path: root/libc/misc/wchar/wchar.c

/*  Copyright (C) 2002, 2003, 2004     Manuel Novoa III
 *
 *  This library is free software; you can redistribute it and/or
 *  modify it under the terms of the GNU Library General Public
 *  License as published by the Free Software Foundation; either
 *  version 2 of the License, or (at your option) any later version.
 *
 *  This library is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 *  Library General Public License for more details.
 *
 *  You should have received a copy of the GNU Library General Public
 *  License along with this library; if not, write to the Free
 *  Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 */

/*  ATTENTION!   ATTENTION!   ATTENTION!   ATTENTION!   ATTENTION!
 *
 *  Besides uClibc, I'm using this code in my libc for elks, which is
 *  a 16-bit environment with a fairly limited compiler.  It would make
 *  things much easier for me if this file isn't modified unnecessarily.
 *  In particular, please put any new or replacement functions somewhere
 *  else, and modify the makefile to use your version instead.
 *  Thanks.  Manuel
 *
 *  ATTENTION!   ATTENTION!   ATTENTION!   ATTENTION!   ATTENTION! */


/* May 23, 2002     Initial Notes:
 *
 * I'm still tweaking this stuff, but it passes the tests I've thrown
 * at it, and Erik needs it for the gcc port.  The glibc extension
 * __wcsnrtombs() hasn't been tested, as I didn't find a test for it
 * in the glibc source.  I also need to fix the behavior of
 * _wchar_utf8sntowcs() if the max number of wchars to convert is 0.
 *
 * UTF-8 -> wchar -> UTF-8 conversion tests on Markus Kuhn's UTF-8-demo.txt
 * file on my platform (x86) show about 5-10% faster conversion speed than
 * glibc with mbsrtowcs()/wcsrtombs() and almost twice as fast as glibc with
 * individual mbrtowc()/wcrtomb() calls.
 *
 * If 'DECODER' is defined, then _wchar_utf8sntowcs() will be compiled
 * as a fail-safe UTF-8 decoder appropriate for a terminal, etc.  which
 * needs to deal gracefully with whatever is sent to it.  In that mode,
 * it passes Markus Kuhn's UTF-8-test.txt stress test.  I plan to add
 * an arg to force that behavior, so the interface will be changing.
 *
 * I need to fix the error checking for 16-bit wide chars.  This isn't
 * an issue for uClibc, but may be for ELKS.  I'm currently not sure
 * if I'll use 16-bit, 32-bit, or configureable wchars in ELKS.
 *
 * July 1, 2002
 *
 * Fixed _wchar_utf8sntowcs() for the max number of wchars == 0 case.
 * Fixed nul-char bug in btowc(), and another in __mbsnrtowcs() for 8-bit
 *    locales.
 * Enabled building of a C/POSIX-locale-only version, so full locale support
 *    no longer needs to be enabled.
 *
 * Nov 4, 2002
 *
 * Fixed a bug in _wchar_wcsntoutf8s().  Don't store wcs position if dst is NULL.
 * Also, introduce an awful hack into _wchar_wcsntoutf8s() and wcsrtombs() in
 *   order to support %ls in printf.  See comments below for details.
 * Change behaviour of wc<->mb functions when in the C locale.  Now they do
 *   a 1-1 map for the range 0x80-UCHAR_MAX.  This is for backwards compatibility
 *   and consistency with the stds requirements that a printf format string by
 *   a valid multibyte string beginning and ending in it's initial shift state.
 *
 * Nov 5, 2002
 *
 * Forgot to change btowc and wctob when I changed the wc<->mb functions yesterday.
 *
 * Nov 7, 2002
 *
 * Add wcwidth and wcswidth, based on Markus Kuhn's wcwidth of 2002-05-08.
 *   Added some size/speed optimizations and integrated it into my locale
 *   framework.  Minimally tested at the moment, but the stub C-locale
 *   version (which most people would probably be using) should be fine.
 *
 * Nov 21, 2002
 *
 * Revert the wc<->mb changes from earlier this month involving the C-locale.
 * Add a couple of ugly hacks to support *wprintf.
 * Add a mini iconv() and iconv implementation (requires locale support).
 *
 * Aug 1, 2003
 * Bug fix for mbrtowc.
 *
 * Aug 18, 2003
 * Bug fix: _wchar_utf8sntowcs and _wchar_wcsntoutf8s now set errno if EILSEQ.
 *
 * Feb 11, 2004
 * Bug fix: Fix size check for remaining output space in iconv().
 *
 * Manuel
 */
#ifdef _LIBC
#include <errno.h>
#include <stddef.h>
#include <limits.h>
#include <stdint.h>
#include <inttypes.h>
#include <stdlib.h>
#include <stdio.h>
#include <assert.h>
#include <locale.h>
#include <wchar.h>
#include <bits/uClibc_uwchar.h>

