pa-fazil
/
android_bionic
mirror of https://github.com/fazilsheik96/android_bionic.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge "Add %b and %B support to the scanf/wscanf and strto*/wcsto* families."

This commit is contained in:
Elliott Hughes 2022-08-11 21:22:23 +00:00 committed by Gerrit Code Review
parent ea7f84435d 1f462dec34
commit d1c3d4a454
17 changed files with 272 additions and 530 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
docs/status.md
View File

@ -55,7 +55,9 @@ New libc functions in U (API level 34):
* `close_range` and `copy_file_range` (Linux-specific GNU extensions).
New libc behavior in U (API level 34):
* Support for `%b` and `%B` in the printf and wprintf family.
* Support for `%b` and `%B` in the printf/wprintf family, `%b` in the
scanf/wscanf family, and `0b` prefixes with base 0 in the strtol/wcstol
family.
New libc functions in T (API level 33):
* `backtrace`, `backtrace_symbols`, `backtrace_symbols_fd` (`<execinfo.h>`).

8
libc/Android.bp
View File

@ -35,7 +35,7 @@ libc_common_src_files = [
"stdio/stdio.cpp",
"stdio/stdio_ext.cpp",
"stdio/vfscanf.cpp",
"stdio/vfwscanf.c",
"stdio/vfwscanf.cpp",
]
// off64_t/time64_t support on LP32.
@ -500,13 +500,7 @@ cc_library_static {
"upstream-openbsd/lib/libc/locale/mbstowcs.c",
"upstream-openbsd/lib/libc/locale/mbtowc.c",
"upstream-openbsd/lib/libc/locale/wcscoll.c",
"upstream-openbsd/lib/libc/locale/wcstoimax.c",
"upstream-openbsd/lib/libc/locale/wcstol.c",
"upstream-openbsd/lib/libc/locale/wcstoll.c",
"upstream-openbsd/lib/libc/locale/wcstombs.c",
"upstream-openbsd/lib/libc/locale/wcstoul.c",
"upstream-openbsd/lib/libc/locale/wcstoull.c",
"upstream-openbsd/lib/libc/locale/wcstoumax.c",
"upstream-openbsd/lib/libc/locale/wcsxfrm.c",
"upstream-openbsd/lib/libc/locale/wctob.c",
"upstream-openbsd/lib/libc/locale/wctomb.c",

