jpcpack.c File Reference

Pack up a data field into a JPEG2000 code stream. More...

#include <stdlib.h>
#include <math.h>
#include "grib2_int.h"

Go to the source code of this file.

Functions
void	jpcpack (float *fld, g2int width, g2int height, g2int *idrstmpl, unsigned char *cpack, g2int *lcpack)
	This subroutine packs up a data field into a JPEG2000 code stream. More...
static void	jpcpack_int (void *fld, int fld_is_double, g2int width, g2int height, g2int *idrstmpl, unsigned char *cpack, g2int *lcpack)
	This subroutine packs up a float or double array into a JPEG2000 code stream. More...
void	jpcpackd (double *fld, g2int width, g2int height, g2int *idrstmpl, unsigned char *cpack, g2int *lcpack)
	This subroutine packs up a data field into a JPEG2000 code stream. More...

Detailed Description

Pack up a data field into a JPEG2000 code stream.

Author

Stephen Gilbert

Date

2003-08-17

Program History Log

Date	Programmer	Comments
2003-08-17	Gilbert	Initial.
2004-11-92	Gilbert	Fixed bug packing a near constant field.
2004-07-19	Gilbert	If jpeg2000 encoding fails, try again with different encoder options.
2005-05-10	Gilbert	Imposed minimum size on cpack.
2022-08-12	Hartnett	Now handle doubles too.

Definition in file jpcpack.c.

Function Documentation

jpcpack()

void jpcpack	(	float *	fld,
		g2int	width,
		g2int	height,
		g2int *	idrstmpl,
		unsigned char *	cpack,
		g2int *	lcpack
	)

This subroutine packs up a data field into a JPEG2000 code stream.

After the data field is scaled, and the reference value is subtracted out, it is treated as a grayscale image and passed to a JPEG2000 encoder. It also fills in GRIB2 Data Representation Template 5.40 or 5.40000 with the appropriate values.

Parameters

fld	Pointer to the float or double data values to pack.
width	The number of points in the x direction.
height	The number of points in the y direction.
idrstmpl	Contains the array of values for Data Representation Template Table 5.40 or 5.40000. 0 Reference value - ignored on input, set by jpcpack routine. 1 Binary Scale Factor - used on input, unchanged by jpcpack routine. 2 Decimal Scale Factor - used on input, unchanged by jpcpack routine. 3 number of bits for each data value - ignored on input 4 Original field type - currently ignored on input Data values assumed to be reals. Set to 0 on output. 5 if 0 use lossless compression, if 1 use lossy compression. 6 Desired compression ratio, if idrstmpl[5]=1. Set to 255, if idrstmpl[5]=0.
cpack	A pointer that will get the packed data field. Must be allocated before this function is called. Pass the allocated size in the lcpack parameter.
lcpack	Pointer that gets the length of packed field in cpack. This must be set by the calling function to the size available in cpack.

Author

Stephen Gilbert, Ed Hartnett

Definition at line 253 of file jpcpack.c.

References jpcpack_int().

Referenced by g2_addfield().

jpcpack_int()

static void jpcpack_int ( void *  fld, int  fld_is_double, g2int  width, g2int  height, g2int *  idrstmpl, unsigned char *  cpack, g2int *  lcpack  )

static

This subroutine packs up a float or double array into a JPEG2000 code stream.

After the data field is scaled, and the reference value is subtracted out, it is treated as a grayscale image and passed to a JPEG2000 encoder. It also fills in GRIB2 Data Representation Template 5.40 or 5.40000 with the appropriate values.

Parameters

fld	Pointer to the float or double data values to pack.
fld_is_double	If non-zero, then fld points to array of doubles, otherwise an array of floats.
width	The number of points in the x direction.
height	The number of points in the y direction.
idrstmpl	Contains the array of values for Data Representation Template Table 5.40 or 5.40000. 0 Reference value - ignored on input, set by jpcpack routine. 1 Binary Scale Factor - used on input, unchanged by jpcpack routine. 2 Decimal Scale Factor - used on input, unchanged by jpcpack routine. 3 number of bits for each data value - ignored on input 4 Original field type - currently ignored on input Data values assumed to be reals. Set to 0 on output. 5 if 0 use lossless compression, if 1 use lossy compression. 6 Desired compression ratio, if idrstmpl[5]=1. Set to 255, if idrstmpl[5]=0.
cpack	A pointer that will get the packed data field. Must be allocated before this function is called. Pass the allocated size in the lcpack parameter.
lcpack	Pointer that gets the length of packed field in cpack. This must be set by the calling function to the size available in cpack.

Returns

N/A

Author

Stephen Gilbert, Ed Hartnett

Definition at line 57 of file jpcpack.c.

References ALOG2, enc_jpeg2000(), int_power(), LOG, mkieee(), and sbits().

Referenced by jpcpack(), and jpcpackd().

jpcpackd()

void jpcpackd	(	double *	fld,
		g2int	width,
		g2int	height,
		g2int *	idrstmpl,
		unsigned char *	cpack,
		g2int *	lcpack
	)

This subroutine packs up a data field into a JPEG2000 code stream.

After the data field is scaled, and the reference value is subtracted out, it is treated as a grayscale image and passed to a JPEG2000 encoder. It also fills in GRIB2 Data Representation Template 5.40 or 5.40000 with the appropriate values.

Parameters

fld	Pointer to the float or double data values to pack.
width	The number of points in the x direction.
height	The number of points in the y direction.
idrstmpl	Contains the array of values for Data Representation Template Table 5.40 or 5.40000. 0 Reference value - ignored on input, set by jpcpack routine. 1 Binary Scale Factor - used on input, unchanged by jpcpack routine. 2 Decimal Scale Factor - used on input, unchanged by jpcpack routine. 3 number of bits for each data value - ignored on input 4 Original field type - currently ignored on input Data values assumed to be reals. Set to 0 on output. 5 if 0 use lossless compression, if 1 use lossy compression. 6 Desired compression ratio, if idrstmpl[5]=1. Set to 255, if idrstmpl[5]=0.
cpack	A pointer that will get the packed data field. Must be allocated before this function is called. Pass the allocated size in the lcpack parameter.
lcpack	Pointer that gets the length of packed field in cpack. This must be set by the calling function to the size available in cpack.

Author

Ed Hartnett

Definition at line 294 of file jpcpack.c.

References jpcpack_int().