ldap/common/convutf8.c

1N/A/*
1N/A * Copyright (c) 2003, 2011, Oracle and/or its affiliates. All rights reserved.
1N/A */
1N/A
1N/A/*
1N/A * The contents of this file are subject to the Netscape Public
1N/A * License Version 1.1 (the "License"); you may not use this file
1N/A * except in compliance with the License. You may obtain a copy of
1N/A * the License at http://www.mozilla.org/NPL/
1N/A *
1N/A * Software distributed under the License is distributed on an "AS
1N/A * IS" basis, WITHOUT WARRANTY OF ANY KIND, either express or
1N/A * implied. See the License for the specific language governing
1N/A * rights and limitations under the License.
1N/A *
1N/A * The Original Code is Mozilla Communicator client code, released
1N/A * March 31, 1998.
1N/A *
1N/A * The Initial Developer of the Original Code is Netscape
1N/A * Communications Corporation. Portions created by Netscape are
1N/A * Copyright (C) 1998-1999 Netscape Communications Corporation. All
1N/A * Rights Reserved.
1N/A *
1N/A * Contributor(s):
1N/A */
1N/A
1N/A#include <stdio.h>
1N/A#include <stdlib.h>
1N/A#include <string.h>
1N/A#include <locale.h>
1N/A#include <ctype.h>
1N/A
1N/A#ifndef HAVE_LIBICU
1N/A
1N/A#ifdef SOLARIS_LDAP_CMD
1N/A#include <errno.h>
1N/A#include <langinfo.h>
1N/A#include <iconv.h>
1N/A#endif
1N/A
1N/A#ifdef __cplusplus
1N/Aextern "C" {
1N/A#endif
1N/A
1N/Aextern char *ldaptool_charset;
1N/Achar        *ldaptool_convdir = NULL;
1N/Astatic      int charsetset = 0;
1N/Achar        *ldaptool_local2UTF8( const char *src );
1N/A
1N/A#ifdef SOLARIS_LDAP_CMD
1N/Astatic char     *ldaptool_convert( const char *src, const char *fcode,
1N/A                const char *tcode);
1N/Achar        *ldaptool_UTF82local( const char *src );
1N/A#endif  /* SOLARIS_LDAP_CMD */
1N/A
1N/A#ifdef SOLARIS_LDAP_CMD
1N/A/*
1N/A * ICU version always returns string, unless strdup fails.
1N/A * As in ICU version, in case of error strdup(src)
1N/A * Usually strdup(src) will be ASCII and legal anyways.
1N/A */
1N/A
1N/Astatic char *
1N/Aldaptool_convert( const char *src, const char *fcode,
1N/A                 const char *tcode) {
1N/A    char    *dest, *tptr, *tmp;
1N/A    char    *fptr;
1N/A    iconv_t cd;
1N/A    size_t  ileft, oleft, ret, size;
1N/A
1N/A    if (src == NULL)
1N/A    return (NULL);
1N/A
1N/A    if (fcode == NULL || tcode == NULL)
1N/A    return (strdup(src));
1N/A
1N/A    if (strcasecmp(fcode, tcode) == 0)
1N/A    return (strdup(src));
1N/A
1N/A    if ((cd = iconv_open(tcode, fcode)) == (iconv_t)-1) {
1N/A    /* conversion table not available */
1N/A    return (strdup(src));
1N/A    }
1N/A
1N/A    ileft = strlen(src);
1N/A    oleft = 2 * ileft;
1N/A    size = oleft;
1N/A    ret = -1;
1N/A    if ((dest = (char *)malloc(size)) == NULL) {
1N/A    (void) iconv_close(cd);
1N/A    /* maybe sizeof strlen(src) memory still exists */
1N/A    return (strdup(src));
1N/A    }
1N/A    tptr = dest;
1N/A    fptr = (char *)src;
1N/A
1N/A    for (;;) {
1N/A    ret = iconv(cd, &fptr, &ileft, &tptr, &oleft);
1N/A
1N/A    if (ret != (size_t)-1) {
1N/A        /*
1N/A         * Success. Place 'cd' into its initial shift
1N/A         * state before returning.