65
libc/bionic/strtol.cpp
View File

@ -32,11 +32,13 @@
#include <inttypes.h>
#include <limits.h>
#include <stdlib.h>
#include <wchar.h>
template <typename T, T Min, T Max> T StrToI(const char* nptr, char** endptr, int base) {
template <typename T, T Min, T Max, typename CharT>
T StrToI(const CharT* nptr, CharT** endptr, int base) {
// Ensure that base is between 2 and 36 inclusive, or the special value of 0.
if (base < 0 || base == 1 || base > 36) {
if (endptr != nullptr) *endptr = const_cast<char*>(nptr);
if (endptr != nullptr) *endptr = const_cast<CharT*>(nptr);
errno = EINVAL;
return 0;
}
@ -44,7 +46,7 @@ template <typename T, T Min, T Max> T StrToI(const char* nptr, char** endptr, in
// Skip white space and pick up leading +/- sign if any.
// If base is 0, allow 0x for hex and 0 for octal, else
// assume decimal; if base is already 16, allow 0x.
const char* s = nptr;
const CharT* s = nptr;
int c;
do {
c = *s++;
@ -62,6 +64,11 @@ template <typename T, T Min, T Max> T StrToI(const char* nptr, char** endptr, in
s += 2;
base = 16;
}
if ((base == 0 || base == 2) && c == '0' && (*s == 'b' || *s == 'B') && isdigit(s[1])) {
c = s[1];
s += 2;
base = 2;
}
if (base == 0) base = (c == '0') ? 8 : 10;
// We always work in the negative space because the most negative value has a
@ -91,7 +98,7 @@ template <typename T, T Min, T Max> T StrToI(const char* nptr, char** endptr, in
acc -= c;
}
}
if (endptr != nullptr) *endptr = const_cast<char*>(any ? s - 1 : nptr);
if (endptr != nullptr) *endptr = const_cast<CharT*>(any ? s - 1 : nptr);
if (!neg) {
if (acc == Min) {
errno = ERANGE;
@ -103,14 +110,15 @@ template <typename T, T Min, T Max> T StrToI(const char* nptr, char** endptr, in
return acc;
}
template <typename T, T Max> T StrToU(const char* nptr, char** endptr, int base) {
template <typename T, T Max, typename CharT>
T StrToU(const CharT* nptr, CharT** endptr, int base) {
if (base < 0 || base == 1 || base > 36) {
if (endptr != nullptr) *endptr = const_cast<char*>(nptr);
if (endptr != nullptr) *endptr = const_cast<CharT*>(nptr);
errno = EINVAL;
return 0;
}
const char* s = nptr;
const CharT* s = nptr;
int c;
do {
c = *s++;
@ -128,6 +136,11 @@ template <typename T, T Max> T StrToU(const char* nptr, char** endptr, int base)
s += 2;
base = 16;
}
if ((base == 0 || base == 2) && c == '0' && (*s == 'b' || *s == 'B') && isdigit(s[1])) {
c = s[1];
s += 2;
base = 2;
}
if (base == 0) base = (c == '0') ? 8 : 10;
T cutoff = Max / static_cast<T>(base);
@ -155,7 +168,7 @@ template <typename T, T Max> T StrToU(const char* nptr, char** endptr, int base)
}
}
if (neg && any > 0) acc = -acc;
if (endptr != nullptr) *endptr = const_cast<char*>(any ? s - 1 : nptr);
if (endptr != nullptr) *endptr = const_cast<CharT*>(any ? s - 1 : nptr);
return acc;
}
@ -172,30 +185,54 @@ long long atoll(const char* s) {
}
intmax_t strtoimax(const char* s, char** end, int base) {
return StrToI<intmax_t, INTMAX_MIN, INTMAX_MAX>(s, end, base);
return StrToI<intmax_t, INTMAX_MIN, INTMAX_MAX, char>(s, end, base);
}
intmax_t wcstoimax(const wchar_t* s, wchar_t** end, int base) {
return StrToI<intmax_t, INTMAX_MIN, INTMAX_MAX, wchar_t>(s, end, base);
}
long strtol(const char* s, char** end, int base) {
return StrToI<long, LONG_MIN, LONG_MAX>(s, end, base);
return StrToI<long, LONG_MIN, LONG_MAX, char>(s, end, base);
}
long wcstol(const wchar_t* s, wchar_t** end, int base) {
return StrToI<long, LONG_MIN, LONG_MAX, wchar_t>(s, end, base);
}
long long strtoll(const char* s, char** end, int base) {
return StrToI<long long, LLONG_MIN, LLONG_MAX>(s, end, base);
return StrToI<long long, LLONG_MIN, LLONG_MAX, char>(s, end, base);
}
long long wcstoll(const wchar_t* s, wchar_t** end, int base) {
return StrToI<long long, LLONG_MIN, LLONG_MAX, wchar_t>(s, end, base);
}
// Public API since L, but not in any header.
__strong_alias(strtoq, strtoll);
unsigned long strtoul(const char* s, char** end, int base) {
return StrToU<unsigned long, ULONG_MAX>(s, end, base);
return StrToU<unsigned long, ULONG_MAX, char>(s, end, base);
}
unsigned long wcstoul(const wchar_t* s, wchar_t** end, int base) {
return StrToU<unsigned long, ULONG_MAX, wchar_t>(s, end, base);
}
unsigned long long strtoull(const char* s, char** end, int base) {
return StrToU<unsigned long long, ULLONG_MAX>(s, end, base);
return StrToU<unsigned long long, ULLONG_MAX, char>(s, end, base);
}
unsigned long long wcstoull(const wchar_t* s, wchar_t** end, int base) {
return StrToU<unsigned long long, ULLONG_MAX, wchar_t>(s, end, base);
}
uintmax_t strtoumax(const char* s, char** end, int base) {
return StrToU<uintmax_t, UINTMAX_MAX>(s, end, base);
return StrToU<uintmax_t, UINTMAX_MAX, char>(s, end, base);
}
uintmax_t wcstoumax(const wchar_t* s, wchar_t** end, int base) {
return StrToU<uintmax_t, UINTMAX_MAX, wchar_t>(s, end, base);
}
// Public API since L, but not in any header.

