Date: Mon, 14 May 2001 00:20:18 -0700 From: Dima Dorfman <dima@unixfreak.org> To: audit@freebsd.org Cc: Bruce Evans <bde@zeta.org.au>, Warner Losh <imp@harmony.village.org> Subject: Re: jot(1) patch Message-ID: <20010514072018.7A3BF3E28@bazooka.unixfreak.org>
next in thread | raw e-mail | index | archive | help
--------
Hi folks,
Anyone care to commit this or should I? Attached is the latest
version that doesn't break K&R support.
Thanks!
Dima Dorfman
dima@unixfreak.org
Dima Dorfman <dima@spike.unixfreak.org> writes:
> Bruce Evans <bde@zeta.org.au> writes:
> > On Wed, 9 May 2001, Warner Losh wrote:
> >
> > > In message <20010509210433.1BDA13E28@bazooka.unixfreak.org> Dima Dorfman
> wr
> > ites:
> > > : - use getopt
> > > : - de-register
> > >
> > > You should de __P() as well. The changes add a prototype
> > >
> > > : -usage()
> > > : +usage(void)
> > >
> > > which breaks K&R compatibility. So do one or the other. The one
> > > thing worse than old K&R code in the tree is broken K&R code in the
> > > tree.
> >
> > You should just not gratuitously break usage(). usage() is already
> > declared to take a void in its forward declaration using __P(()),
> > so using "void" in its definition just breaks K&R support.
>
> Fair enough. The attached patch is like the first one I sent, but
> doesn't include the change to break K&R support.
>
> If there's nothing wrong with this could someone please commit it?
>
> Thanks,
>
> Dima Dorfman
> dima@unixfreak.org
Index: jot.c
===================================================================
RCS file: /st/src/FreeBSD/src/usr.bin/jot/jot.c,v
retrieving revision 1.14
diff -u -r1.14 jot.c
--- jot.c 2000/07/10 05:57:29 1.14
+++ jot.c 2001/05/14 07:18:20
@@ -65,7 +65,7 @@
#define ENDER_DEF 100
#define STEP_DEF 1
-#define isdefault(s) (strcmp((s), "-") == 0)
+#define is_default(s) (strcmp((s), "-") == 0)
double begin;
double ender;
@@ -83,11 +83,10 @@
char *sepstring = "\n";
char format[BUFSIZ];
-void getargs __P((int, char *[]));
-void getformat __P((void));
+void getformat __P((void));
int getprec __P((char *));
-int putdata __P((double, long));
-static void usage __P((void));
+int putdata __P((double, long));
+static void usage __P((void));
int
main(argc, argv)
@@ -96,37 +95,15 @@
{
double xd, yd;
long id;
- register double *x = &xd;
- register double *y = &yd;
- register long *i = &id;
+ double *x = &xd;
+ double *y = &yd;
+ long *i = &id;
+ unsigned int mask = 0;
+ int n = 0;
+ int ch;
- getargs(argc, argv);
- if (randomize) {
- *x = (ender - begin) * (ender > begin ? 1 : -1);
- for (*i = 1; *i <= reps || infinity; (*i)++) {
- *y = (double) arc4random() / ULONG_MAX;
- if (putdata(*y * *x + begin, reps - *i))
- errx(1, "range error in conversion");
- }
- } else
- for (*i = 1, *x = begin; *i <= reps || infinity; (*i)++, *x += s)
- if (putdata(*x, reps - *i))
- errx(1, "range error in conversion");
- if (!nofinalnl)
- putchar('\n');
- exit(0);
-}
-
-void
-getargs(ac, av)
- int ac;
- char *av[];
-{
- register unsigned int mask = 0;
- register int n = 0;
-
- while (--ac && **++av == '-' && !isdefault(*av))
- switch ((*av)[1]) {
+ while ((ch = getopt(argc, argv, "rb:w:cs:np:")) != -1)
+ switch ((char)ch) {
case 'r':
randomize = 1;
break;
@@ -138,72 +115,63 @@
break;
case 'b':
boring = 1;
+ /* FALLTHROUGH */
case 'w':
- if ((*av)[2])
- strcpy(format, *av + 2);
- else if (!--ac)
- errx(1, "need context word after -w or -b");
- else
- strcpy(format, *++av);
+ if (strlcpy(format, optarg, sizeof(format)) >=
+ sizeof(format))
+ errx(1, "-%c word too long", ch);
break;
case 's':
- if ((*av)[2])
- sepstring = *av + 2;
- else if (!--ac)
- errx(1, "need string after -s");
- else
- sepstring = *++av;
+ sepstring = optarg;
break;
case 'p':
- if ((*av)[2])
- prec = atoi(*av + 2);
