Provided by: libg2c-dev_1.6.4-3_amd64
NAME
reduce.c - Determines whether the number of groups should be increased in order to reduce the size of the large groups, and to make that adjustment.
SYNOPSIS
#include <stdlib.h> #include 'grib2.h' Typedefs typedef g2int integer Integer type. typedef g2float real Float type. Functions int reduce (integer *kfildo, integer *jmin, integer *jmax, integer *lbit, integer *nov, integer *lx, integer *ndg, integer *ibit, integer *jbit, integer *kbit, integer *novref, integer *ibxx2, integer *ier) Determines whether the number of groups should be increased in order to reduce the size of the large groups, and to make that adjustment.
Detailed Description
Determines whether the number of groups should be increased in order to reduce the size of the large groups, and to make that adjustment. Author Glahn Date November 2001 Definition in file reduce.c.
Typedef Documentation
typedef g2int integer Integer type. Definition at line 11 of file reduce.c. typedef g2float real Float type. Definition at line 12 of file reduce.c.
Function Documentation
int reduce (integer * kfildo, integer * jmin, integer * jmax, integer * lbit, integer * nov, integer * lx, integer * ndg, integer * ibit, integer * jbit, integer * kbit, integer * novref, integer * ibxx2, integer * ier) Determines whether the number of groups should be increased in order to reduce the size of the large groups, and to make that adjustment. By reducing the size of the large groups, less bits may be necessary for packing the group sizes and all the information about the groups. The reference for nov was removed in the calling routine so that kbit could be determined. This furnishes a starting point for the iterations in reduce. However, the reference must be considered. PROGRAM HISTORY LOG • November 2001 Glahn tdl grib2 • March 2002 Glahn comment ier = 715 • March 2002 Glahn modified to accommodate lx=1 on entry DATA SET USE • kfildo - unit number for output (print) file. (output) Parameters kfildo unit number for output (print) file. (input) jmin the minimum of each group (j=1,lx). it is possible after splitting the groups, jmin( ) will not be the minimum of the new group. this doesn't matter; jmin( ) is really the group reference and doesn't have to be the smallest value. (input/output) jmax the maximum of each group (j=1,lx). (input/output) lbit the number of bits necessary to pack each group (j=1,lx). (input/output) nov the number of values in each group (j=1,lx). (input/output) lx the number of groups. this will be increased if groups are split. (input/output) ndg the dimension of jmin( ), jmax( ), lbit( ), and nov( ). (input) ibit the number of bits necessary to pack the jmin(j) values, j=1,lx. (input) jbit the number of bits necessary to pack the lbit(j) values, j=1,lx. (input) kbit the number of bits necessary to pack the nov(j) values, j=1,lx. if the groups are split, kbit is reduced. (input/output) novref reference value for nov( ). (input) ibxx2 2**j (j=0,30). (input) ier error return. (output) • 0 = good return. • 714 = problem in algorithm. reduce aborted. • 715 = ngp not large enough. reduce aborted. Returns always returns 0, see parameter ier for error code. ntotbt(j) = the total bits used for the packing bits j (j=1,30). (internal) nboxj(j) = new boxes needed for the packing bits j (j=1,30). (internal) newbox(l) = number of new boxes (groups) for each original group (l=1,lx) for the current j. (automatic) (internal) newboxp(l) = same as newbox( ) but for the previous j. this eliminates recomputation. (automatic) (internal) cfeed = contains the character representation of a printer form feed. (character) (internal) ifeed = contains the integer value of a printer form feed. (internal) iorigb = the original number of bits necessary for the group values. (internal) Author Glahn Date November 2001 Definition at line 84 of file reduce.c. Referenced by pack_gp().
Author
Generated automatically by Doxygen for NCEPLIBS-g2c from the source code.