]>
git.saurik.com Git - apple/libc.git/blob - stdio/vfscanf.c
35bb094d4351ccc67edd8409cc024c58729864e1
2 * Copyright (c) 1999 Apple Computer, Inc. All rights reserved.
4 * @APPLE_LICENSE_HEADER_START@
6 * The contents of this file constitute Original Code as defined in and
7 * are subject to the Apple Public Source License Version 1.1 (the
8 * "License"). You may not use this file except in compliance with the
9 * License. Please obtain a copy of the License at
10 * http://www.apple.com/publicsource and read it before using this file.
12 * This Original Code and all software distributed under the License are
13 * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER
14 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
15 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
16 * FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT. Please see the
17 * License for the specific language governing rights and limitations
20 * @APPLE_LICENSE_HEADER_END@
23 * Copyright (c) 1990, 1993
24 * The Regents of the University of California. All rights reserved.
26 * This code is derived from software contributed to Berkeley by
29 * Redistribution and use in source and binary forms, with or without
30 * modification, are permitted provided that the following conditions
32 * 1. Redistributions of source code must retain the above copyright
33 * notice, this list of conditions and the following disclaimer.
34 * 2. Redistributions in binary form must reproduce the above copyright
35 * notice, this list of conditions and the following disclaimer in the
36 * documentation and/or other materials provided with the distribution.
37 * 3. All advertising materials mentioning features or use of this software
38 * must display the following acknowledgement:
39 * This product includes software developed by the University of
40 * California, Berkeley and its contributors.
41 * 4. Neither the name of the University nor the names of its contributors
42 * may be used to endorse or promote products derived from this software
43 * without specific prior written permission.
45 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
46 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
47 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
48 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
49 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
50 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
51 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
52 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
53 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
54 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
69 #define FLOATING_POINT
72 #define BUF 513 /* Maximum length of numeric string. */
75 * Flags used during conversion.
77 #define LONG 0x01 /* l: long or double */
78 #define LONGDBL 0x02 /* L: long double; unimplemented */
79 #define SHORT 0x04 /* h: short */
80 #define SUPPRESS 0x08 /* suppress assignment */
81 #define POINTER 0x10 /* weird %p pointer (`fake hex') */
82 #define NOSKIP 0x20 /* do not skip blanks */
86 * The following are used in numeric conversions only:
87 * SIGNOK, NDIGITS, DPTOK, and EXPOK are for floating point;
88 * SIGNOK, NDIGITS, PFXOK, and NZDIGITS are for integral.
90 #define SIGNOK 0x40 /* +/- is (still) legal */
91 #define NDIGITS 0x80 /* no digits detected */
93 #define DPTOK 0x100 /* (float) decimal point is still legal */
94 #define EXPOK 0x200 /* (float) exponent (e+3, etc) still legal */
96 #define PFXOK 0x100 /* 0x prefix is (still) legal */
97 #define NZDIGITS 0x200 /* no zero digits detected */
102 #define CT_CHAR 0 /* %c conversion */
103 #define CT_CCL 1 /* %[...] conversion */
104 #define CT_STRING 2 /* %s conversion */
105 #define CT_INT 3 /* integer, i.e., strtoq or strtouq */
106 #define CT_FLOAT 4 /* floating, i.e., strtod */
108 #define u_char unsigned char
109 #define u_long unsigned long
111 static u_char
*__sccl(char *, u_char
*);
117 __svfscanf(fp
, fmt0
, ap
)
122 register u_char
*fmt
= (u_char
*)fmt0
;
123 register int c
; /* character from format, or conversion */
124 register size_t width
; /* field width, or 0 */
125 register char *p
; /* points into all kinds of strings */
126 register int n
; /* handy integer */
127 register int flags
; /* flags as defined above */
128 register char *p0
; /* saves original value of p when necessary */
129 int nassigned
; /* number of fields assigned */
130 int nread
; /* number of characters consumed from fp */
131 int base
; /* base argument to strtoq/strtouq */
132 u_quad_t (*ccfn
)(); /* conversion function (strtoq/strtouq) */
133 char ccltab
[256]; /* character class table for %[...] */
134 char buf
[BUF
]; /* buffer for numeric conversions */
136 /* `basefix' is used to avoid `if' tests in the integer scanner */
137 static short basefix
[17] =
138 { 10, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 };
142 base
= 0; /* XXX just to keep gcc happy */
143 ccfn
= NULL
; /* XXX just to keep gcc happy */
150 if (fp
->_r
<= 0 && __srefill(fp
))
152 if (!isspace(*fp
->_p
))
154 nread
++, fp
->_r
--, fp
->_p
++;
163 * switch on the format. continue if done;
164 * break once format type is derived.
170 if (fp
->_r
<= 0 && __srefill(fp
))
191 if( sizeof(size_t) == sizeof(long) )
193 if( sizeof(size_t) == sizeof(quad_t
) )
203 case '0': case '1': case '2': case '3': case '4':
204 case '5': case '6': case '7': case '8': case '9':
205 width
= width
* 10 + c
- '0';
210 * Those marked `compat' are for 4.[123]BSD compatibility.
