source: branches/samba-3.0/source/lib/charcnv.c@ 120

/*
   Unix SMB/CIFS implementation.
   Character set conversion Extensions
   Copyright (C) Igor Vergeichik <iverg@mail.ru> 2001
   Copyright (C) Andrew Tridgell 2001
   Copyright (C) Simo Sorce 2001
   Copyright (C) Martin Pool 2003

   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 2 of the License, or
   (at your option) any later version.

   This program 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 General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this program; if not, write to the Free Software
   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.

*/
#include "includes.h"

/* We can parameterize this if someone complains.... JRA. */

char lp_failed_convert_char(void)
{
    return '_';
}

/**
 * @file
 *
 * @brief Character-set conversion routines built on our iconv.
 *
 * @note Samba's internal character set (at least in the 3.0 series)
 * is always the same as the one for the Unix filesystem.  It is
 * <b>not</b> necessarily UTF-8 and may be different on machines that
 * need i18n filenames to be compatible with Unix software.  It does
 * have to be a superset of ASCII.  All multibyte sequences must start
 * with a byte with the high bit set.
 *
 * @sa lib/iconv.c
 */


static smb_iconv_t conv_handles[NUM_CHARSETS][NUM_CHARSETS];
static BOOL conv_silent; /* Should we do a debug if the conversion fails ? */

/**
 * Return the name of a charset to give to iconv().
 **/
static const char *charset_name(charset_t ch)
{
    const char *ret = NULL;

    if (ch == CH_UTF16LE) ret = "UTF-16LE";
    else if (ch == CH_UTF16BE) ret = "UTF-16BE";
    else if (ch == CH_UNIX) ret = lp_unix_charset();
    else if (ch == CH_DOS) ret = lp_dos_charset();
    else if (ch == CH_DISPLAY) ret = lp_display_charset();
    else if (ch == CH_UTF8) ret = "UTF8";

#if defined(HAVE_NL_LANGINFO) && defined(CODESET)
    if (ret && !strcmp(ret, "LOCALE")) {
        const char *ln = NULL;

#ifdef HAVE_SETLOCALE
        setlocale(LC_ALL, "");
#endif
        ln = nl_langinfo(CODESET);
        if (ln) {
            /* Check whether the charset name is supported
               by iconv */
            smb_iconv_t handle = smb_iconv_open(ln,"UCS-2LE");
            if (handle == (smb_iconv_t) -1) {
                DEBUG(5,("Locale charset '%s' unsupported, using ASCII instead\n", ln));
                ln = NULL;
            } else {
                DEBUG(5,("Substituting charset '%s' for LOCALE\n", ln));
                smb_iconv_close(handle);
            }
        }
        ret = ln;
    }
#endif

    if (!ret || !*ret) ret = "ASCII";
    return ret;
}

void lazy_initialize_conv(void)
{
    static int initialized = False;

    if (!initialized) {
        initialized = True;
        load_case_tables();
        init_iconv();
    }
}

/**
 * Destroy global objects allocated by init_iconv()
 **/
void gfree_charcnv(void)
{
    int c1, c2;

    for (c1=0;c1<NUM_CHARSETS;c1++) {
        for (c2=0;c2<NUM_CHARSETS;c2++) {
            if ( conv_handles[c1][c2] ) {
                smb_iconv_close( conv_handles[c1][c2] );
                conv_handles[c1][c2] = 0;
            }
        }
    }
}

/**
 * Initialize iconv conversion descriptors.
 *
 * This is called the first time it is needed, and also called again
 * every time the configuration is reloaded, because the charset or
 * codepage might have changed.
 **/
void init_iconv(void)
{
    int c1, c2;
    BOOL did_reload = False;

