duchow@watnxt3.ucr.edu (John Duchowski) (05/26/91)
Hi,
I am a bit of a novice at this, so this is probably something simple,
nevertheless I can't seem to be able to get around it. The program below
reads a file which looks like this:
PellinBroca 770-810 1900V 20mW 3sINT
10cmFL-617OL
108
2
0
7.7448E-12
10
7.8324E-12
20
9.425E-12
30
8.753E-12
40
1.0755E-11
.
.
and converts it to a file which looks like this:
PellinBroca 770-810 1900V 20mW 3sINT
10cmFL-617OL
0.0000 0.0005
10.0000 0.0005
20.0000 0.0006
30.0000 0.0005
40.0000 0.0007
.
.
(normalized here, but a scaling factor can be used as well). Anyway, the
main problem is that in its present form (please see below) it runs ok on
a NeXT ('030, Mach 2.0, cc compiler) and on a Vax (8820, VMS 5.4, Vax C)
but not on my AT (DR DOS 5.0, bc++). At home the array registers seem to
be out of step, i.e., an extra point gets written in before the 0th point.
I have tried both: for(i=1; i<=n; i++) and for(i=0, i<n, i++) and only
the former (less elegant) version works. I don't understand why neither
the Vax nor the NeXT object to this, but my AT does. The second, lesser
problem is that of elegance. I can't seem to get around the following
warnings:
Borland C++ Version 2.0 Copyright (c) 1991 Borland International
hp2plot.c:
Warning hp2plot.c 73: Parameter 'argc' is never used in function input
Warning hp2plot.c 127: Parameter 'argc' is never used in function output
Warning hp2plot.c 127: Function should return a value in function output
When I remove argc from output, the file name then look weird (some binary
bits or something). When I try return(0) in output the values actually
get zeroed and when I declare output as void, I get a domain error (I think).
Could anyone please shed more light on this ? Any hints, comments, etc.,
will be greatly appreciated. Thank you !
- John Duchowski
----------------------------------------------------------------------------
/* Program hp2plot.c - convert HP900 series 300 files to PC format */
/* Add plotting routine */
/* John K. Duchowski, May 20, 1991 */
#include <stdio.h>
#include <math.h>
#define EOL '\n'
typedef struct {
char head[6][20];
} header_struct;
void read_header(fptr, header)
FILE *fptr;
header_struct *header;
{
char junk[20];
int i;
if(fptr)
{
for(i = 0; i < 6; i++)
fscanf(fptr,"%s", header->head[i]);
fscanf(fptr,"%s%s",junk,junk);
}
}
void write_header(fptr, header)
FILE *fptr;
header_struct *header;
{
int i;
if(fptr)
{
for(i = 0; i < 5; i++)
fprintf(fptr,"%s ", header->head[i]);
fprintf(fptr,"\n");
fprintf(fptr,"%s ",header->head[5]);
fprintf(fptr,"\n");
}
}
input(argc, argv, x, y, header)
int *argc;
char *argv[];
float *x;
float *y;
header_struct *header;
{
FILE *fptr;
char c;
int i = 1;
if((fptr = fopen(argv[1], "r")) == NULL)
{
fprintf(stderr,"Can't open file %s\n",argv[1]);
exit(-1);
}
read_header(fptr,header);
printf("\nReading points from the input file %s\n", argv[1]);
while((c = fgetc(fptr)) != EOF)
{
ungetc(c,fptr);
fscanf(fptr,"%f",&x[i]);
fscanf(fptr,"%e",&y[i]);
i++;
while((c = fgetc(fptr)) != EOL);
c = fgetc(fptr);
}
printf("\n");
fclose(fptr);
return(i); /* return the number of points read */
}
output(argc, argv, x, y, n, header)
int *argc;
char *argv[];
float *x;
float *y;
int n;
header_struct *header;
{
FILE *fptr;
int i, norm;
float high, scale;
if((fptr = fopen(argv[2], "w")) == NULL)
{
fprintf(stderr,"Can't open file %s\n", argv[2]);
exit(-1);
}
write_header(fptr,header);
printf("\nNormalize (max value = 1.0) ? (yes = 1) ");
scanf("%d", &norm);
if(norm == 1)
{
high = y[0];
for(i = 0; i < n; i++)
{
if(y[i] > high)
high = y[i];
}
for(i = 0; i < n; i++)
y[i] = y[i]/high;
}
else
{
printf("\nEnter the scaling factor: ");
scanf("%f", &scale);
for(i = 0; i < n; i++)
y[i] = y[i]*scale;
}
printf("\nWriting points to the output file %s\n", argv[2]);
for(i = 0; i < n; i++)
{
fprintf(fptr,"%10.4f %10.4f\n", x[i], y[i]);
fflush(fptr);
}
fclose(fptr);
}
main(argc, argv)
int argc;
char *argv[];
{
header_struct header;
float x[5000];
float y[5000];
int n;
if(argc != 3)
{
fprintf(stderr,"Usage: %s <infile> <outfile> \n", argv[0]);
exit(-1);
}
n = input(argc, argv, x, y, &header);
printf("%d points read\n",n);
output(argc, argv, x, y, n, &header);
return(0);
}ok@goanna.cs.rmit.oz.au (Richard A. O'Keefe) (05/27/91)
This is not the real problem, but in article <14724@ucrmath.ucr.edu>, duchow@watnxt3.ucr.edu (John Duchowski) writes: > input(argc, argv, x, y, header) > int *argc; <------ > char *argv[]; ... > output(argc, argv, x, y, n, header) > int *argc; <------ > char *argv[]; ... > main(argc, argv) > int argc; > char *argv[]; > { ... > n = input(argc, argv, x, y, &header); <------ ... ^^^^ > output(argc, argv, x, y, n, &header); <------ ^^^^ That suggests that the functions should have been declared void input(argc, argv, x, y, header) int argc; char *argv[]; ... void output(argc, argv, x, y, n, header) int argc; char *argv[]; ... and that there may be other problems lurking around that 'lint' (surely the NeXt has 'lint'?) would catch. Myself, I always write "char **argv;"; C doesn't really understand arrays as parameters, and using a notation that suggests that it does will only get you in trouble. -- I rejoiced that at least So-and-So could spell "hierarchical", but the _real_ explanation was that he couldn't spell "heir". -me