212 * (According to ANSI, E and X formats are supposed
213 * to the same as e and x. Sorry about that.)
215 case 'D': /* compat */
220 ccfn
= (u_quad_t (*)())strtoq
;
226 ccfn
= (u_quad_t (*)())strtoq
;
230 case 'O': /* compat */
245 case 'X': /* compat XXX */
249 flags
|= PFXOK
; /* enable 0x prefixing */
255 #ifdef FLOATING_POINT
256 case 'E': /* compat XXX */
257 case 'F': /* compat */
260 case 'e': case 'f': case 'g':
270 fmt
= __sccl(ccltab
, fmt
);
280 case 'p': /* pointer format is like hex */
281 flags
|= POINTER
| PFXOK
;
288 if (flags
& SUPPRESS
) /* ??? */
291 *va_arg(ap
, short *) = nread
;
292 else if (flags
& LONG
)
293 *va_arg(ap
, long *) = nread
;
294 else if (flags
& QUAD
)
295 *va_arg(ap
, quad_t
*) = nread
;
297 *va_arg(ap
, int *) = nread
;
301 * Disgusting backwards compatibility hacks. XXX
303 case '\0': /* compat */
306 default: /* compat */
310 ccfn
= (u_quad_t (*)())strtoq
;
316 * We have a conversion that requires input.
318 if (fp
->_r
<= 0 && __srefill(fp
))
322 * Consume leading white space, except for formats
323 * that suppress this.
325 if ((flags
& NOSKIP
) == 0) {
326 while (isspace(*fp
->_p
)) {
330 else if (__srefill(fp
))
334 * Note that there is at least one character in
335 * the buffer, so conversions that do not set NOSKIP
336 * ca no longer result in an input failure.
346 /* scan arbitrary characters (sets NOSKIP) */
349 if (flags
& SUPPRESS
) {
352 if ((n
= fp
->_r
) < width
) {
370 size_t r
= fread((void *)va_arg(ap
, char *), 1,
381 /* scan a (nonempty) character class (sets NOSKIP) */
383 width
= ~0; /* `infinity' */
384 /* take only those things in the class */
385 if (flags
& SUPPRESS
) {
387 while (ccltab
[*fp
->_p
]) {
388 n
++, fp
->_r
--, fp
->_p
++;
391 if (fp
->_r
<= 0 && __srefill(fp
)) {
400 p0
= p
= va_arg(ap
, char *);
401 while (ccltab
[*fp
->_p
]) {
406 if (fp
->_r
<= 0 && __srefill(fp
)) {
422 /* like CCL, but zero-length string OK, & no NOSKIP */
425 if (flags
& SUPPRESS
) {
427 while (!isspace(*fp
->_p
)) {
428 n
++, fp
->_r
--, fp
->_p
++;
431 if (fp
->_r
<= 0 && __srefill(fp
))
436 p0
= p
= va_arg(ap
, char *);
437 while (!isspace(*fp
->_p
)) {
442 if (fp
->_r
<= 0 && __srefill(fp
))
452 /* scan an integer as if by strtoq/strtouq */
454 if (width
== 0 || width
> sizeof(buf
) - 1)
455 width
= sizeof(buf
) - 1;
457 /* size_t is unsigned, hence this optimisation */
458 if (--width
> sizeof(buf
) - 2)
459 width
= sizeof(buf
) - 2;
462 flags
|= SIGNOK
| NDIGITS
| NZDIGITS
;
463 for (p
= buf
; width
; width
--) {
466 * Switch on the character; `goto ok'
467 * if we accept it as a part of number.
472 * The digit 0 is always legal, but is
473 * special. For %i conversions, if no
474 * digits (zero or nonzero) have been
475 * scanned (only signs), we will have
476 * base==0. In that case, we should set
477 * it to 8 and enable 0x prefixing.
478 * Also, if we have not scanned zero digits
479 * before this, do not turn off prefixing
480 * (someone else will turn it off if we
481 * have scanned any nonzero digits).
488 if (flags
& NZDIGITS
)
489 flags
&= ~(SIGNOK
|NZDIGITS
|NDIGITS
);
491 flags
&= ~(SIGNOK
|PFXOK
|NDIGITS
);
494 /* 1 through 7 always legal */
495 case '1': case '2': case '3':
496 case '4': case '5': case '6': case '7':
497 base
= basefix
[base
];
498 flags
&= ~(SIGNOK
| PFXOK
| NDIGITS
);
501 /* digits 8 and 9 ok iff decimal or hex */
503 base
= basefix
[base
];
505 break; /* not legal here */
506 flags
&= ~(SIGNOK
| PFXOK
| NDIGITS
);
509 /* letters ok iff hex */
510 case 'A': case 'B': case 'C':
511 case 'D': case 'E': case 'F':
512 case 'a': case 'b': case 'c':
513 case 'd': case 'e': case 'f':
514 /* no need to fix base here */
516 break; /* not legal here */
517 flags
&= ~(SIGNOK
| PFXOK
| NDIGITS
);
520 /* sign ok only as first character */
522 if (flags
& SIGNOK
) {
528 /* x ok iff flag still set & 2nd char */
530 if (flags
& PFXOK
&& p
== buf
+ 1) {
531 base
= 16; /* if %i */
539 * If we got here, c is not a legal character
540 * for a number. Stop accumulating digits.