/**********************************************************************/
#ifdef __UCLIBC_HAS_LOCALE__
#ifdef __UCLIBC_MJN3_ONLY__
#ifdef L_iswspace
/* generates one warning */
#warning TODO: Fix Cc2wc* and Cwc2c* defines!
#endif
#endif /* __UCLIBC_MJN3_ONLY__ */

#define ENCODING		(__UCLIBC_CURLOCALE->encoding)

#define Cc2wc_IDX_SHIFT		__LOCALE_DATA_Cc2wc_IDX_SHIFT
#define Cc2wc_ROW_LEN		__LOCALE_DATA_Cc2wc_ROW_LEN
#define Cwc2c_DOMAIN_MAX	__LOCALE_DATA_Cwc2c_DOMAIN_MAX
#define Cwc2c_TI_SHIFT		__LOCALE_DATA_Cwc2c_TI_SHIFT
#define Cwc2c_TT_SHIFT		__LOCALE_DATA_Cwc2c_TT_SHIFT
#define Cwc2c_TI_LEN		__LOCALE_DATA_Cwc2c_TI_LEN

#ifndef __CTYPE_HAS_UTF_8_LOCALES
#warning __CTYPE_HAS_UTF_8_LOCALES not set!
#endif

#else  /* __UCLIBC_HAS_LOCALE__ */

#ifdef __UCLIBC_MJN3_ONLY__
#ifdef L_btowc
/* emit only once */
#warning fix preprocessor logic testing locale settings
#endif
#endif

#define ENCODING (__ctype_encoding_7_bit)
#ifdef __CTYPE_HAS_8_BIT_LOCALES
#error __CTYPE_HAS_8_BIT_LOCALES is defined!
#endif
#ifdef __CTYPE_HAS_UTF_8_LOCALES
#error __CTYPE_HAS_UTF_8_LOCALES is defined!
#endif
#undef L__wchar_utf8sntowcs
#undef L__wchar_wcsntoutf8s

#endif /* __UCLIBC_HAS_LOCALE__ */
/**********************************************************************/

#if WCHAR_MAX > 0xffffUL
#define UTF_8_MAX_LEN 6
#else
#define UTF_8_MAX_LEN 3
#endif

#define KUHN 1

/* Implementation-specific work functions. */

extern size_t _wchar_utf8sntowcs(wchar_t *__restrict pwc, size_t wn,
					const char **__restrict src, size_t n,
					mbstate_t *ps, int allow_continuation) attribute_hidden;

extern size_t _wchar_wcsntoutf8s(char *__restrict s, size_t n,
					const wchar_t **__restrict src, size_t wn) attribute_hidden;
#endif
/**********************************************************************/
#ifdef L_btowc


wint_t btowc(int c)
{
#ifdef __CTYPE_HAS_8_BIT_LOCALES

	wchar_t wc;
	unsigned char buf[1];
	mbstate_t mbstate;

	if (c != EOF) {
		*buf = (unsigned char) c;
		mbstate.__mask = 0;		/* Initialize the mbstate. */
		if (mbrtowc(&wc, (char*) buf, 1, &mbstate) <= 1) {
			return wc;
		}
	}
	return WEOF;

#else  /* !__CTYPE_HAS_8_BIT_LOCALES */

#ifdef __UCLIBC_HAS_LOCALE__
	assert((ENCODING == __ctype_encoding_7_bit)
		   || (ENCODING == __ctype_encoding_utf8));
#endif

	/* If we don't have 8-bit locale support, then this is trivial since
	 * anything outside of 0-0x7f is illegal in C/POSIX and UTF-8 locales. */
	return (((unsigned int)c) < 0x80) ? c : WEOF;

#endif /* !__CTYPE_HAS_8_BIT_LOCALES */
}
libc_hidden_def(btowc)

#endif
/**********************************************************************/
#ifdef L_wctob

/* Note: We completely ignore ps in all currently supported conversions. */


int wctob(wint_t c)
{
#ifdef __CTYPE_HAS_8_BIT_LOCALES

	unsigned char buf[MB_LEN_MAX];

	return (wcrtomb((char*) buf, c, NULL) == 1) ? *buf : EOF;

#else  /*  __CTYPE