    /* so that charset_name() works we need to get the UNIX<->UCS2 going
       first */
    if (!conv_handles[CH_UNIX][CH_UTF16LE])
        conv_handles[CH_UNIX][CH_UTF16LE] = smb_iconv_open(charset_name(CH_UTF16LE), "ASCII");

    if (!conv_handles[CH_UTF16LE][CH_UNIX])
        conv_handles[CH_UTF16LE][CH_UNIX] = smb_iconv_open("ASCII", charset_name(CH_UTF16LE));

    for (c1=0;c1<NUM_CHARSETS;c1++) {
        for (c2=0;c2<NUM_CHARSETS;c2++) {
            const char *n1 = charset_name((charset_t)c1);
            const char *n2 = charset_name((charset_t)c2);
            if (conv_handles[c1][c2] &&
                strcmp(n1, conv_handles[c1][c2]->from_name) == 0 &&
                strcmp(n2, conv_handles[c1][c2]->to_name) == 0)
                continue;

            did_reload = True;

            if (conv_handles[c1][c2])
                smb_iconv_close(conv_handles[c1][c2]);

            conv_handles[c1][c2] = smb_iconv_open(n2,n1);
            if (conv_handles[c1][c2] == (smb_iconv_t)-1) {
                DEBUG(0,("init_iconv: Conversion from %s to %s not supported\n",
                     charset_name((charset_t)c1), charset_name((charset_t)c2)));
                if (c1 != CH_UTF16LE && c1 != CH_UTF16BE) {
                    n1 = "ASCII";
                }
                if (c2 != CH_UTF16LE && c2 != CH_UTF16BE) {
                    n2 = "ASCII";
                }
                DEBUG(0,("init_iconv: Attempting to replace with conversion from %s to %s\n",
                    n1, n2 ));
                conv_handles[c1][c2] = smb_iconv_open(n2,n1);
                if (!conv_handles[c1][c2]) {
                    DEBUG(0,("init_iconv: Conversion from %s to %s failed", n1, n2));
                    smb_panic("init_iconv: conv_handle initialization failed.");
                }
            }
        }
    }

    if (did_reload) {
        /* XXX: Does this really get called every time the dos
         * codepage changes? */
        /* XXX: Is the did_reload test too strict? */
        conv_silent = True;
        init_doschar_table();
        init_valid_table();
        conv_silent = False;
    }
}

/**
 * Convert string from one encoding to another, making error checking etc
 * Slow path version - uses (slow) iconv.
 *
 * @param src pointer to source string (multibyte or singlebyte)
 * @param srclen length of the source string in bytes
 * @param dest pointer to destination string (multibyte or singlebyte)
 * @param destlen maximal length allowed for string
 * @param allow_bad_conv determines if a "best effort" conversion is acceptable (never returns errors)
 * @returns the number of bytes occupied in the destination
 *
 * Ensure the srclen contains the terminating zero.
 *
 **/

static size_t convert_string_internal(charset_t from, charset_t to,
              void const *src, size_t srclen,
              void *dest, size_t destlen, BOOL allow_bad_conv)
{
    size_t i_len, o_len;
    size_t retval;
    const char* inbuf = (const char*)src;
    char* outbuf = (char*)dest;
    smb_iconv_t descriptor;

    lazy_initialize_conv();

    descriptor = conv_handles[from][to];

    if (srclen == (size_t)-1) {
        if (from == CH_UTF16LE || from == CH_UTF16BE) {
            srclen = (strlen_w((const smb_ucs2_t *)src)+1) * 2;
        } else {
            srclen = strlen((const char *)src)+1;
        }
    }


    if (descriptor == (smb_iconv_t)-1 || descriptor == (smb_iconv_t)0) {
        if (!conv_silent)
            DEBUG(0,("convert_string_internal: Conversion not supported.\n"));
        return (size_t)-1;
    }

    i_len=srclen;
    o_len=destlen;

 again:

    retval = smb_iconv(descriptor, &inbuf, &i_len, &outbuf, &o_len);
    if(retval==(size_t)-1) {
            const char *reason="unknown error";
        switch(errno) {
            case EINVAL:
                reason="Incomplete multibyte sequence";
                if (!conv_silent)
                    DEBUG(3,("convert_string_internal: Conversion error: %s(%s)\n",reason,inbuf));
                if (allow_bad_conv)
                    goto use_as_is;
                break;
            case E2BIG:
                reason="No more room";
                if (!conv_silent) {
                    if (from == CH_UNIX) {
                        DEBUG(3,("E2BIG: convert_string(%s,%s): srclen=%u destlen=%u - '%s'\n",
                            charset_name(from), charset_name(to),
                            (unsigned int)srclen, (unsigned int)destlen, (const char *)src));
                    } else {
                        DEBUG(3,("E2BIG: convert_string(%s,%s): srclen=%u destlen=%u\n",
                            charset_name(from), charset_name(to),
                            (unsigned int)srclen, (unsigned int)destlen));
                    }
                }
                break;
            case EILSEQ:
                reason="Illegal multibyte sequence";
                if (!conv_silent)
                    DEBUG(3,("convert_string_internal: Conversion error: %s(%s)\n",reason,inbuf));
                if (allow_bad_conv)
                    goto use_as_is;
                break;
            default:
                if (!conv_silent)
                    DEBUG(0,("convert_string_internal: Conversion error: %s(%s)\n",reason,inbuf));
                break;
        }
        /* smb_panic(reason); */
    }
    return destlen-o_len;

 use_as_is:

    /*
     * Conversion not supported. This is actually an error, but there are so
     * many misconfigured iconv systems and smb.conf's out there we can't just
     * fail. Do a very bad conversion instead.... JRA.
     */

    {
        if (o_len == 0 || i_len == 0)
            return destlen - o_len;

        if (((from == CH_UTF16LE)||(from == CH_UTF16BE)) &&
                ((to != CH_UTF16LE)||(to != CH_UTF16BE))) {
            /* Can't convert from utf16 any endian to multibyte.
               Replace with the default fail char.
            */
            if (i_len < 2)
                return destlen - o_len;
            if (i_len >= 2) {
                *outbuf = lp_failed_convert_char();

                outbuf++;
                o_len--;

                inbuf += 2;
                i_len -= 2;
            }

            if (o_len == 0 || i_len == 0)
                return destlen - o_len;

            /* Keep trying with the next char... */
            goto again;

        } else if (from != CH_UTF16LE && from != CH_UTF16BE && to == CH_UTF16LE) {
            /* Can't convert to UTF16LE - just widen by adding the
               default fail char then zero.
            */
            if (o_len < 2)
                return destlen - o_len;

            outbuf[0] = lp_failed_convert_char();
            outbuf[1] = '\0';

            inbuf++;
            i_len--;

            outbuf += 2;
            o_len -= 2;

            if (o_len == 0 || i_len == 0)
                return destlen - o_len;

            /* Keep trying with the next char... */
            goto again;

        } else if (from != CH_UTF16LE && from != CH_UTF16BE &&
                to != CH_UTF16LE && to != CH_UTF16BE) {
            /* Failed multibyte to multibyte. Just copy the default fail char and
                try again. */
            outbuf[0] = lp_failed_convert_char();

            inbuf++;
            i_len--;

            outbuf++;
            o_len--;

            if (o_len == 0 || i_len == 0)
                return destlen - o_len;

            /* Keep trying with the next char... */
            goto again;

        } else {
            /* Keep compiler happy.... */
            return destlen - o_len;
        }
    }
}

/**
 * Convert string from one encoding to another, making error checking etc
 * Fast path version - handles ASCII first.
 *
 * @param src pointer to source string (multibyte or singlebyte)
 * @param srclen length of the source string in bytes, or -1 for nul terminated.
 * @param dest pointer to destination string (multibyte or singlebyte)
 * @param destlen maximal length allowed for string - *NEVER* -1.
 * @param allow_bad_conv determines if a "best effort" conversion is acceptable (never returns errors)
 * @returns the number of bytes occupied in the destination
 *
 * Ensure the srclen contains the terminating zero.
 *
 * This function has been hand-tuned to provide a fast path.
 * Don't change unless you really know what you are doing. JRA.
 **/

size_t convert_string(charset_t from, charset_t to,
              void const *src, size_t srclen,
              void *dest, size_t destlen, BOOL allow_bad_conv)
{
    /*
     * NB. We deliberately don't do a strlen here if srclen == -1.
     * This is very expensive over millions of calls and is taken
     * care of in the slow path in convert_string_internal. JRA.
     */

#ifdef DEVELOPER
    SMB_ASSERT(destlen != (size_t)-1);
#endif

    if (srclen == 0)
        return 0;

    if (from != CH_UTF16LE && from != CH_UTF16BE && to != CH_UTF16LE && to != CH_UTF16BE) {
        const unsigned char *p = (const unsigned char *)src;
        unsigned char *q = (unsigned char *)dest;
        size_t slen = srclen;
        size_t dlen = destlen;
        unsigned char lastp = '\0';
        size_t retval = 0;