63
libc/stdio/vfscanf.cpp
View File

@ -69,7 +69,8 @@
#define HAVESIGN 0x04000 // Sign detected
#define NDIGITS 0x08000 // No digits detected
#define PFXOK 0x10000 // "0x" prefix is (still) legal
#define NZDIGITS 0x20000 // No zero digits detected
#define PFBOK 0x20000 // "0b" prefix is (still) legal
#define NZDIGITS 0x40000 // No zero digits detected
// Conversion types.
#define CT_CHAR 0 // %c conversion
@ -101,9 +102,6 @@ int __svfscanf(FILE* fp, const char* fmt0, va_list ap) {
void* allocation = nullptr; // Allocated but unassigned result for %mc/%ms/%m[.
size_t capacity = 0; // Number of char/wchar_t units allocated in `allocation`.
/* `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 };
_SET_ORIENTATION(fp, -1);
nassigned = 0;
@ -188,6 +186,12 @@ literal:
* Conversions.
* Those marked `compat' are for 4.[123]BSD compatibility.
*/
case 'b':
c = CT_INT;
base = 2;
flags |= PFBOK; /* enable 0b prefixing */
break;
case 'D': /* compat */
flags |= LONG;
__BIONIC_FALLTHROUGH;
@ -558,7 +562,7 @@ literal:
* 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.
* it to 8 and enable 0b/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
@ -567,15 +571,24 @@ literal:
case '0':
if (base == 0) {
base = 8;
flags |= PFXOK;
flags |= PFBOK | PFXOK;
}
if (flags & NZDIGITS)
if (flags & NZDIGITS) {
flags &= ~(SIGNOK | NZDIGITS | NDIGITS);
else
flags &= ~(SIGNOK | PFXOK | NDIGITS);
} else {
flags &= ~(SIGNOK | PFBOK | PFXOK | NDIGITS);
}
goto ok;
/* 1 through 7 always legal */
case 'B':
case 'b':
// Is this 'b' or 'B' potentially part of an "0b" prefix?
if ((flags & PFBOK) && p == buf + 1 + !!(flags & HAVESIGN)) {
base = 2;
flags &= ~PFBOK;
goto ok;
}
// No? Fall through and see if it's a hex digit instead then...
__BIONIC_FALLTHROUGH;
case '1':
case '2':
case '3':
@ -583,34 +596,21 @@ literal:
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);
if (base == 0) base = 10;
if (base != 16 && (c - '0') >= base) break; /* not legal here */
flags &= ~(SIGNOK | PFBOK | PFXOK | NDIGITS);
goto ok;
/* sign ok only as first character */
@ -653,17 +653,16 @@ literal:
break; /* EOF */
}
/*
* 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 we had only a sign, it is no good; push back the sign.
* If the number was `[-+]0[BbXx]`, push back and treat it
* as `[-+]0`.
*/
if (flags & NDIGITS) {
if (p > buf) (void)ungetc(*(u_char*)--p, fp);
goto match_failure;
}
c = ((u_char*)p)[-1];
if (c == 'x' || c == 'X') {
if ((base == 2 && (c == 'b' || c == 'B')) || c == 'x' || c == 'X') {
--p;
(void)ungetc(c, fp);
}

