664 lines
16 KiB
C
664 lines
16 KiB
C
/*-
|
|
* SPDX-License-Identifier: BSD-3-Clause
|
|
*
|
|
* Copyright (c) 1990, 1993
|
|
* The Regents of the University of California. All rights reserved.
|
|
*
|
|
* This code is derived from software contributed to Berkeley by
|
|
* Chris Torek.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
* 3. Neither the name of the University nor the names of its contributors
|
|
* may be used to endorse or promote products derived from this software
|
|
* without specific prior written permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
* SUCH DAMAGE.
|
|
*
|
|
* From: Id: vfscanf.c,v 1.13 1998/09/25 12:20:27 obrien Exp
|
|
* From: static char sccsid[] = "@(#)strtol.c 8.1 (Berkeley) 6/4/93";
|
|
* From: static char sccsid[] = "@(#)strtoul.c 8.1 (Berkeley) 6/4/93";
|
|
*/
|
|
|
|
#include <sys/cdefs.h>
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
#include <sys/param.h>
|
|
#include <sys/systm.h>
|
|
#include <sys/ctype.h>
|
|
#include <sys/limits.h>
|
|
#include <sys/stddef.h>
|
|
|
|
/*
|
|
* Note that stdarg.h and the ANSI style va_start macro is used for both
|
|
* ANSI and traditional C compilers.
|
|
*/
|
|
#include <machine/stdarg.h>
|
|
|
|
#define BUF 32 /* Maximum length of numeric string. */
|
|
|
|
/*
|
|
* Flags used during conversion.
|
|
*/
|
|
#define LONG 0x01 /* l: long or double */
|
|
#define SHORT 0x04 /* h: short */
|
|
#define SUPPRESS 0x08 /* suppress assignment */
|
|
#define POINTER 0x10 /* weird %p pointer (`fake hex') */
|
|
#define NOSKIP 0x20 /* do not skip blanks */
|
|
#define QUAD 0x400
|
|
#define INTMAXT 0x800 /* j: intmax_t */
|
|
#define PTRDIFFT 0x1000 /* t: ptrdiff_t */
|
|
#define SIZET 0x2000 /* z: size_t */
|
|
#define SHORTSHORT 0x4000 /** hh: char */
|
|
|
|
/*
|
|
* The following are used in numeric conversions only:
|
|
* SIGNOK, NDIGITS, DPTOK, and EXPOK are for floating point;
|
|
* SIGNOK, NDIGITS, PFXOK, and NZDIGITS are for integral.
|
|
*/
|
|
#define SIGNOK 0x40 /* +/- is (still) legal */
|
|
#define NDIGITS 0x80 /* no digits detected */
|
|
|
|
#define DPTOK 0x100 /* (float) decimal point is still legal */
|
|
#define EXPOK 0x200 /* (float) exponent (e+3, etc) still legal */
|
|
|
|
#define PFXOK 0x100 /* 0x prefix is (still) legal */
|
|
#define NZDIGITS 0x200 /* no zero digits detected */
|
|
|
|
/*
|
|
* Conversion types.
|
|
*/
|
|
#define CT_CHAR 0 /* %c conversion */
|
|
#define CT_CCL 1 /* %[...] conversion */
|
|
#define CT_STRING 2 /* %s conversion */
|
|
#define CT_INT 3 /* integer, i.e., strtoq or strtouq */
|
|
typedef u_quad_t (*ccfntype)(const char *, char **, int);
|
|
|
|
static const u_char *__sccl(char *, const u_char *);
|
|
|
|
int
|
|
sscanf(const char *ibuf, const char *fmt, ...)