        /* If all characters are ascii, fast path here. */
        while (slen && dlen) {
            if ((lastp = *p) <= 0x7f) {
                *q++ = *p++;
                if (slen != (size_t)-1) {
                    slen--;
                }
                dlen--;
                retval++;
                if (!lastp)
                    break;
            } else {
#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
                goto general_case;
#else
                return retval + convert_string_internal(from, to, p, slen, q, dlen, allow_bad_conv);
#endif
            }
        }
        if (!dlen) {
            /* Even if we fast path we should note if we ran out of room. */
            if (((slen != (size_t)-1) && slen) ||
                    ((slen == (size_t)-1) && lastp)) {
                errno = E2BIG;
            }
        }
        return retval;
    } else if (from == CH_UTF16LE && to != CH_UTF16LE) {
        const unsigned char *p = (const unsigned char *)src;
        unsigned char *q = (unsigned char *)dest;
        size_t retval = 0;
        size_t slen = srclen;
        size_t dlen = destlen;
        unsigned char lastp = '\0';

        /* If all characters are ascii, fast path here. */
        while (((slen == (size_t)-1) || (slen >= 2)) && dlen) {
            if (((lastp = *p) <= 0x7f) && (p[1] == 0)) {
                *q++ = *p;
                if (slen != (size_t)-1) {
                    slen -= 2;
                }
                p += 2;
                dlen--;
                retval++;
                if (!lastp)
                    break;
            } else {
#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
                goto general_case;
#else
                return retval + convert_string_internal(from, to, p, slen, q, dlen, allow_bad_conv);
#endif
            }
        }
        if (!dlen) {
            /* Even if we fast path we should note if we ran out of room. */
            if (((slen != (size_t)-1) && slen) ||
                    ((slen == (size_t)-1) && lastp)) {
                errno = E2BIG;
            }
        }
        return retval;
    } else if (from != CH_UTF16LE && from != CH_UTF16BE && to == CH_UTF16LE) {
        const unsigned char *p = (const unsigned char *)src;
        unsigned char *q = (unsigned char *)dest;
        size_t retval = 0;
        size_t slen = srclen;
        size_t dlen = destlen;
        unsigned char lastp = '\0';

        /* If all characters are ascii, fast path here. */
        while (slen && (dlen >= 2)) {
            if ((lastp = *p) <= 0x7F) {
                *q++ = *p++;
                *q++ = '\0';
                if (slen != (size_t)-1) {
                    slen--;
                }
                dlen -= 2;
                retval += 2;
                if (!lastp)
                    break;
            } else {
#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
                goto general_case;
#else
                return retval + convert_string_internal(from, to, p, slen, q, dlen, allow_bad_conv);
#endif
            }
        }
        if (!dlen) {
            /* Even if we fast path we should note if we ran out of room. */
            if (((slen != (size_t)-1) && slen) ||
                    ((slen == (size_t)-1) && lastp)) {
                errno = E2BIG;
            }
        }
        return retval;
    }

#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
  general_case:
#endif
    return convert_string_internal(from, to, src, srclen, dest, destlen, allow_bad_conv);
}

/**
 * Convert between character sets, allocating a new buffer for the result.
 *
 * @param ctx TALLOC_CTX to use to allocate with. If NULL use malloc.
 * @param srclen length of source buffer.
 * @param dest always set at least to NULL
 * @note -1 is not accepted for srclen.
 *
 * @returns Size in bytes of the converted string; or -1 in case of error.
 *
 * Ensure the srclen contains the terminating zero.
 *
 * I hate the goto's in this function. It's embarressing.....
 * There has to be a cleaner way to do this. JRA.
 **/

size_t convert_string_allocate(TALLOC_CTX *ctx, charset_t from, charset_t to,
                   void const *src, size_t srclen, void *dst, BOOL allow_bad_conv)
{
    size_t i_len, o_len, destlen = MAX(srclen, 512);
    size_t retval;
    const char *inbuf = (const char *)src;
    char *outbuf = NULL, *ob = NULL;
    smb_iconv_t descriptor;
    void **dest = (void **)dst;