77
libc/stdio/vfwscanf.c → libc/stdio/vfwscanf.cpp
View File

@ -42,6 +42,8 @@
#include <wctype.h>
#include "local.h"
#include <platform/bionic/macros.h>
#define BUF 513 /* Maximum length of numeric string. */
/*
@ -65,15 +67,16 @@
* SIGNOK, HAVESIGN, NDIGITS, DPTOK, and EXPOK are for floating point;
* SIGNOK, HAVESIGN, NDIGITS, PFXOK, and NZDIGITS are for integral.
*/
#define SIGNOK 0x01000 /* +/- is (still) legal */
#define SIGNOK 0x01000 /* +/- is (still) legal */
#define HAVESIGN 0x02000 /* sign detected */
#define NDIGITS 0x04000 /* no digits detected */
#define NDIGITS 0x04000 /* no digits detected */
#define DPTOK 0x08000 /* (float) decimal point is still legal */
#define EXPOK 0x10000 /* (float) exponent (e+3, etc) still legal */
#define DPTOK 0x08000 /* (float) decimal point is still legal */
#define EXPOK 0x10000 /* (float) exponent (e+3, etc) still legal */
#define PFXOK 0x08000 /* 0x prefix is (still) legal */
#define NZDIGITS 0x10000 /* no zero digits detected */
#define PFBOK 0x20000 /* 0x prefix is (still) legal */
#define PFXOK 0x40000 /* 0x prefix is (still) legal */
#define NZDIGITS 0x80000 /* no zero digits detected */
/*
* Conversion types.
@ -147,9 +150,6 @@ int __vfwscanf(FILE* __restrict fp, const wchar_t* __restrict fmt, __va_list ap)
char mbbuf[MB_LEN_MAX]; /* temporary mb. character buffer */
mbstate_t mbs;
/* `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 };
_SET_ORIENTATION(fp, 1);
nassigned = 0;
@ -239,9 +239,15 @@ int __vfwscanf(FILE* __restrict fp, const wchar_t* __restrict fmt, __va_list ap)
* Conversions.
* Those marked `compat' are for 4.[123]BSD compatibility.
*/
case 'b':
c = CT_INT;
base = 2;
flags |= PFBOK; /* enable 0b prefixing */
break;
case 'D': /* compat */
flags |= LONG;
/* FALLTHROUGH */
__BIONIC_FALLTHROUGH;
case 'd':
c = CT_INT;
base = 10;
@ -254,7 +260,7 @@ int __vfwscanf(FILE* __restrict fp, const wchar_t* __restrict fmt, __va_list ap)
case 'O': /* compat */
flags |= LONG;
/* FALLTHROUGH */
__BIONIC_FALLTHROUGH;
case 'o':
c = CT_INT;
flags |= UNSIGNED;
@ -468,7 +474,7 @@ int __vfwscanf(FILE* __restrict fp, const wchar_t* __restrict fmt, __va_list ap)
* 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.
* it to 8 and enable 0b/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
@ -477,15 +483,26 @@ int __vfwscanf(FILE* __restrict fp, const wchar_t* __restrict fmt, __va_list ap)
case '0':
if (base == 0) {
base = 8;
flags |= PFXOK;
flags |= PFBOK | PFXOK;
}
if (flags & NZDIGITS)
if (flags & NZDIGITS) {
flags &= ~(SIGNOK | NZDIGITS | NDIGITS);
else
flags &= ~(SIGNOK | PFXOK | NDIGITS);
} else {
flags &= ~(SIGNOK | PFBOK | PFXOK | NDIGITS);
}
goto ok;
/* 1 through 7 always legal */
case 'B':
case 'b':
// Is this 'b' potentially part of an "0b" prefix?
if ((flags & PFBOK) && p == buf + 1 + !!(flags & HAVESIGN)) {
base = 2;
flags &= ~PFBOK;
goto ok;
}
// No? Fall through and see if it's a hex digit instead then...
__BIONIC_FALLTHROUGH;
case '1':
case '2':
case '3':
@ -493,34 +510,21 @@ int __vfwscanf(FILE* __restrict fp, const wchar_t* __restrict fmt, __va_list ap)
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);
if (base == 0) base = 10;
if (base != 16 && (int)(c - '0') >= base) break; /* not legal here */
flags &= ~(SIGNOK | PFBOK | PFXOK | NDIGITS);
goto ok;
/* sign ok only as first character */
@ -560,17 +564,16 @@ int __vfwscanf(FILE* __restrict fp, const wchar_t* __restrict fmt, __va_list ap)
*p++ = (wchar_t)c;
}
/*
* 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 we had only a sign, it is no good; push back the sign.
* If the number was `[-+]0[BbXx]`, push back and treat it
* as `[-+]0`.
*/
if (flags & NDIGITS) {
if (p > buf) __ungetwc(*--p, fp);
goto match_failure;
}
c = p[-1];
if (c == 'x' || c == 'X') {
if ((base == 2 && (c == 'b' || c == 'B')) || c == 'x' || c == 'X') {
--p;
__ungetwc(c, fp);
}