1N/A         */
1N/A        if (fptr == NULL) /* already in initial state  */
1N/A            break;
1N/A        fptr = NULL;
1N/A        ileft = 0;
1N/A        continue;
1N/A    } if (errno == E2BIG) {
1N/A        /*
1N/A         * Lack of space in output buffer.
1N/A         * Hence double the size and retry.
1N/A         * But before calling  iconv(), oleft
1N/A         * and tptr have to re-adjusted, so that
1N/A         * iconv() doesn't overwrite the data
1N/A         * which has already been converted.
1N/A         */
1N/A        oleft += size;
1N/A        size *= 2;
1N/A        if ((tmp = (char *) realloc(dest, size)) == NULL)
1N/A            break;
1N/A        tptr = tmp + (tptr - dest);
1N/A        dest = tmp;
1N/A        continue;
1N/A    } else {
1N/A        /* Other errors */
1N/A        break;
1N/A    }
1N/A    }
1N/A
1N/A    if (dest != NULL) {
1N/A    if (ret == -1) {
1N/A            /* Free malloc'ed memory on failure */
1N/A        free(dest);
1N/A        dest = NULL;
1N/A    } else if (oleft > 0) {
1N/A        /* NULL terminate the return value */
1N/A        *(dest + (size - oleft)) = '\0';
1N/A    } else {
1N/A        /* realloc one more byte and NULL terminate */
1N/A        if ((tmp = (char *) realloc(dest, size + 1)) == NULL) {
1N/A            free(dest);
1N/A            dest = NULL;
1N/A        } else {
1N/A            *(dest + size) = '\0';
1N/A        }
1N/A    }
1N/A    }
1N/A
1N/A    (void) iconv_close(cd);
1N/A    if (dest == NULL) {
1N/A    /* last chance in case some other failure along the way occurs */
1N/A    return (strdup(src));
1N/A    }
1N/A    return (dest);
1N/A}
1N/A
1N/Achar *
1N/Aldaptool_UTF82local( const char *src )
1N/A{
1N/A    char *to_code;
1N/A    if ((to_code = nl_langinfo(CODESET)) == NULL)
1N/A    return (strdup(src));
1N/A    return (ldaptool_convert(src, "UTF-8", (const char *)to_code));
1N/A}
1N/A#endif  /* SOLARIS_LDAP_CMD */
1N/A
1N/Achar *
1N/Aldaptool_local2UTF8( const char *src )
1N/A{
1N/A#ifdef SOLARIS_LDAP_CMD
1N/A    char *from_code;
1N/A    if ((from_code = nl_langinfo(CODESET)) == NULL)
1N/A    return (strdup(src));
1N/A    return (ldaptool_convert(src, (const char *)from_code, "UTF-8"));
1N/A#else
1N/A    char *utf8;
1N/A    charsetset = 0;
1N/A    if (src == NULL)
1N/A    {
1N/A    return NULL;
1N/A    }
1N/A    utf8 = strdup(src);
1N/A    return ( utf8 );
1N/A#endif  /* SOLARIS_LDAP_CMD */
1N/A}
1N/A
1N/A#else /* HAVE_LIBICU */
1N/A
1N/A#include "unicode/utypes.h"
1N/A#include "unicode/ucnv.h"
1N/A
1N/A#define NSPR20
1N/A
1N/A#ifdef XP_WIN32
1N/A#define  VC_EXTRALEAN
1N/A#include <afxwin.h>
1N/A#include <winnls.h>
1N/A#endif
1N/A
1N/Aextern char *ldaptool_charset;
1N/Astatic int charsetset = 0;
1N/A
1N/Aextern "C" {
1N/Achar *ldaptool_convdir = NULL;
1N/Achar *ldaptool_local2UTF8( const char * );
1N/A}
1N/A
1N/A#ifndef XP_WIN32
1N/Achar * GetNormalizedLocaleName(void);
1N/A
1N/A
1N/Achar *
1N/AGetNormalizedLocaleName(void)