    *dest = NULL;

    if (src == NULL || srclen == (size_t)-1)
        return (size_t)-1;
    if (srclen == 0)
        return 0;

    lazy_initialize_conv();

    descriptor = conv_handles[from][to];

    if (descriptor == (smb_iconv_t)-1 || descriptor == (smb_iconv_t)0) {
        if (!conv_silent)
            DEBUG(0,("convert_string_allocate: Conversion not supported.\n"));
        return (size_t)-1;
    }

  convert:

    if ((destlen*2) < destlen) {
        /* wrapped ! abort. */
        if (!conv_silent)
            DEBUG(0, ("convert_string_allocate: destlen wrapped !\n"));
        if (!ctx)
            SAFE_FREE(outbuf);
        return (size_t)-1;
    } else {
        destlen = destlen * 2;
    }

    if (ctx) {
        ob = (char *)TALLOC_REALLOC(ctx, ob, destlen);
    } else {
        ob = (char *)SMB_REALLOC(ob, destlen);
    }

    if (!ob) {
        DEBUG(0, ("convert_string_allocate: realloc failed!\n"));
        return (size_t)-1;
    }
    outbuf = ob;
    i_len = srclen;
    o_len = destlen;

 again:

    retval = smb_iconv(descriptor,
               &inbuf, &i_len,
               &outbuf, &o_len);
    if(retval == (size_t)-1)        {
            const char *reason="unknown error";
        switch(errno) {
            case EINVAL:
                reason="Incomplete multibyte sequence";
                if (!conv_silent)
                    DEBUG(3,("convert_string_allocate: Conversion error: %s(%s)\n",reason,inbuf));
                if (allow_bad_conv)
                    goto use_as_is;
                break;
            case E2BIG:
                goto convert;
            case EILSEQ:
                reason="Illegal multibyte sequence";
                if (!conv_silent)
                    DEBUG(3,("convert_string_allocate: Conversion error: %s(%s)\n",reason,inbuf));
                if (allow_bad_conv)
                    goto use_as_is;
                break;
        }
        if (!conv_silent)
            DEBUG(0,("Conversion error: %s(%s)\n",reason,inbuf));
        /* smb_panic(reason); */
        return (size_t)-1;
    }

  out:

    destlen = destlen - o_len;
    if (ctx) {
        ob = (char *)TALLOC_REALLOC(ctx,ob,destlen);
    } else {
        ob = (char *)SMB_REALLOC(ob,destlen);
    }

    if (destlen && !ob) {
        DEBUG(0, ("convert_string_allocate: out of memory!\n"));
        return (size_t)-1;
    }

    *dest = ob;
    return destlen;

 use_as_is:

    /*
     * Conversion not supported. This is actually an error, but there are so
     * many misconfigured iconv systems and smb.conf's out there we can't just
     * fail. Do a very bad conversion instead.... JRA.
     */

    {
        if (o_len == 0 || i_len == 0)
            goto out;

        if (((from == CH_UTF16LE)||(from == CH_UTF16BE)) &&
                ((to != CH_UTF16LE)||(to != CH_UTF16BE))) {
            /* Can't convert from utf16 any endian to multibyte.
               Replace with the default fail char.
            */

            if (i_len < 2)
                goto out;

            if (i_len >= 2) {
                *outbuf = lp_failed_convert_char();

                outbuf++;
                o_len--;

                inbuf += 2;
                i_len -= 2;
            }

            if (o_len == 0 || i_len == 0)
                goto out;

            /* Keep trying with the next char... */
            goto again;

        } else if (from != CH_UTF16LE && from != CH_UTF16BE && to == CH_UTF16LE) {
            /* Can't convert to UTF16LE - just widen by adding the
               default fail char then zero.
            */
            if (o_len < 2)
                goto out;

            outbuf[0] = lp_failed_convert_char();
            outbuf[1] = '\0';

            inbuf++;
            i_len--;

            outbuf += 2;
            o_len -= 2;

            if (o_len == 0 || i_len == 0)
                goto out;

            /* Keep trying with the next char... */
            goto again;