136
libc/upstream-openbsd/lib/libc/locale/_wcstol.h
View File

@ -1,136 +0,0 @@
/* $OpenBSD: _wcstol.h,v 1.3 2015/10/01 02:32:07 guenther Exp $ */
/* $NetBSD: _wcstol.h,v 1.2 2003/08/07 16:43:03 agc Exp $ */
/*-
* Copyright (c) 1990, 1993
* The Regents of the University of California. All rights reserved.
*
* 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.
*
* Original version ID:
* @(#)strtol.c 8.1 (Berkeley) 6/4/93
* NetBSD: wcstol.c,v 1.1 2001/09/27 16:30:36 yamt Exp
* Citrus: xpg4dl/FreeBSD/lib/libc/locale/wcstol.c,v 1.2 2001/09/21 16:11:41 yamt Exp
*/
/*
* function template for wcstol, wcstoll and wcstoimax.
*
* parameters:
* FUNCNAME : function name
* int_type : return type
* MIN_VALUE : lower limit of the return type
* MAX_VALUE : upper limit of the return type
*/
int_type
FUNCNAME(const wchar_t *nptr, wchar_t **endptr, int base)
{
const wchar_t *s;
int_type acc, cutoff;
wint_t wc;
int i;
int neg, any, cutlim;
/* check base value */
if (base && (base < 2 || base > 36)) {
errno = EINVAL;
return 0;
}
/*
* Skip white space and pick up leading +/- sign if any.
* If base is 0, allow 0x for hex and 0 for octal, else
* assume decimal; if base is already 16, allow 0x.
*/
s = nptr;
do {
wc = (wchar_t) *s++;
} while (iswspace(wc));
if (wc == L'-') {
neg = 1;
wc = *s++;
} else {
neg = 0;
if (wc == L'+')
wc = *s++;
}
if ((base == 0 || base == 16) &&
wc == L'0' && (*s == L'x' || *s == L'X')) {
wc = s[1];
s += 2;
base = 16;
}
if (base == 0)
base = wc == L'0' ? 8 : 10;
/*
* See strtol for comments as to the logic used.
*/
cutoff = neg ? MIN_VALUE : MAX_VALUE;
cutlim = (int)(cutoff % base);
cutoff /= base;
if (neg) {
if (cutlim > 0) {
cutlim -= base;
cutoff += 1;
}
cutlim = -cutlim;
}
for (acc = 0, any = 0;; wc = (wchar_t) *s++) {
i = wctoint(wc);
if (i == -1)
break;
if (i >= base)
break;
if (any < 0)
continue;
if (neg) {
if (acc < cutoff || (acc == cutoff && i > cutlim)) {
any = -1;
acc = MIN_VALUE;
errno = ERANGE;
} else {
any = 1;
acc *= base;
acc -= i;
}
} else {
if (acc > cutoff || (acc == cutoff && i > cutlim)) {
any = -1;
acc = MAX_VALUE;
errno = ERANGE;
} else {
any = 1;
acc *= base;
acc += i;
}
}
}
if (endptr != 0)
*endptr = (wchar_t *)(any ? s - 1 : nptr);
return (acc);
}
DEF_STRONG(FUNCNAME);

116
libc/upstream-openbsd/lib/libc/locale/_wcstoul.h
View File

@ -1,116 +0,0 @@
/* $OpenBSD: _wcstoul.h,v 1.3 2015/10/01 02:32:07 guenther Exp $ */
/* $NetBSD: _wcstoul.h,v 1.2 2003/08/07 16:43:03 agc Exp $ */
/*
* Copyright (c) 1990, 1993
* The Regents of the University of California. All rights reserved.
*
* 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.
*
* Original version ID:
* @(#)strtoul.c 8.1 (Berkeley) 6/4/93
* Citrus: xpg4dl/FreeBSD/lib/libc/locale/wcstoul.c,v 1.2 2001/09/21 16:11:41 yamt Exp
* NetBSD: wcstoul.c,v 1.1 2001/09/27 16:30:37 yamt Exp
*/
/*
* function template for wcstoul, wcstoull and wcstoumax.
*
* parameters:
* FUNCNAME : function name
* uint_type : return type
* MAX_VALUE : upper limit of the return type
*/
uint_type
FUNCNAME(const wchar_t *nptr, wchar_t **endptr, int base)
{
const wchar_t *s;
uint_type acc, cutoff;
wint_t wc;
int i;
int neg, any, cutlim;
if (base && (base < 2 || base > 36)) {
errno = EINVAL;
return 0;
}
/*
* Skip white space and pick up leading +/- sign if any.
* If base is 0, allow 0x for hex and 0 for octal, else
* assume decimal; if base is already 16, allow 0x.
*/
s = nptr;
do {
wc = (wchar_t) *s++;
} while (iswspace(wc));
if (wc == L'-') {
neg = 1;
wc = *s++;
} else {
neg = 0;
if (wc == L'+')
wc = *s++;
}
if ((base == 0 || base == 16) &&
wc == L'0' && (*s == L'x' || *s == L'X')) {
wc = s[1];
s += 2;
base = 16;
}
if (base == 0)
base = wc == L'0' ? 8 : 10;
/*
* See strtoul for comments as to the logic used.
*/
cutoff = MAX_VALUE / (uint_type)base;
cutlim = (int)(MAX_VALUE % (uint_type)base);
for (acc = 0, any = 0;; wc = (wchar_t) *s++) {
i = wctoint(wc);
if (i == (wint_t)-1)
break;
if (i >= base)
break;
if (any < 0)
continue;
if (acc > cutoff || (acc == cutoff && i > cutlim)) {
any = -1;
acc = MAX_VALUE;
errno = ERANGE;
} else {
any = 1;
acc *= (uint_type)base;
acc += i;
}
}
if (neg && any > 0)
acc = -acc;
if (endptr != 0)
*endptr = (wchar_t *)(any ? s - 1 : nptr);
return (acc);
}
DEF_STRONG(FUNCNAME);

