBLEPRECISION datatype. Note REAL, DOUBLE PRECISION, and FLOAT also accept three special values in addition to ordinary numeric values. Those values represent the IEEE 754 special values of ‘Infinity’, ‘ Infinity’ (negative infinity), and ‘NaN’ (not a number). On input, these strings are recognized in a case-insensitive manner.
Looking for affordable and reliable webhost to host and run your business application? Then look no more and go to servlet web hosting services.