        } else if (from != CH_UTF16LE && from != CH_UTF16BE &&
                to != CH_UTF16LE && to != CH_UTF16BE) {
            /* Failed multibyte to multibyte. Just copy the default fail char and
               try again. */
            outbuf[0] = lp_failed_convert_char();

            inbuf++;
            i_len--;

            outbuf++;
            o_len--;

            if (o_len == 0 || i_len == 0)
                goto out;

            /* Keep trying with the next char... */
            goto again;

        } else {
            /* Keep compiler happy.... */
            goto out;
        }
    }
}

/**
 * Convert between character sets, allocating a new buffer using talloc for the result.
 *
 * @param srclen length of source buffer.
 * @param dest always set at least to NULL
 * @note -1 is not accepted for srclen.
 *
 * @returns Size in bytes of the converted string; or -1 in case of error.
 **/
size_t convert_string_talloc(TALLOC_CTX *ctx, charset_t from, charset_t to,
                 void const *src, size_t srclen, void *dst,
                 BOOL allow_bad_conv)
{
    void **dest = (void **)dst;
    size_t dest_len;

    *dest = NULL;
    dest_len=convert_string_allocate(ctx, from, to, src, srclen, dest, allow_bad_conv);
    if (dest_len == (size_t)-1)
        return (size_t)-1;
    if (*dest == NULL)
        return (size_t)-1;
    return dest_len;
}

size_t unix_strupper(const char *src, size_t srclen, char *dest, size_t destlen)
{
    size_t size;
    smb_ucs2_t *buffer;

    size = push_ucs2_allocate(&buffer, src);
    if (size == (size_t)-1) {
        smb_panic("failed to create UCS2 buffer");
    }
    if (!strupper_w(buffer) && (dest == src)) {
        free(buffer);
        return srclen;
    }

    size = convert_string(CH_UTF16LE, CH_UNIX, buffer, size, dest, destlen, True);
    free(buffer);
    return size;
}

/**
 strdup() a unix string to upper case.
 Max size is pstring.
**/

char *strdup_upper(const char *s)
{
    pstring out_buffer;
    const unsigned char *p = (const unsigned char *)s;
    unsigned char *q = (unsigned char *)out_buffer;

    /* this is quite a common operation, so we want it to be
       fast. We optimise for the ascii case, knowing that all our
       supported multi-byte character sets are ascii-compatible
       (ie. they match for the first 128 chars) */

    while (1) {
        if (*p & 0x80)
            break;
        *q++ = toupper_ascii(*p);
        if (!*p)
            break;
        p++;
        if (p - ( const unsigned char *)s >= sizeof(pstring))
            break;
    }

    if (*p) {
        /* MB case. */
        size_t size;
        wpstring buffer;
        size = convert_string(CH_UNIX, CH_UTF16LE, s, -1, buffer, sizeof(buffer), True);
        if (size == (size_t)-1) {
            return NULL;
        }

        strupper_w(buffer);

        size = convert_string(CH_UTF16LE, CH_UNIX, buffer, -1, out_buffer, sizeof(out_buffer), True);
        if (size == (size_t)-1) {
            return NULL;
        }
    }

    return SMB_STRDUP(out_buffer);
}

size_t unix_strlower(const char *src, size_t srclen, char *dest, size_t destlen)
{
    size_t size;
    smb_ucs2_t *buffer = NULL;

    size = convert_string_allocate(NULL, CH_UNIX, CH_UTF16LE, src, srclen,
                       (void **)(void *)&buffer, True);
    if (size == (size_t)-1 || !buffer) {
        smb_panic("failed to create UCS2 buffer");
    }
    if (!strlower_w(buffer) && (dest == src)) {
        SAFE_FREE(buffer);
        return srclen;
    }
    size = convert_string(CH_UTF16LE, CH_UNIX, buffer, size, dest, destlen, True);
    SAFE_FREE(buffer);
    return size;
}