19
libc/upstream-openbsd/lib/libc/locale/wcstoimax.c
View File

@ -1,19 +0,0 @@
/* $OpenBSD: wcstoimax.c,v 1.1 2009/01/13 18:13:51 kettenis Exp $ */
/* $NetBSD: wcstol.c,v 1.2 2003/03/11 09:21:23 tshiozak Exp $ */
#include <ctype.h>
#include <errno.h>
#include <inttypes.h>
#include <stdint.h>
#include <stdlib.h>
#include <wchar.h>
#include <wctype.h>
#include "wctoint.h"
#define FUNCNAME wcstoimax
typedef intmax_t int_type;
#define MIN_VALUE INTMAX_MIN
#define MAX_VALUE INTMAX_MAX
#include "_wcstol.h"

18
libc/upstream-openbsd/lib/libc/locale/wcstol.c
View File

@ -1,18 +0,0 @@
/* $OpenBSD: wcstol.c,v 1.2 2005/08/08 08:05:35 espie Exp $ */
/* $NetBSD: wcstol.c,v 1.2 2003/03/11 09:21:23 tshiozak Exp $ */
#include <ctype.h>
#include <errno.h>
#include <limits.h>
#include <stdlib.h>
#include <wchar.h>
#include <wctype.h>
#include "wctoint.h"
#define FUNCNAME wcstol
typedef long int_type;
#define MIN_VALUE LONG_MIN
#define MAX_VALUE LONG_MAX
#include "_wcstol.h"

18
libc/upstream-openbsd/lib/libc/locale/wcstoll.c
View File

@ -1,18 +0,0 @@
/* $OpenBSD: wcstoll.c,v 1.2 2005/08/08 08:05:35 espie Exp $ */
/* $NetBSD: wcstoll.c,v 1.1 2003/03/11 09:21:23 tshiozak Exp $ */
#include <ctype.h>
#include <errno.h>
#include <limits.h>
#include <stdlib.h>
#include <wchar.h>
#include <wctype.h>
#include "wctoint.h"
#define FUNCNAME wcstoll
typedef long long int int_type;
#define MIN_VALUE LLONG_MIN
#define MAX_VALUE LLONG_MAX
#include "_wcstol.h"

17
libc/upstream-openbsd/lib/libc/locale/wcstoul.c
View File

@ -1,17 +0,0 @@
/* $OpenBSD: wcstoul.c,v 1.2 2005/08/08 08:05:35 espie Exp $ */
/* $NetBSD: wcstoul.c,v 1.2 2003/03/11 09:21:24 tshiozak Exp $ */
#include <ctype.h>
#include <errno.h>
#include <limits.h>
#include <stdlib.h>
#include <wchar.h>
#include <wctype.h>
#include "wctoint.h"
#define FUNCNAME wcstoul
typedef unsigned long uint_type;
#define MAX_VALUE ULONG_MAX
#include "_wcstoul.h"

17
libc/upstream-openbsd/lib/libc/locale/wcstoull.c
View File

@ -1,17 +0,0 @@
/* $OpenBSD: wcstoull.c,v 1.2 2005/08/08 08:05:35 espie Exp $ */
/* $NetBSD: wcstoull.c,v 1.1 2003/03/11 09:21:24 tshiozak Exp $ */
#include <ctype.h>
#include <errno.h>
#include <limits.h>
#include <stdlib.h>
#include <wchar.h>
#include <wctype.h>
#include "wctoint.h"
#define FUNCNAME wcstoull
typedef unsigned long long int uint_type;
#define MAX_VALUE ULLONG_MAX
#include "_wcstoul.h"

18
libc/upstream-openbsd/lib/libc/locale/wcstoumax.c
View File

@ -1,18 +0,0 @@
/* $OpenBSD: wcstoumax.c,v 1.1 2009/01/13 18:13:51 kettenis Exp $ */
/* $NetBSD: wcstoul.c,v 1.2 2003/03/11 09:21:24 tshiozak Exp $ */
#include <ctype.h>
#include <errno.h>
#include <inttypes.h>
#include <stdint.h>
#include <stdlib.h>
#include <wchar.h>
#include <wctype.h>
#include "wctoint.h"
#define FUNCNAME wcstoumax
typedef uintmax_t uint_type;
#define MAX_VALUE UINTMAX_MAX
#include "_wcstoul.h"