|
|
{
|
|
va_list ap;
|
|
int ret;
|
|
|
|
va_start(ap, fmt);
|
|
ret = vsscanf(ibuf, fmt, ap);
|
|
va_end(ap);
|
|
return(ret);
|
|
}
|
|
|
|
int
|
|
vsscanf(const char *inp, char const *fmt0, va_list ap)
|
|
{
|
|
int inr;
|
|
const u_char *fmt = (const u_char *)fmt0;
|
|
int c; /* character from format, or conversion */
|
|
size_t width; /* field width, or 0 */
|
|
char *p; /* points into all kinds of strings */
|
|
int n; /* handy integer */
|
|
int flags; /* flags as defined above */
|
|
char *p0; /* saves original value of p when necessary */
|
|
int nassigned; /* number of fields assigned */
|
|
int nconversions; /* number of conversions */
|
|
int nread; /* number of characters consumed from fp */
|
|
int base; /* base argument to strtoq/strtouq */
|
|
ccfntype ccfn; /* conversion function (strtoq/strtouq) */
|
|
char ccltab[256]; /* character class table for %[...] */
|
|
char buf[BUF]; /* buffer for numeric conversions */
|
|
|
|
/* `basefix' is used to avoid `if' tests in the integer scanner */
|
|
static short basefix[17] =
|
|
{ 10, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 };
|
|
|
|
inr = strlen(inp);
|
|
|
|
nassigned = 0;
|
|
nconversions = 0;
|
|
nread = 0;
|
|
base = 0; /* XXX just to keep gcc happy */
|
|
ccfn = NULL; /* XXX just to keep gcc happy */
|
|
for (;;) {
|
|
c = *fmt++;
|
|
if (c == 0)
|
|
return (nassigned);
|
|
if (isspace(c)) {
|
|
while (inr > 0 && isspace(*inp))
|
|
nread++, inr--, inp++;
|
|
continue;
|
|
}
|
|
if (c != '%')
|
|
goto literal;
|
|
width = 0;
|
|
flags = 0;
|
|
/*
|
|
* switch on the format. continue if done;
|
|
* break once format type is derived.
|
|
*/
|
|
again: c = *fmt++;
|
|
switch (c) {
|
|
case '%':
|
|
literal:
|
|
if (inr <= 0)
|
|
goto input_failure;
|
|
if (*inp != c)
|
|
goto match_failure;
|
|
inr--, inp++;
|
|
nread++;
|
|
continue;
|
|
|
|
case '*':
|
|
flags |= SUPPRESS;
|
|
goto again;
|
|
case 'j':
|
|
flags |= INTMAXT;
|
|
goto again;
|
|
case 'l':
|
|
if (flags & LONG){
|
|
flags &= ~LONG;
|
|
flags |= QUAD;
|
|
} else {
|
|
flags |= LONG;
|
|
}
|
|
goto again;
|
|
case 'q':
|
|
flags |= QUAD;
|
|
goto again;
|
|
case 't':
|
|
flags |= PTRDIFFT;
|
|
goto again;
|
|
case 'z':
|
|
flags |= SIZET;
|
|
goto again;
|
|
case 'h':
|
|
if (flags & SHORT){
|
|
flags &= ~SHORT;
|
|
flags |= SHORTSHORT;
|
|
} else {
|
|
flags |= SHORT;
|
|
}
|
|
goto again;
|
|
|
|
case '0': case '1': case '2': case '3': case '4':
|
|
case '5': case '6': case '7': case '8': case '9':
|
|
width = width * 10 + c - '0';
|
|
goto again;
|
|
|
|
/*
|
|
* Conversions.
|
|
*
|
|
*/
|
|
case 'd':
|
|
c = CT_INT;
|
|
ccfn = (ccfntype)strtoq;
|
|
base = 10;
|
|
break;
|
|
|
|
case 'i':
|
|
c = CT_INT;
|
|
ccfn = (ccfntype)strtoq;
|
|
base = 0;
|
|
break;
|
|
|
|
case 'o':
|
|
c = CT_INT;
|
|
ccfn = strtouq;
|
|
base = 8;
|
|
break;
|
|
|
|
case 'u':
|
|
c = CT_INT;
|
|
ccfn = strtouq;
|
|
base = 10;
|
|
break;
|
|
|
|
case 'x':
|
|
flags |= PFXOK; /* enable 0x prefixing */
|
|
c = CT_INT;
|
|
ccfn = strtouq;
|
|
base = 16;
|
|
break;
|
|
|
|
case 's':
|
|
c = CT_STRING;
|
|
break;
|
|
|
|
case '[':
|
|
fmt = __sccl(ccltab, fmt);
|
|
flags |= NOSKIP;
|
|
c = CT_CCL;
|
|
break;
|
|
|
|
case 'c':
|
|
flags |= NOSKIP;
|
|
c = CT_CHAR;
|
|
break;
|
|
|
|
case 'p': /* pointer format is like hex */
|
|
flags |= POINTER | PFXOK;
|
|
c = CT_INT;
|
|
ccfn = strtouq;
|
|
base = 16;
|
|
break;
|
|
|
|
case 'n':
|
|
nconversions++;
|
|
if (flags & SUPPRESS) /* ??? */
|
|
continue;
|
|
if (flags & SHORTSHORT)
|
|
*va_arg(ap, char *) = nread;
|
|
else if (flags & SHORT)
|
|
*va_arg(ap, short *) = nread;
|
|
else if (flags & LONG)
|
|
*va_arg(ap, long *) = nread;
|
|
else if (flags & QUAD)
|
|
*va_arg(ap, quad_t *) = nread;
|
|
else if (flags & INTMAXT)
|
|
*va_arg(ap, intmax_t *) = nread;
|
|
else if (flags & SIZET)
|
|
*va_arg(ap, size_t *) = nread;
|
|
else if (flags & PTRDIFFT)
|
|
*va_arg(ap, ptrdiff_t *) = nread;
|
|
else
|
|
*va_arg(ap, int *) = nread;
|
|
continue;
|
|
}
|
|
|
|
/*
|
|
* We have a conversion that requires input.