545 * c is legal: store it and look at the next.
550 else if (__srefill(fp
))
554 * If we had only a sign, it is no good; push
555 * back the sign. If the number ends in `x',
556 * it was [sign] '0' 'x', so push back the x
557 * and treat it as [sign] '0'.
559 if (flags
& NDIGITS
) {
561 (void) ungetc(*(u_char
*)--p
, fp
);
564 c
= ((u_char
*)p
)[-1];
565 if (c
== 'x' || c
== 'X') {
567 (void) ungetc(c
, fp
);
569 if ((flags
& SUPPRESS
) == 0) {
573 res
= (*ccfn
)(buf
, (char **)NULL
, base
);
575 *va_arg(ap
, void **) = (void *)res
;
576 else if (flags
& SHORT
)
577 *va_arg(ap
, short *) = res
;
578 else if (flags
& QUAD
)
579 *va_arg(ap
, quad_t
*) = res
;
580 else if (flags
& LONG
)
581 *va_arg(ap
, long *) = res
;
583 *va_arg(ap
, int *) = res
;
589 #ifdef FLOATING_POINT
591 /* scan a floating point number as if by strtod */
593 if (width
== 0 || width
> sizeof(buf
) - 1)
594 width
= sizeof(buf
) - 1;
596 /* size_t is unsigned, hence this optimisation */
597 if (--width
> sizeof(buf
) - 2)
598 width
= sizeof(buf
) - 2;
601 flags
|= SIGNOK
| NDIGITS
| DPTOK
| EXPOK
;
602 for (p
= buf
; width
; width
--) {
605 * This code mimicks the integer conversion
606 * code, but is much simpler.
610 case '0': case '1': case '2': case '3':
611 case '4': case '5': case '6': case '7':
613 flags
&= ~(SIGNOK
| NDIGITS
);
617 if (flags
& SIGNOK
) {
624 flags
&= ~(SIGNOK
| DPTOK
);
629 /* no exponent without some digits */
630 if ((flags
&(NDIGITS
|EXPOK
)) == EXPOK
) {
632 (flags
& ~(EXPOK
|DPTOK
)) |
643 else if (__srefill(fp
))
647 * If no digits, might be missing exponent digits
648 * (just give back the exponent) or might be missing
649 * regular digits, but had sign and/or decimal point.
651 if (flags
& NDIGITS
) {
653 /* no digits at all */
655 ungetc(*(u_char
*)--p
, fp
);
658 /* just a bad exponent (e and maybe sign) */
660 if (c
!= 'e' && c
!= 'E') {
661 (void) ungetc(c
, fp
);/* sign */
664 (void) ungetc(c
, fp
);
666 if ((flags
& SUPPRESS
) == 0) {
670 res
= strtod(buf
,(char **) NULL
);
672 *va_arg(ap
, long double *) = res
;
673 else if (flags
& LONG
)
674 *va_arg(ap
, double *) = res
;
676 *va_arg(ap
, float *) = res
;
681 #endif /* FLOATING_POINT */
685 return (nassigned
? nassigned
: -1);
691 * Fill in the given table from the scanset at the given format
692 * (just after `['). Return a pointer to the character past the
693 * closing `]'. The table has a 1 wherever characters should be
694 * considered part of the scanset.
699 register u_char
*fmt
;
701 register int c
, n
, v
;
703 /* first `clear' the whole table */
704 c
= *fmt
++; /* first char hat => negated scanset */
706 v
= 1; /* default => accept */
707 c
= *fmt
++; /* get new first char */
709 v
= 0; /* default => reject */
710 /* should probably use memset here */
711 for (n
= 0; n
< 256; n
++)
714 return (fmt
- 1);/* format ended before closing ] */
717 * Now set the entries corresponding to the actual scanset
718 * to the opposite of the above.
720 * The first character may be ']' (or '-') without being special;
721 * the last character may be '-'.
725 tab
[c
] = v
; /* take character c */
727 n
= *fmt
++; /* and examine the next */
730 case 0: /* format ended too soon */
735 * A scanset of the form
737 * is defined as `the digit 0, the digit 1,
738 * the character +, the character -', but
739 * the effect of a scanset such as
741 * is implementation defined. The V7 Unix
742 * scanf treats `a-z' as `the letters a through
743 * z', but treats `a-a' as `the letter a, the
744 * character -, and the letter a'.
746 * For compatibility, the `-' is not considerd
747 * to define a range if the character following
748 * it is either a close bracket (required by ANSI)
749 * or is not numerically greater than the character
750 * we just stored in the table (c).
753 if (n
== ']' || n
< c
) {
755 break; /* resume the for(;;) */
758 do { /* fill in the range */
761 #if 1 /* XXX another disgusting compatibility hack */
763 * Alas, the V7 Unix scanf also treats formats
764 * such as [a-c-e] as `the letters a through e'.
765 * This too is permitted by the standard....
777 case ']': /* end of scanset */
780 default: /* just another character */