79
libc/upstream-openbsd/lib/libc/locale/wctoint.h
View File

@ -1,79 +0,0 @@
/* $OpenBSD: wctoint.h,v 1.2 2015/09/13 11:38:08 guenther Exp $ */
/* $NetBSD: __wctoint.h,v 1.1 2001/09/28 11:25:37 yamt Exp $ */
/*-
* Copyright (c)2001 Citrus Project,
* All rights reserved.
*
* 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.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
*
* $Citrus: xpg4dl/FreeBSD/lib/libc/locale/__wctoint.h,v 1.1 2001/09/21 13:52:32 yamt Exp $
*/
inline static int
wctoint(wchar_t wc)
{
int n;
switch (wc) {
case L'0': n = 0; break;
case L'1': n = 1; break;
case L'2': n = 2; break;
case L'3': n = 3; break;
case L'4': n = 4; break;
case L'5': n = 5; break;
case L'6': n = 6; break;
case L'7': n = 7; break;
case L'8': n = 8; break;
case L'9': n = 9; break;
case L'A': case L'a': n = 10; break;
case L'B': case L'b': n = 11; break;
case L'C': case L'c': n = 12; break;
case L'D': case L'd': n = 13; break;
case L'E': case L'e': n = 14; break;
case L'F': case L'f': n = 15; break;
case L'G': case L'g': n = 16; break;
case L'H': case L'h': n = 17; break;
case L'I': case L'i': n = 18; break;
case L'J': case L'j': n = 19; break;
case L'K': case L'k': n = 20; break;
case L'L': case L'l': n = 21; break;
case L'M': case L'm': n = 22; break;
case L'N': case L'n': n = 23; break;
case L'O': case L'o': n = 24; break;
case L'P': case L'p': n = 25; break;
case L'Q': case L'q': n = 26; break;
case L'R': case L'r': n = 27; break;
case L'S': case L's': n = 28; break;
case L'T': case L't': n = 29; break;
case L'U': case L'u': n = 30; break;
case L'V': case L'v': n = 31; break;
case L'W': case L'w': n = 32; break;
case L'X': case L'x': n = 33; break;
case L'Y': case L'y': n = 34; break;
case L'Z': case L'z': n = 35; break;
default: n = -1; break; /* error */
}
return n;
}

136
tests/stdio_test.cpp
View File

@ -3045,3 +3045,139 @@ TEST(STDIO_TEST, swprintf_B) {
EXPECT_EQ(std::wstring(L"<0>"), buf);
#pragma clang diagnostic pop
}
TEST(STDIO_TEST, scanf_i_decimal) {
int i;
EXPECT_EQ(1, sscanf("<123789>", "<%i>", &i));
EXPECT_EQ(123789, i);
long long int lli;
char ch;
EXPECT_EQ(2, sscanf("1234567890abcdefg", "%lli%c", &lli, &ch));
EXPECT_EQ(1234567890, lli);
EXPECT_EQ('a', ch);
}
TEST(STDIO_TEST, scanf_i_hex) {
int i;
EXPECT_EQ(1, sscanf("<0x123abf>", "<%i>", &i));
EXPECT_EQ(0x123abf, i);
long long int lli;
char ch;
EXPECT_EQ(2, sscanf("0x1234567890abcdefg", "%lli%c", &lli, &ch));
EXPECT_EQ(0x1234567890abcdefLL, lli);
EXPECT_EQ('g', ch);
}
TEST(STDIO_TEST, scanf_i_octal) {
int i;
EXPECT_EQ(1, sscanf("<01234567>", "<%i>", &i));
EXPECT_EQ(01234567, i);
long long int lli;
char ch;
EXPECT_EQ(2, sscanf("010234567890abcdefg", "%lli%c", &lli, &ch));
EXPECT_EQ(010234567, lli);
EXPECT_EQ('8', ch);
}
TEST(STDIO_TEST, scanf_i_binary) {
int i;
EXPECT_EQ(1, sscanf("<0b101>", "<%i>", &i));
EXPECT_EQ(0b101, i);
long long int lli;
char ch;
EXPECT_EQ(2, sscanf("0b10234567890abcdefg", "%lli%c", &lli, &ch));
EXPECT_EQ(0b10, lli);
EXPECT_EQ('2', ch);
}
TEST(STDIO_TEST, wscanf_i_decimal) {
int i;
EXPECT_EQ(1, swscanf(L"<123789>", L"<%i>", &i));
EXPECT_EQ(123789, i);
long long int lli;
char ch;
EXPECT_EQ(2, swscanf(L"1234567890abcdefg", L"%lli%c", &lli, &ch));
EXPECT_EQ(1234567890, lli);
EXPECT_EQ('a', ch);
}
TEST(STDIO_TEST, wscanf_i_hex) {
int i;
EXPECT_EQ(1, swscanf(L"<0x123abf>", L"<%i>", &i));
EXPECT_EQ(0x123abf, i);
long long int lli;
char ch;
EXPECT_EQ(2, swscanf(L"0x1234567890abcdefg", L"%lli%c", &lli, &ch));
EXPECT_EQ(0x1234567890abcdefLL, lli);
EXPECT_EQ('g', ch);
}
TEST(STDIO_TEST, wscanf_i_octal) {
int i;
EXPECT_EQ(1, swscanf(L"<01234567>", L"<%i>", &i));
EXPECT_EQ(01234567, i);
long long int lli;
char ch;
EXPECT_EQ(2, swscanf(L"010234567890abcdefg", L"%lli%c", &lli, &ch));
EXPECT_EQ(010234567, lli);
EXPECT_EQ('8', ch);
}
TEST(STDIO_TEST, wscanf_i_binary) {
int i;
EXPECT_EQ(1, swscanf(L"<0b101>", L"<%i>", &i));
EXPECT_EQ(0b101, i);
long long int lli;
char ch;
EXPECT_EQ(2, swscanf(L"0b10234567890abcdefg", L"%lli%c", &lli, &ch));
EXPECT_EQ(0b10, lli);
EXPECT_EQ('2', ch);
}
TEST(STDIO_TEST, scanf_b) {
// Our clang doesn't know about %b yet.
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wformat"
#pragma clang diagnostic ignored "-Wformat-invalid-specifier"
int i;
char ch;
EXPECT_EQ(2, sscanf("<1012>", "<%b%c>", &i, &ch));
EXPECT_EQ(0b101, i);
EXPECT_EQ('2', ch);
EXPECT_EQ(1, sscanf("<00000101>", "<%08b>", &i));
EXPECT_EQ(0b00000101, i);
EXPECT_EQ(1, sscanf("<0b1010>", "<%b>", &i));
EXPECT_EQ(0b1010, i);
EXPECT_EQ(2, sscanf("-0b", "%i%c", &i, &ch));
EXPECT_EQ(0, i);
EXPECT_EQ('b', ch);
#pragma clang diagnostic pop
}
TEST(STDIO_TEST, swscanf_b) {
// Our clang doesn't know about %b yet.
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wformat"
#pragma clang diagnostic ignored "-Wformat-invalid-specifier"
int i;
char ch;
EXPECT_EQ(2, swscanf(L"<1012>", L"<%b%c>", &i, &ch));
EXPECT_EQ(0b101, i);
EXPECT_EQ('2', ch);
EXPECT_EQ(1, swscanf(L"<00000101>", L"<%08b>", &i));
EXPECT_EQ(0b00000101, i);
EXPECT_EQ(1, swscanf(L"<0b1010>", L"<%b>", &i));
EXPECT_EQ(0b1010, i);
EXPECT_EQ(2, swscanf(L"-0b", L"%i%c", &i, &ch));
EXPECT_EQ(0, i);
EXPECT_EQ('b', ch);
#pragma clang diagnostic pop
}

