/* *********************************************************************************** *********************************************************************************** Sequential Evaluation Algorithm for Simultaneous and Concurrent Retrieval of Atmospheric Parameter Estimates (SEASCRAPE) Jet Propulsion Laboratory Copyright (c) 1995, California Institute of Technology. U. S. Sponsorship under NASA Contract NAS7-1270 is acknowledged. *********************************************************************************** *********************************************************************************** File: arithmetic_types.h Library: Environs Abstract: This file defines a set of arithmetic types (and related macros) that can be used to specify the precision of real and integral variables. These definitions are currently machine-independent but may become machine-dependent when source code that employs this file is ported to new environments. History: - Date - - Author - - Comment - CURRENT_DATE L. Sparks Initial version. RCS info: $Log$ *********************************************************************************** *********************************************************************************** */ /* Header dependencies: */ #include #ifndef COMPILATION #include "compilation.h" #endif HEADER_COMPILATION(arithmetic_types, "$Header$") /* Definition of macro used to verify file inclusion: */ #define ARITHMETIC_TYPES /* Definition of arithmetic types: */ typedef long int Int; /* Type for integers. */ typedef short int Integer2; /* Type for two byte (sixteen bit) integers. */ typedef long int Integer4; /* Type for four byte (thirty-two bit) integers. */ typedef unsigned long int U_Int; /* Type for unsigned integers. */ typedef short int U_Integer2; /* Type for two byte (sixteen bit) unsigned integers. */ typedef unsigned long int U_Integer4; /* Type for four byte (thirty-two bit) unsigned integers. */ typedef double Real; /* Type for real floating-point numbers. */ typedef float Float4; /* Type for four byte (thirty-two bit) real floating-point numbers. */ typedef double Float8; /* Type for eight byte (sixty-four bit) real floating-point numbers. */ /* Definition of field macros for reading input variables and writing output variables of the various arithmetic types defined above (to write an "Int", for example, one can use the format "%" XD which concatenates to form "%ld"): */ #define XD "ld" /* Conversion for reading and writing integer numbers of type "Int" or type "U_Int". */ #define X2D "hd" /* Conversion for reading and writing integer numbers of type "Integer2" or type "U_Integer2". */ #define X4D "ld" /* Conversion for reading and writing integer numbers of type "Integer4" or type "U_Integer4". */ #define XE "le" /* Conversion for reading and writing real numbers of type "Real" in exponential notation. */ #define X4E "e" /* Conversion for reading and writing real numbers of type "Float4" in exponential notation. */ #define X8E "le" /* Conversion for reading and writing real numbers of type "Float8" in exponential notation. */ #define XF "lf" /* Conversion for reading and writing real numbers of type "Real" in decimal notation. */ #define X4F "f" /* Conversion for reading and writing real numbers of type "Float4" in decimal notation. */ #define X8F "lf" /* Conversion for reading and writing real numbers of type "Float8" in decimal notation. */ #define XG "lg" /* Conversion for reading and writing real numbers of type "Real" in exponential or decimal notation, whichever is smaller. */ #define X4G "g" /* Conversion for reading and writing real numbers of type "Float4" in exponential or decimal notation, whichever is smaller. */ #define X8G "lg" /* Conversion for reading and writing real numbers of type "Float8" in exponential or decimal notation, whichever is smaller. */ /* Definition of constants (the first set should be coordinated with the definition of "Real"): */ #define MAX_NUMBER_DIGITS DBL_DIG #define BIG_NUMBER DBL_MAX #define SMALL_NUMBER DBL_MIN #define TOLERANCE (100. * DBL_EPSILON) #define MAX_FLOAT8_DIGITS DBL_DIG #define BIG_FLOAT8 DBL_MAX #define SMALL_FLOAT8 DBL_MIN #define TOLERANCE8 (100. * DBL_EPSILON) #define MAX_FLOAT4_DIGITS FLT_DIG #define BIG_FLOAT4 FLT_MAX #define SMALL_FLOAT4 FLT_MIN #define TOLERANCE4 (100. * FLT_EPSILON)