1N/A{
1N/A#ifdef _HPUX_SOURCE
1N/A
1N/A    int    len;
1N/A    char    *locale;
1N/A
1N/A    locale = setlocale(LC_CTYPE, "");
1N/A    if (locale && *locale) {
1N/A        len = strlen(locale);
1N/A    } else {
1N/A        locale = "C";
1N/A        len = 1;
1N/A    }
1N/A
1N/A    if ((!strncmp(locale, "/\x03:", 3)) &&
1N/A        (!strcmp(&locale[len - 2], ";/"))) {
1N/A        locale += 3;
1N/A        len -= 5;
1N/A    }
1N/A
1N/A    locale = strdup(locale);
1N/A    if (locale) {
1N/A        locale[len] = 0;
1N/A    }
1N/A
1N/A    return locale;
1N/A
1N/A#else
1N/A
1N/A    char    *locale;
1N/A
1N/A    locale = setlocale(LC_CTYPE, "");
1N/A    if (locale && *locale) {
1N/A        return strdup(locale);
1N/A    }
1N/A
1N/A    return strdup("C");
1N/A
1N/A#endif
1N/A}
1N/A
1N/A#if defined(IRIX)
1N/Aconst char *CHARCONVTABLE[] =
1N/A{
1N/A"! This table maps the host's locale names to IANA charsets",
1N/A"!",
1N/A"C:             ISO_8859-1:1987",
1N/A"cs:            ISO_8859-2:1987",
1N/A"da:            ISO_8859-1:1987",
1N/A"de:            ISO_8859-1:1987",
1N/A"de_AT:         ISO_8859-1:1987",
1N/A"de_CH:         ISO_8859-1:1987",
1N/A"en:            ISO_8859-1:1987",
1N/A"en_AU:         ISO_8859-1:1987",
1N/A"en_CA:         ISO_8859-1:1987",
1N/A"en_TH:         ISO_8859-1:1987",
1N/A"en_US:         ISO_8859-1:1987",
1N/A"es:            ISO_8859-1:1987",
1N/A"fi:            ISO_8859-1:1987",
1N/A"fr:            ISO_8859-1:1987",
1N/A"fr_BE:         ISO_8859-1:1987",
1N/A"fr_CA:         ISO_8859-1:1987",
1N/A"fr_CH:         ISO_8859-1:1987",
1N/A"is:            ISO_8859-1:1987",
1N/A"it:            ISO_8859-1:1987",
1N/A"it_CH:         ISO_8859-1:1987",
1N/A"ja_JP.EUC:     Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ko_KR.euc:     EUC-KR",
1N/A"nl:            ISO_8859-1:1987",
1N/A"nl_BE:         ISO_8859-1:1987",
1N/A"no:            ISO_8859-1:1987",
1N/A"pl:            ISO_8859-2:1987",
1N/A"pt:            ISO_8859-1:1987",
1N/A"sh:            ISO_8859-2:1987",
1N/A"sk:            ISO_8859-2:1987",
1N/A"sv:            ISO_8859-1:1987",
1N/A"zh_CN.ugb:     GB2312",
1N/A"zh_TW.ucns:    cns11643_1",
1N/ANULL
1N/A};
1N/A#elif defined(SOLARIS)
1N/Aconst char *CHARCONVTABLE[] =
1N/A{
1N/A"! This table maps the host's locale names to IANA charsets",
1N/A"!",
1N/A"C:             ISO_8859-1:1987",
1N/A"ja:            Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ja_JP.EUC:     Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ja_JP.PCK:     Shift_JIS",
1N/A"en:        ISO_8859-1:1987",
1N/A"en_AU:     ISO_8859-1:1987",
1N/A"en_CA:     ISO_8859-1:1987",
1N/A"en_UK:     ISO_8859-1:1987",
1N/A"en_US:     ISO_8859-1:1987",
1N/A"es:        ISO_8859-1:1987",
1N/A"es_AR:     ISO_8859-1:1987",
1N/A"es_BO:     ISO_8859-1:1987",
1N/A"es_CL:     ISO_8859-1:1987",
1N/A"es_CO:     ISO_8859-1:1987",
1N/A"es_CR:     ISO_8859-1:1987",
1N/A"es_EC:     ISO_8859-1:1987",
1N/A"es_GT:     ISO_8859-1:1987",