|
|
*/
|
|
if (inr <= 0)
|
|
goto input_failure;
|
|
|
|
/*
|
|
* Consume leading white space, except for formats
|
|
* that suppress this.
|
|
*/
|
|
if ((flags & NOSKIP) == 0) {
|
|
while (isspace(*inp)) {
|
|
nread++;
|
|
if (--inr > 0)
|
|
inp++;
|
|
else
|
|
goto input_failure;
|
|
}
|
|
/*
|
|
* Note that there is at least one character in
|
|
* the buffer, so conversions that do not set NOSKIP
|
|
* can no longer result in an input failure.
|
|
*/
|
|
}
|
|
|
|
/*
|
|
* Do the conversion.
|
|
*/
|
|
switch (c) {
|
|
case CT_CHAR:
|
|
/* scan arbitrary characters (sets NOSKIP) */
|
|
if (width == 0)
|
|
width = 1;
|
|
if (flags & SUPPRESS) {
|
|
size_t sum = 0;
|
|
for (;;) {
|
|
if ((n = inr) < width) {
|
|
sum += n;
|
|
width -= n;
|
|
inp += n;
|
|
if (sum == 0)
|
|
goto input_failure;
|
|
break;
|
|
} else {
|
|
sum += width;
|
|
inr -= width;
|
|
inp += width;
|
|
break;
|
|
}
|
|
}
|
|
nread += sum;
|
|
} else {
|
|
bcopy(inp, va_arg(ap, char *), width);
|
|
inr -= width;
|
|
inp += width;
|
|
nread += width;
|
|
nassigned++;
|
|
}
|
|
nconversions++;
|
|
break;
|
|
|
|
case CT_CCL:
|
|
/* scan a (nonempty) character class (sets NOSKIP) */
|
|
if (width == 0)
|
|
width = (size_t)~0; /* `infinity' */
|
|
/* take only those things in the class */
|
|
if (flags & SUPPRESS) {
|
|
n = 0;
|
|
while (ccltab[(unsigned char)*inp]) {
|
|
n++, inr--, inp++;
|
|
if (--width == 0)
|
|
break;
|
|
if (inr <= 0) {
|
|
if (n == 0)
|
|
goto input_failure;
|
|
break;
|
|
}
|
|
}
|
|
if (n == 0)
|
|
goto match_failure;
|
|
} else {
|
|
p0 = p = va_arg(ap, char *);
|
|
while (ccltab[(unsigned char)*inp]) {
|
|
inr--;
|
|
*p++ = *inp++;
|
|
if (--width == 0)
|
|
break;
|
|
if (inr <= 0) {
|
|
if (p == p0)
|
|
goto input_failure;
|
|
break;
|
|
}
|
|
}
|
|
n = p - p0;
|
|
if (n == 0)
|
|
goto match_failure;
|
|
*p = 0;
|
|
nassigned++;
|
|
}
|
|
nread += n;
|
|
nconversions++;
|
|
break;
|
|
|
|
case CT_STRING:
|
|
/* like CCL, but zero-length string OK, & no NOSKIP */
|
|
if (width == 0)
|
|
width = (size_t)~0;
|
|
if (flags & SUPPRESS) {
|
|
n = 0;
|
|
while (!isspace(*inp)) {
|
|
n++, inr--, inp++;
|
|
if (--width == 0)
|
|
break;
|
|
if (inr <= 0)
|
|
break;
|
|
}
|
|
nread += n;
|
|
} else {
|
|
p0 = p = va_arg(ap, char *);
|
|
while (!isspace(*inp)) {
|
|
inr--;
|
|
*p++ = *inp++;
|
|
if (--width == 0)
|
|
break;
|
|
if (inr <= 0)
|
|
break;
|
|
}
|
|
*p = 0;
|
|
nread += p - p0;
|
|
nassigned++;
|
|
}
|
|
nconversions++;
|
|
continue;
|
|
|
|
case CT_INT:
|
|
/* scan an integer as if by strtoq/strtouq */
|
|
#ifdef hardway
|
|
if (width == 0 || width > sizeof(buf) - 1)
|
|
width = sizeof(buf) - 1;
|
|
#else
|
|
/* size_t is unsigned, hence this optimisation */
|
|
if (--width > sizeof(buf) - 2)
|
|
width = sizeof(buf) - 2;
|
|
width++;
|
|
#endif
|
|
flags |= SIGNOK | NDIGITS | NZDIGITS;
|
|
for (p = buf; width; width--) {
|
|
c = *inp;
|
|
/*
|
|
* Switch on the character; `goto ok'
|
|
* if we accept it as a part of number.