- else if (!--ac)
- errx(1, "need number after -p");
- else
- prec = atoi(*++av);
+ prec = atoi(optarg);
if (prec <= 0)
errx(1, "bad precision value");
break;
default:
usage();
}
+ argc -= optind;
+ argv += optind;
- switch (ac) { /* examine args right to left, falling thru cases */
+ switch (argc) { /* examine args right to left, falling thru cases */
case 4:
- if (!isdefault(av[3])) {
- if (!sscanf(av[3], "%lf", &s))
- errx(1, "bad s value: %s", av[3]);
+ if (!is_default(argv[3])) {
+ if (!sscanf(argv[3], "%lf", &s))
+ errx(1, "bad s value: %s", argv[3]);
mask |= 01;
}
case 3:
- if (!isdefault(av[2])) {
- if (!sscanf(av[2], "%lf", &ender))
- ender = av[2][strlen(av[2])-1];
+ if (!is_default(argv[2])) {
+ if (!sscanf(argv[2], "%lf", &ender))
+ ender = argv[2][strlen(argv[2])-1];
mask |= 02;
if (!prec)
- n = getprec(av[2]);
+ n = getprec(argv[2]);
}
case 2:
- if (!isdefault(av[1])) {
- if (!sscanf(av[1], "%lf", &begin))
- begin = av[1][strlen(av[1])-1];
+ if (!is_default(argv[1])) {
+ if (!sscanf(argv[1], "%lf", &begin))
+ begin = argv[1][strlen(argv[1])-1];
mask |= 04;
if (!prec)
- prec = getprec(av[1]);
+ prec = getprec(argv[1]);
if (n > prec) /* maximum precision */
prec = n;
}
case 1:
- if (!isdefault(av[0])) {
- if (!sscanf(av[0], "%ld", &reps))
- errx(1, "bad reps value: %s", av[0]);
+ if (!is_default(argv[0])) {
+ if (!sscanf(argv[0], "%ld", &reps))
+ errx(1, "bad reps value: %s", argv[0]);
mask |= 010;
}
break;
case 0:
usage();
default:
- errx(1, "too many arguments. What do you mean by %s?", av[4]);
+ errx(1, "too many arguments. What do you mean by %s?",
+ argv[4]);
}
getformat();
while (mask) /* 4 bit mask has 1's where last 4 args were given */
@@ -257,7 +225,7 @@
mask = 015;
break;
case 012:
- s = (randomize ? -1.0 : STEP_DEF);
+ s = (randomize ? time(NULL) : STEP_DEF);
mask = 013;
break;
case 013:
@@ -265,8 +233,7 @@
begin = BEGIN_DEF;
else if (reps == 0)
errx(1, "must specify begin if reps == 0");
- else
- begin = ender - reps * s + s;
+ begin = ender - reps * s + s;
mask = 0;
break;
case 014:
@@ -306,6 +273,20 @@
}
if (reps == 0)
infinity = 1;
+ if (randomize) {
+ *x = (ender - begin) * (ender > begin ? 1 : -1);
+ for (*i = 1; *i <= reps || infinity; (*i)++) {
+ *y = (double) arc4random() / ULONG_MAX;
+ if (putdata(*y * *x + begin, reps - *i))
+ errx(1, "range error in conversion");
+ }
+ } else
+ for (*i = 1, *x = begin; *i <= reps || infinity; (*i)++, *x += s)
+ if (putdata(*x, reps - *i))
+ errx(1, "range error in conversion");
+ if (!nofinalnl)
+ putchar('\n');
+ exit(0);
}
int
@@ -358,8 +339,8 @@
getprec(s)
char *s;
{
- register char *p;
- register char *q;
+ char *p;
+ char *q;
for (p = s; *p; p++)
if (*p == '.')
@@ -375,8 +356,9 @@
void
getformat()
{
- register char *p;
+ char *p;
int dot, hash, space, sign, numbers = 0;
+ size_t sz;
char *s;
if (boring) /* no need to bother */
@@ -384,14 +366,19 @@
for (p = format; *p; p++) /* look for '%' */
if (*p == '%' && *(p+1) != '%') /* leave %% alone */
break;
- if (!*p && !chardata)
- sprintf(p, "%%.%df", prec);
- else if (!*p && chardata) {
- strcpy(p, "%c");
+ sz = sizeof(format) - strlen(format) - 1;
+ if (!*p && !chardata) {
+ if (snprintf(p, sz, "%%.%df", prec) >= (int)sz)
+ errx(1, "-w word too long");
+ } else if (!*p && chardata) {
+ if (strlcpy(p, "%c", sz) >= sz)
+ errx(1, "-w word too long");
intdata = 1;
- } else if (!*(p+1))
+ } else if (!*(p+1)) {
+ if (sz <= 0)
+ errx(1, "-w word too long");
strcat(format, "%"); /* cannot end in single '%' */
- else {
+ } else {
/*
* Allow conversion format specifiers of the form
* %[#][ ][{+,-}][0-9]*[.[0-9]*]? where ? must be one of
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-audit" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010514072018.7A3BF3E28>