1N/A"es_MX:     ISO_8859-1:1987",
1N/A"es_NI:     ISO_8859-1:1987",
1N/A"es_PA:     ISO_8859-1:1987",
1N/A"es_PE:     ISO_8859-1:1987",
1N/A"es_PY:     ISO_8859-1:1987",
1N/A"es_SV:     ISO_8859-1:1987",
1N/A"es_UY:     ISO_8859-1:1987",
1N/A"es_VE:     ISO_8859-1:1987",
1N/A"fr:        ISO_8859-1:1987",
1N/A"fr_BE:     ISO_8859-1:1987",
1N/A"fr_CA:     ISO_8859-1:1987",
1N/A"fr_CH:     ISO_8859-1:1987",
1N/A"de:        ISO_8859-1:1987",
1N/A"de_AT:     ISO_8859-1:1987",
1N/A"de_CH:     ISO_8859-1:1987",
1N/A"nl:        ISO_8859-1:1987",
1N/A"nl_BE:     ISO_8859-1:1987",
1N/A"it:        ISO_8859-1:1987",
1N/A"sv:        ISO_8859-1:1987",
1N/A"no:        ISO_8859-1:1987",
1N/A"da:        ISO_8859-1:1987",
1N/A"iso_8859_1:    ISO_8859-1:1987",
1N/A"japanese:      Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ko:            EUC-KR",
1N/A"zh:            GB2312",
1N/A"zh_TW:         cns11643_1",
1N/ANULL
1N/A};
1N/A#elif defined(OSF1)
1N/Aconst char *CHARCONVTABLE[] =
1N/A{
1N/A"! This table maps the host's locale names to IANA charsets",
1N/A"!",
1N/A"C:                     ISO_8859-1:1987",
1N/A"cs_CZ.ISO8859-2:       ISO_8859-2:1987",
1N/A"cs_CZ:                 ISO_8859-2:1987",
1N/A"da_DK.ISO8859-1:       ISO_8859-1:1987",
1N/A"de_CH.ISO8859-1:       ISO_8859-1:1987",
1N/A"de_DE.ISO8859-1:       ISO_8859-1:1987",
1N/A"en_GB.ISO8859-1:       ISO_8859-1:1987",
1N/A"en_US.ISO8859-1:       ISO_8859-1:1987",
1N/A"es_ES.ISO8859-1:       ISO_8859-1:1987",
1N/A"fi_FI.ISO8859-1:       ISO_8859-1:1987",
1N/A"fr_BE.ISO8859-1:       ISO_8859-1:1987",
1N/A"fr_CA.ISO8859-1:       ISO_8859-1:1987",
1N/A"fr_CH.ISO8859-1:       ISO_8859-1:1987",
1N/A"fr_FR.ISO8859-1:       ISO_8859-1:1987",
1N/A"hu_HU.ISO8859-2:       ISO_8859-2:1987",
1N/A"hu_HU:                 ISO_8859-2:1987",
1N/A"is_IS.ISO8859-1:       ISO_8859-1:1987",
1N/A"it_IT.ISO8859-1:       ISO_8859-1:1987",
1N/A"ja_JP.SJIS:            Shift_JIS",
1N/A"ja_JP.eucJP:           Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ja_JP:                 Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ko_KR.eucKR:           EUC-KR",
1N/A"ko_KR:                 EUC-KR",
1N/A"nl_BE.ISO8859-1:       ISO_8859-1:1987",
1N/A"nl_NL.ISO8859-1:       ISO_8859-1:1987",
1N/A"no_NO.ISO8859-1:       ISO_8859-1:1987",
1N/A"pl_PL.ISO8859-2:       ISO_8859-2:1987",
1N/A"pl_PL:                 ISO_8859-2:1987",
1N/A"pt_PT.ISO8859-1:       ISO_8859-1:1987",
1N/A"sk_SK.ISO8859-2:       ISO_8859-2:1987",
1N/A"sk_SK:                 ISO_8859-2:1987",
1N/A"sv_SE.ISO8859-1:       ISO_8859-1:1987",
1N/A"zh_CN:                 GB2312",
1N/A"zh_HK.big5:            Big5",
1N/A"zh_HK.eucTW:           cns11643_1",
1N/A"zh_TW.big5:            Big5",
1N/A"zh_TW.big5@chuyin:     Big5",
1N/A"zh_TW.big5@radical:    Big5",
1N/A"zh_TW.big5@stroke:     Big5",
1N/A"zh_TW.eucTW:           cns11643_1",
1N/A"zh_TW.eucTW@chuyin:    cns11643_1",