|
|
*/
|
|
switch (c) {
|
|
/*
|
|
* The digit 0 is always legal, but is
|
|
* special. For %i conversions, if no
|
|
* digits (zero or nonzero) have been
|
|
* scanned (only signs), we will have
|
|
* base==0. In that case, we should set
|
|
* it to 8 and enable 0x prefixing.
|
|
* Also, if we have not scanned zero digits
|
|
* before this, do not turn off prefixing
|
|
* (someone else will turn it off if we
|
|
* have scanned any nonzero digits).
|
|
*/
|
|
case '0':
|
|
if (base == 0) {
|
|
base = 8;
|
|
flags |= PFXOK;
|
|
}
|
|
if (flags & NZDIGITS)
|
|
flags &= ~(SIGNOK|NZDIGITS|NDIGITS);
|
|
else
|
|
flags &= ~(SIGNOK|PFXOK|NDIGITS);
|
|
goto ok;
|
|
|
|
/* 1 through 7 always legal */
|
|
case '1': case '2': case '3':
|
|
case '4': case '5': case '6': case '7':
|
|
base = basefix[base];
|
|
flags &= ~(SIGNOK | PFXOK | NDIGITS);
|
|
goto ok;
|
|
|
|
/* digits 8 and 9 ok iff decimal or hex */
|
|
case '8': case '9':
|
|
base = basefix[base];
|
|
if (base <= 8)
|
|
break; /* not legal here */
|
|
flags &= ~(SIGNOK | PFXOK | NDIGITS);
|
|
goto ok;
|
|
|
|
/* letters ok iff hex */
|
|
case 'A': case 'B': case 'C':
|
|
case 'D': case 'E': case 'F':
|
|
case 'a': case 'b': case 'c':
|
|
case 'd': case 'e': case 'f':
|
|
/* no need to fix base here */
|
|
if (base <= 10)
|
|
break; /* not legal here */
|
|
flags &= ~(SIGNOK | PFXOK | NDIGITS);
|
|
goto ok;
|
|
|
|
/* sign ok only as first character */
|
|
case '+': case '-':
|
|
if (flags & SIGNOK) {
|
|
flags &= ~SIGNOK;
|
|
goto ok;
|
|
}
|
|
break;
|
|
|
|
/* x ok iff flag still set & 2nd char */
|
|
case 'x': case 'X':
|
|
if (flags & PFXOK && p == buf + 1) {
|
|
base = 16; /* if %i */
|
|
flags &= ~PFXOK;
|
|
goto ok;
|
|
}
|
|
break;
|
|
}
|
|
|
|
/*
|
|
* If we got here, c is not a legal character
|
|
* for a number. Stop accumulating digits.
|
|
*/
|
|
break;
|
|
ok:
|
|
/*
|
|
* c is legal: store it and look at the next.
|
|
*/
|
|
*p++ = c;
|
|
if (--inr > 0)
|
|
inp++;
|
|
else
|
|
break; /* end of input */
|
|
}
|
|
/*
|
|
* If we had only a sign, it is no good; push
|
|
* back the sign. If the number ends in `x',
|
|
* it was [sign] '0' 'x', so push back the x
|
|
* and treat it as [sign] '0'.