/**
 strdup() a unix string to lower case.
**/

char *strdup_lower(const char *s)
{
    size_t size;
    smb_ucs2_t *buffer = NULL;
    char *out_buffer;

    size = push_ucs2_allocate(&buffer, s);
    if (size == -1 || !buffer) {
        return NULL;
    }

    strlower_w(buffer);

    size = pull_ucs2_allocate(&out_buffer, buffer);
    SAFE_FREE(buffer);

    if (size == (size_t)-1) {
        return NULL;
    }

    return out_buffer;
}

static size_t ucs2_align(const void *base_ptr, const void *p, int flags)
{
    if (flags & (STR_NOALIGN|STR_ASCII))
        return 0;
    return PTR_DIFF(p, base_ptr) & 1;
}


/**
 * Copy a string from a char* unix src to a dos codepage string destination.
 *
 * @return the number of bytes occupied by the string in the destination.
 *
 * @param flags can include
 * <dl>
 * <dt>STR_TERMINATE</dt> <dd>means include the null termination</dd>
 * <dt>STR_UPPER</dt> <dd>means uppercase in the destination</dd>
 * </dl>
 *
 * @param dest_len the maximum length in bytes allowed in the
 * destination.  If @p dest_len is -1 then no maximum is used.
 **/
size_t push_ascii(void *dest, const char *src, size_t dest_len, int flags)
{
        size_t src_len = strlen(src);
        pstring tmpbuf;
        size_t ret;

        /* No longer allow a length of -1 */
            if (dest_len == (size_t)-1)
                smb_panic("push_ascii - dest_len == -1");

    if (flags & STR_UPPER) {
        pstrcpy(tmpbuf, src);
        strupper_m(tmpbuf);
        src = tmpbuf;
    }

    if (flags & (STR_TERMINATE | STR_TERMINATE_ASCII))
        src_len++;

        ret =convert_string(CH_UNIX, CH_DOS, src, src_len, dest, dest_len, True);
        if (ret == (size_t)-1 &&
                        (flags & (STR_TERMINATE | STR_TERMINATE_ASCII))
                        && dest_len > 0) {
                ((char *)dest)[0] = '\0';
        }
        return ret;

}

size_t push_ascii_fstring(void *dest, const char *src)
{
    return push_ascii(dest, src, sizeof(fstring), STR_TERMINATE);
}

size_t push_ascii_pstring(void *dest, const char *src)
{
    return push_ascii(dest, src, sizeof(pstring), STR_TERMINATE);
}

/********************************************************************
 Push an nstring - ensure null terminated. Written by
 moriyama@miraclelinux.com (MORIYAMA Masayuki).
********************************************************************/

size_t push_ascii_nstring(void *dest, const char *src)
{
    size_t i, buffer_len, dest_len;
    smb_ucs2_t *buffer;

    conv_silent = True;
    buffer_len = push_ucs2_allocate(&buffer, src);
    if (buffer_len == (size_t)-1) {
        smb_panic("failed to create UCS2 buffer");
    }

    /* We're using buffer_len below to count ucs2 characters, not bytes. */
    buffer_len /= sizeof(smb_ucs2_t);

    dest_len = 0;
    for (i = 0; buffer[i] != 0 && (i < buffer_len); i++) {
        unsigned char mb[10];
        /* Convert one smb_ucs2_t character at a time. */
        size_t mb_len = convert_string(CH_UTF16LE, CH_DOS, buffer+i, sizeof(smb_ucs2_t), mb, sizeof(mb), False);
        if ((mb_len != (size_t)-1) && (dest_len + mb_len <= MAX_NETBIOSNAME_LEN - 1)) {
            memcpy((char *)dest + dest_len, mb, mb_len);
            dest_len += mb_len;
        } else {
            errno = E2BIG;
            break;
        }
    }
    ((char *)dest)[dest_len] = '\0';