1N/A"zh_TW.eucTW@radical:   cns11643_1",
1N/A"zh_TW.eucTW@stroke:    cns11643_1",
1N/A"zh_TW:                 cns11643_1",
1N/ANULL
1N/A};
1N/A#elif defined(HPUX)
1N/Aconst char *CHARCONVTABLE[] =
1N/A{
1N/A"! This table maps the host's locale names to IANA charsets",
1N/A"!",
1N/A"C:         ISO_8859-1:1987",
1N/A"ja_JP:         Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ja_JP.SJIS:        Shift_JIS",
1N/A"ja_JP.eucJP:       Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"es_ES:         ISO_8859-1:1987",
1N/A"es_ES.iso88591:    ISO_8859-1:1987",
1N/A"sv_SE:         ISO_8859-1:1987",
1N/A"sv_SE.iso88591:    ISO_8859-1:1987",
1N/A"da_DK:         ISO_8859-1:1987",
1N/A"da_DK.iso88591:    ISO_8859-1:1987",
1N/A"nl_NL:         ISO_8859-1:1987",
1N/A"nl_NL.iso88591:    ISO_8859-1:1987",
1N/A"en:            ISO_8859-1:1987",
1N/A"en_GB:         ISO_8859-1:1987",
1N/A"en_GB.iso88591:    ISO_8859-1:1987",
1N/A"en_US:         ISO_8859-1:1987",
1N/A"en_US.iso88591:    ISO_8859-1:1987",
1N/A"fi_FI:         ISO_8859-1:1987",
1N/A"fi_FI.iso88591:    ISO_8859-1:1987",
1N/A"fr_CA:         ISO_8859-1:1987",
1N/A"fr_CA.iso88591:    ISO_8859-1:1987",
1N/A"fr_FR:         ISO_8859-1:1987",
1N/A"fr_FR.iso88591:    ISO_8859-1:1987",
1N/A"de_DE:         ISO_8859-1:1987",
1N/A"de_DE.iso88591:    ISO_8859-1:1987",
1N/A"is_IS:         ISO_8859-1:1987",
1N/A"is_IS.iso88591:    ISO_8859-1:1987",
1N/A"it_IT:         ISO_8859-1:1987",
1N/A"it_IT.iso88591:    ISO_8859-1:1987",
1N/A"no_NO:         ISO_8859-1:1987",
1N/A"no_NO.iso88591:    ISO_8859-1:1987",
1N/A"pt_PT:         ISO_8859-1:1987",
1N/A"pt_PT.iso88591:    ISO_8859-1:1987",
1N/A"hu_HU:         ISO_8859-2:1987",
1N/A"hu_HU.iso88592:    ISO_8859-2:1987",
1N/A"cs_CZ:         ISO_8859-2:1987",
1N/A"cs_CZ.iso88592:    ISO_8859-2:1987",
1N/A"pl_PL:         ISO_8859-2:1987",
1N/A"pl_PL.iso88592:    ISO_8859-2:1987",
1N/A"ro_RO:         ISO_8859-2:1987",
1N/A"ro_RO.iso88592:    ISO_8859-2:1987",
1N/A"hr_HR:         ISO_8859-2:1987",
1N/A"hr_HR.iso88592:    ISO_8859-2:1987",
1N/A"sk_SK:         ISO_8859-2:1987",
1N/A"sk_SK.iso88592:    ISO_8859-2:1987",
1N/A"sl_SI:         ISO_8859-2:1987",
1N/A"sl_SI.iso88592:    ISO_8859-2:1987",
1N/A"american.iso88591:     ISO_8859-1:1987",
1N/A"bulgarian:             ISO_8859-2:1987",
1N/A"c-french.iso88591:     ISO_8859-1:1987",
1N/A"chinese-s:             GB2312",
1N/A"chinese-t.big5:                Big5",
1N/A"czech:                 ISO_8859-2:1987",
1N/A"danish.iso88591:       ISO_8859-1:1987",
1N/A"dutch.iso88591:                ISO_8859-1:1987",
1N/A"english.iso88591:      ISO_8859-1:1987",
1N/A"finnish.iso88591:      ISO_8859-1:1987",
1N/A"french.iso88591:       ISO_8859-1:1987",
1N/A"german.iso88591:       ISO_8859-1:1987",
1N/A"hungarian:             ISO_8859-2:1987",
1N/A"icelandic.iso88591:    ISO_8859-1:1987",
1N/A"italian.iso88591:      ISO_8859-1:1987",
1N/A"japanese.euc:          Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"japanese:              Shift_JIS",