|
|
*/
|
|
if (flags & NDIGITS) {
|
|
if (p > buf) {
|
|
inp--;
|
|
inr++;
|
|
}
|
|
goto match_failure;
|
|
}
|
|
c = ((u_char *)p)[-1];
|
|
if (c == 'x' || c == 'X') {
|
|
--p;
|
|
inp--;
|
|
inr++;
|
|
}
|
|
if ((flags & SUPPRESS) == 0) {
|
|
u_quad_t res;
|
|
|
|
*p = 0;
|
|
res = (*ccfn)(buf, (char **)NULL, base);
|
|
if (flags & POINTER)
|
|
*va_arg(ap, void **) =
|
|
(void *)(uintptr_t)res;
|
|
else if (flags & SHORTSHORT)
|
|
*va_arg(ap, char *) = res;
|
|
else if (flags & SHORT)
|
|
*va_arg(ap, short *) = res;
|
|
else if (flags & LONG)
|
|
*va_arg(ap, long *) = res;
|
|
else if (flags & QUAD)
|
|
*va_arg(ap, quad_t *) = res;
|
|
else if (flags & INTMAXT)
|
|
*va_arg(ap, intmax_t *) = res;
|
|
else if (flags & PTRDIFFT)
|
|
*va_arg(ap, ptrdiff_t *) = res;
|
|
else if (flags & SIZET)
|
|
*va_arg(ap, size_t *) = res;
|
|
else
|
|
*va_arg(ap, int *) = res;
|
|
nassigned++;
|
|
}
|
|
nread += p - buf;
|
|
nconversions++;
|
|
break;
|
|
}
|
|
}
|
|
input_failure:
|
|
return (nconversions != 0 ? nassigned : -1);
|
|
match_failure:
|
|
return (nassigned);
|
|
}
|
|
|
|
/*
|
|
* Fill in the given table from the scanset at the given format
|
|
* (just after `['). Return a pointer to the character past the
|
|
* closing `]'. The table has a 1 wherever characters should be
|
|
* considered part of the scanset.
|
|
*/
|
|
static const u_char *
|
|
__sccl(char *tab, const u_char *fmt)
|
|
{
|
|
int c, n, v;
|
|
|
|
/* first `clear' the whole table */
|
|
c = *fmt++; /* first char hat => negated scanset */
|
|
if (c == '^') {
|
|
v = 1; /* default => accept */
|
|
c = *fmt++; /* get new first char */
|
|
} else
|
|
v = 0; /* default => reject */
|
|
|
|
/* XXX: Will not work if sizeof(tab*) > sizeof(char) */
|
|
for (n = 0; n < 256; n++)
|
|
tab[n] = v; /* memset(tab, v, 256) */
|
|
|
|
if (c == 0)
|
|
return (fmt - 1);/* format ended before closing ] */
|
|
|
|
/*
|
|
* Now set the entries corresponding to the actual scanset
|
|
* to the opposite of the above.
|
|
*
|
|
* The first character may be ']' (or '-') without being special;
|
|
* the last character may be '-'.
|
|
*/
|
|
v = 1 - v;
|
|
for (;;) {
|
|
tab[c] = v; /* take character c */
|
|
doswitch:
|
|
n = *fmt++; /* and examine the next */
|
|
switch (n) {
|
|
case 0: /* format ended too soon */
|
|
return (fmt - 1);
|
|
|
|
case '-':
|
|
/*
|
|
* A scanset of the form
|
|
* [01+-]
|
|
* is defined as `the digit 0, the digit 1,
|
|
* the character +, the character -', but
|
|
* the effect of a scanset such as
|
|
* [a-zA-Z0-9]
|
|
* is implementation defined. The V7 Unix
|
|
* scanf treats `a-z' as `the letters a through
|
|
* z', but treats `a-a' as `the letter a, the
|
|
* character -, and the letter a'.
|
|
*
|
|
* For compatibility, the `-' is not considered
|
|
* to define a range if the character following
|
|
* it is either a close bracket (required by ANSI)
|
|
* or is not numerically greater than the character
|
|
* we just stored in the table (c).
|
|
*/
|
|
n = *fmt;
|
|
if (n == ']' || n < c) {
|
|
c = '-';
|
|
break; /* resume the for(;;) */
|
|
}
|
|
fmt++;
|
|
/* fill in the range */
|
|
do {
|
|
tab[++c] = v;
|
|
} while (c < n);
|
|
c = n;
|
|
/*
|
|
* Alas, the V7 Unix scanf also treats formats
|
|
* such as [a-c-e] as `the letters a through e'.
|
|
* This too is permitted by the standard....
|
|
*/
|
|
goto doswitch;
|
|
break;
|
|
|
|
case ']': /* end of scanset */
|
|
return (fmt);
|
|
|
|
default: /* just another character */
|
|
c = n;
|
|
break;
|
|
}
|
|
}
|
|
/* NOTREACHED */
|
|
}
|