10
tests/stdlib_test.cpp
View File

@ -843,11 +843,19 @@ static void CheckStrToInt(T fn(const char* s, char** end, int base)) {
ASSERT_EQ(T(-123), fn("-123", &end_p, 10));
ASSERT_EQ(T(123), fn("+123", &end_p, 10));
// If we see "0b" *not* followed by a binary digit, we shouldn't swallow the 'b'.
ASSERT_EQ(T(0), fn("0b", &end_p, 2));
ASSERT_EQ('b', *end_p);
// Binary (the "0b" prefix) is case-insensitive.
ASSERT_EQ(T(0b101), fn("0b101", &end_p, 0));
ASSERT_EQ(T(0b101), fn("0B101", &end_p, 0));
// If we see "0x" *not* followed by a hex digit, we shouldn't swallow the 'x'.
ASSERT_EQ(T(0), fn("0xy", &end_p, 16));
ASSERT_EQ('x', *end_p);
// Hexadecimal (both the 0x and the digits) is case-insensitive.
// Hexadecimal (both the "0x" prefix and the digits) is case-insensitive.
ASSERT_EQ(T(0xab), fn("0xab", &end_p, 0));
ASSERT_EQ(T(0xab), fn("0Xab", &end_p, 0));
ASSERT_EQ(T(0xab), fn("0xAB", &end_p, 0));

1
tests/wchar_test.cpp
View File

@ -460,6 +460,7 @@ void TestWcsToInt(WcsToIntFn<T> fn) {
TestSingleWcsToInt(fn, L" 123 45", 0, static_cast<T>(123), 6);
TestSingleWcsToInt(fn, L" -123", 0, static_cast<T>(-123), 6);
TestSingleWcsToInt(fn, L"0x10000", 0, static_cast<T>(65536), 7);
TestSingleWcsToInt(fn, L"0b1011", 0, static_cast<T>(0b1011), 6);
}
template <typename T>