1N/A"katakana:              Shift_JIS",
1N/A"korean:                        EUC-KR",
1N/A"norwegian.iso88591:    ISO_8859-1:1987",
1N/A"polish:                        ISO_8859-2:1987",
1N/A"portuguese.iso88591:   ISO_8859-1:1987",
1N/A"rumanian:              ISO_8859-2:1987",
1N/A"serbocroatian:         ISO_8859-2:1987",
1N/A"slovene:               ISO_8859-2:1987",
1N/A"spanish.iso88591:      ISO_8859-1:1987",
1N/A"swedish.iso88591:      ISO_8859-1:1987",
1N/ANULL
1N/A};
1N/A#elif defined(AIX)
1N/Aconst char *CHARCONVTABLE[] =
1N/A{
1N/A"! This table maps the host's locale names to IANA charsets",
1N/A"!",
1N/A"C:                     ISO_8859-1:1987",
1N/A"En_JP.IBM-932:         Shift_JIS",
1N/A"En_JP:                 Shift_JIS",
1N/A"Ja_JP.IBM-932:         Shift_JIS",
1N/A"Ja_JP:                 Shift_JIS",
1N/A"da_DK.ISO8859-1:       ISO_8859-1:1987",
1N/A"da_DK:                 ISO_8859-1:1987",
1N/A"de_CH.ISO8859-1:       ISO_8859-1:1987",
1N/A"de_CH:                 ISO_8859-1:1987",
1N/A"de_DE.ISO8859-1:       ISO_8859-1:1987",
1N/A"de_DE:                 ISO_8859-1:1987",
1N/A"en_GB.ISO8859-1:       ISO_8859-1:1987",
1N/A"en_GB:                 ISO_8859-1:1987",
1N/A"en_JP.IBM-eucJP:       Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"en_JP:                 Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"en_KR.IBM-eucKR:       EUC-KR",
1N/A"en_KR:                 EUC-KR",
1N/A"en_TW.IBM-eucTW:       cns11643_1",
1N/A"en_TW:                 cns11643_1",
1N/A"en_US.ISO8859-1:       ISO_8859-1:1987",
1N/A"en_US:                 ISO_8859-1:1987",
1N/A"es_ES.ISO8859-1:       ISO_8859-1:1987",
1N/A"es_ES:                 ISO_8859-1:1987",
1N/A"fi_FI.ISO8859-1:       ISO_8859-1:1987",
1N/A"fi_FI:                 ISO_8859-1:1987",
1N/A"fr_BE.ISO8859-1:       ISO_8859-1:1987",
1N/A"fr_BE:                 ISO_8859-1:1987",
1N/A"fr_CA.ISO8859-1:       ISO_8859-1:1987",
1N/A"fr_CA:                 ISO_8859-1:1987",
1N/A"fr_CH.ISO8859-1:       ISO_8859-1:1987",
1N/A"fr_CH:                 ISO_8859-1:1987",
1N/A"fr_FR.ISO8859-1:       ISO_8859-1:1987",
1N/A"fr_FR:                 ISO_8859-1:1987",
1N/A"is_IS.ISO8859-1:       ISO_8859-1:1987",
1N/A"is_IS:                 ISO_8859-1:1987",
1N/A"it_IT.ISO8859-1:       ISO_8859-1:1987",
1N/A"it_IT:                 ISO_8859-1:1987",
1N/A"ja_JP.IBM-eucJP:       Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ja_JP:                 Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ko_KR.IBM-eucKR:       EUC-KR",
1N/A"ko_KR:                 EUC-KR",
1N/A"nl_BE.ISO8859-1:       ISO_8859-1:1987",
1N/A"nl_BE:                 ISO_8859-1:1987",
1N/A"nl_NL.ISO8859-1:       ISO_8859-1:1987",
1N/A"nl_NL:                 ISO_8859-1:1987",
1N/A"no_NO.ISO8859-1:       ISO_8859-1:1987",
1N/A"no_NO:                 ISO_8859-1:1987",
1N/A"pt_PT.ISO8859-1:       ISO_8859-1:1987",
1N/A"pt_PT:                 ISO_8859-1:1987",
1N/A"sv_SE.ISO8859-1:       ISO_8859-1:1987",
1N/A"sv_SE:                 ISO_8859-1:1987",
1N/A"zh_TW.IBM-eucTW:       cns11643_1",
1N/A"zh_TW:                 cns11643_1",
1N/ANULL
1N/A};
1N/A#else   // sunos by default
1N/Aconst char *CHARCONVTABLE[] =
1N/A{
1N/A"! This table maps the host's locale names to IANA charsets",
1N/A"!",
1N/A"C:             ISO_8859-1:1987",
1N/A"de:            ISO_8859-1:1987",
1N/A"en_US:         ISO_8859-1:1987",
1N/A"es:            ISO_8859-1:1987",
1N/A"fr:            ISO_8859-1:1987",
1N/A"iso_8859_1:    ISO_8859-1:1987",
1N/A"it:            ISO_8859-1:1987",
1N/A"ja:            Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ja_JP.EUC:     Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"japanese:      Extended_UNIX_Code_Packed_Format_for_Japanese",
1N/A"ko:            EUC-KR",
1N/A"sv:            ISO_8859-1:1987",
1N/A"zh:            GB2312",
1N/A"zh_TW:         cns11643_1",
1N/ANULL
1N/A};
1N/A#endif
1N/A
1N/A#define BSZ     256
1N/A
1N/Achar *
1N/AGetCharsetFromLocale(char *locale)
1N/A{
1N/A    char *tmpcharset = NULL;
1N/A    char buf[BSZ];
1N/A    char *p;
1N/A    const char *line;
1N/A    int i=0;
1N/A
1N/A    line = CHARCONVTABLE[i];
1N/A    while (line != NULL)
1N/A    {
1N/A       if (*line == 0)
1N/A       {
1N/A          break;
1N/A       }
1N/A
1N/A       strcpy(buf, line);
1N/A       line = CHARCONVTABLE[++i];
1N/A
1N/A       if (strlen(buf) == 0 || buf[0] == '!')
1N/A       {
1N/A          continue;
1N/A       }
1N/A       p = strchr(buf, ':');
1N/A       if (p == NULL)
1N/A       {
1N/A          tmpcharset = NULL;
1N/A          break;
1N/A       }
1N/A       *p = 0;
1N/A       if (strcmp(buf, locale) == 0) {
1N/A          while (*++p == ' ' || *p == '\t')
1N/A             ;
1N/A          if (isalpha(*p)) {
1N/A             tmpcharset = strdup(p);
1N/A          } else
1N/A             tmpcharset = NULL;
1N/A
1N/A          break;
1N/A       }
1N/A    }
1N/A    return tmpcharset;
1N/A}
1N/A
1N/A#endif /* Not defined XP_WIN32 */
1N/A
1N/A#ifdef XP_WIN32
1N/Achar *_convertor(const char *instr, int bFromUTF8)
1N/A{
1N/A    char  *outstr = NULL;
1N/A    int    inlen, wclen, outlen;
1N/A    LPWSTR wcstr;
1N/A
1N/A    if (instr == NULL)
1N/A            return NULL;
1N/A
1N/A    if ((inlen = strlen(instr)) <= 0)
1N/A            return NULL;
1N/A
1N/A    /* output never becomes longer than input,
1N/A     * thus we don't have to ask for the length
1N/A     */
1N/A    wcstr = (LPWSTR) malloc( sizeof( WCHAR ) * (inlen+1) );
1N/A    if (!wcstr)
1N/A        return NULL;
1N/A
1N/A    wclen = MultiByteToWideChar(bFromUTF8 ? CP_UTF8 : CP_ACP, 0, instr,
1N/A                                 inlen, wcstr, inlen);
1N/A    outlen = WideCharToMultiByte(bFromUTF8 ? CP_ACP : CP_UTF8, 0, wcstr,
1N/A                                  wclen, NULL, 0, NULL, NULL);
1N/A
1N/A    if (outlen > 0) {
1N/A        outstr = (char *) malloc(outlen + 2);
1N/A        outlen = WideCharToMultiByte(bFromUTF8 ? CP_ACP : CP_UTF8, 0, wcstr,
1N/A                                      wclen, outstr, outlen, NULL, NULL);
1N/A        if (outlen > 0)
1N/A            *(outstr+outlen) = _T('\0');
1N/A        else
1N/A            return NULL;
1N/A    }
1N/A    free( wcstr );
1N/A    return outstr;
1N/A}
1N/A#endif
1N/A
1N/Achar *
1N/Aldaptool_local2UTF8( const char *src )
1N/A{
1N/A    char *utf8;
1N/A#ifndef XP_WIN32
1N/A    char *locale, *newcharset;
1N/A    size_t outLen, resultLen;
1N/A    UErrorCode err = U_ZERO_ERROR;
1N/A    UConverter *cnv;
1N/A
1N/A    if (src == NULL)
1N/A    {
1N/A      return NULL;
1N/A    }
1N/A    else if (*src == 0 || (ldaptool_charset == NULL)
1N/A         || (!strcmp( ldaptool_charset, "" )))
1N/A    {
1N/A    /* no option specified, so assume it's already in utf-8 */
1N/A        utf8 = strdup(src);
1N/A        return utf8;
1N/A    }
1N/A
1N/A    if( !strcmp( ldaptool_charset, "0" )
1N/A        && (!charsetset) )
1N/A    {
1N/A    /* zero option specified, so try to get default codepage
1N/A       this sucker is strdup'd immediately so it's OK to cast */
1N/A    newcharset = (char *)ucnv_getDefaultName();
1N/A    if (newcharset != NULL) {
1N/A        free( ldaptool_charset );
1N/A        /* the default codepage lives in ICU */
1N/A        ldaptool_charset = strdup(newcharset);
1N/A        if (ldaptool_charset == NULL) {
1N/A        return strdup(src);
1N/A        }
1N/A    }
1N/A    charsetset = 1;
1N/A    }
1N/A    else
1N/A    if( strcmp( ldaptool_charset, "" ) && (!charsetset) )
1N/A    {
1N/A    /* -i option specified with charset name */
1N/A        charsetset = 1;
1N/A    }
1N/A
1N/A    /* do the preflight - get the size needed for the target buffer */
1N/A    outLen = (size_t) ucnv_convert( "utf-8", ldaptool_charset, NULL, 0, src,
1N/A                                      strlen( src ) * sizeof(char), &err);
1N/A
1N/A    if ((err != U_BUFFER_OVERFLOW_ERROR) || (outLen == 0)) {
1N/A      /* default to just a copy of the string - this covers
1N/A         the case of an illegal charset also */
1N/A      return strdup(src);
1N/A    }
1N/A
1N/A    utf8 =  (char *) malloc( outLen + 1);
1N/A    if( utf8 == NULL ) {
1N/A      /* if we're already out of memory, does strdup just return NULL? */
1N/A       return strdup(src);
1N/A    }
1N/A
1N/A    /* do the actual conversion this time */
1N/A    err = U_ZERO_ERROR;
1N/A    resultLen = ucnv_convert( "utf-8", ldaptool_charset, utf8, (outLen + 1), src,
1N/A               strlen(src) * sizeof(char), &err );
1N/A
1N/A    if (!U_SUCCESS(err)) {
1N/A      free(utf8);
1N/A      return strdup(src);
1N/A    }
1N/A
1N/A#else
1N/A    utf8 = _convertor(src, FALSE);
1N/A    if( utf8 == NULL )
1N/A        utf8 = strdup(src);
1N/A#endif
1N/A
1N/A    return utf8;
1N/A}
1N/A#endif /* HAVE_LIBICU */
1N/A
1N/A#ifndef HAVE_LIBICU
1N/A#ifdef __cplusplus
1N/A}
1N/A#endif
1N/A#endif