    SAFE_FREE(buffer);
    conv_silent = False;
    return dest_len;
}

/**
 * Copy a string from a dos codepage source to a unix char* destination.
 *
 * The resulting string in "dest" is always null terminated.
 *
 * @param flags can have:
 * <dl>
 * <dt>STR_TERMINATE</dt>
 * <dd>STR_TERMINATE means the string in @p src
 * is null terminated, and src_len is ignored.</dd>
 * </dl>
 *
 * @param src_len is the length of the source area in bytes.
 * @returns the number of bytes occupied by the string in @p src.
 **/
size_t pull_ascii(char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
{
    size_t ret;

    if (dest_len == (size_t)-1)
        dest_len = sizeof(pstring);

    if (flags & STR_TERMINATE) {
        if (src_len == (size_t)-1) {
            src_len = strlen((const char *)src) + 1;
        } else {
            size_t len = strnlen((const char *)src, src_len);
            if (len < src_len)
                len++;
            src_len = len;
        }
    }

    ret = convert_string(CH_DOS, CH_UNIX, src, src_len, dest, dest_len, True);
    if (ret == (size_t)-1) {
        ret = 0;
        dest_len = 0;
    }

    if (dest_len && ret) {
        /* Did we already process the terminating zero ? */
        if (dest[MIN(ret-1, dest_len-1)] != 0) {
            dest[MIN(ret, dest_len-1)] = 0;
        }
    } else  {
        dest[0] = 0;
    }

    return src_len;
}

size_t pull_ascii_pstring(char *dest, const void *src)
{
    return pull_ascii(dest, src, sizeof(pstring), -1, STR_TERMINATE);
}

size_t pull_ascii_fstring(char *dest, const void *src)
{
    return pull_ascii(dest, src, sizeof(fstring), -1, STR_TERMINATE);
}

/* When pulling an nstring it can expand into a larger size (dos cp -> utf8). Cope with this. */

size_t pull_ascii_nstring(char *dest, size_t dest_len, const void *src)
{
    return pull_ascii(dest, src, dest_len, sizeof(nstring)-1, STR_TERMINATE);
}

/**
 * Copy a string from a char* src to a unicode destination.
 *
 * @returns the number of bytes occupied by the string in the destination.
 *
 * @param flags can have:
 *
 * <dl>
 * <dt>STR_TERMINATE <dd>means include the null termination.
 * <dt>STR_UPPER     <dd>means uppercase in the destination.
 * <dt>STR_NOALIGN   <dd>means don't do alignment.
 * </dl>
 *
 * @param dest_len is the maximum length allowed in the
 * destination. If dest_len is -1 then no maxiumum is used.
 **/

size_t push_ucs2(const void *base_ptr, void *dest, const char *src, size_t dest_len, int flags)
{
    size_t len=0;
    size_t src_len;
    size_t ret;

    /* treat a pstring as "unlimited" length */
    if (dest_len == (size_t)-1)
        dest_len = sizeof(pstring);

    if (flags & STR_TERMINATE)
        src_len = (size_t)-1;
    else
        src_len = strlen(src);

    if (ucs2_align(base_ptr, dest, flags)) {
        *(char *)dest = 0;
        dest = (void *)((char *)dest + 1);
        if (dest_len)
            dest_len--;
        len++;
    }

    /* ucs2 is always a multiple of 2 bytes */
    dest_len &= ~1;

    ret =  convert_string(CH_UNIX, CH_UTF16LE, src, src_len, dest, dest_len, True);
    if (ret == (size_t)-1) {
        return 0;
    }

    len += ret;

    if (flags & STR_UPPER) {
        smb_ucs2_t *dest_ucs2 = (smb_ucs2_t *)dest;
        size_t i;

        /* We check for i < (ret / 2) below as the dest string isn't null
           terminated if STR_TERMINATE isn't set. */

        for (i = 0; i < (ret / 2) && i < (dest_len / 2) && dest_ucs2[i]; i++) {
            smb_ucs2_t v = toupper_w(dest_ucs2[i]);
            if (v != dest_ucs2[i]) {
                dest_ucs2[i] = v;
            }
        }
    }

    return len;
}


/**
 * Copy a string from a unix char* src to a UCS2 destination,
 * allocating a buffer using talloc().
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 *         or -1 in case of error.
 **/
size_t push_ucs2_talloc(TALLOC_CTX *ctx, smb_ucs2_t **dest, const char *src)
{
    size_t src_len = strlen(src)+1;

    *dest = NULL;
    return convert_string_talloc(ctx, CH_UNIX, CH_UTF16LE, src, src_len, (void **)dest, True);
}


/**
 * Copy a string from a unix char* src to a UCS2 destination, allocating a buffer
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 *         or -1 in case of error.
 **/

size_t push_ucs2_allocate(smb_ucs2_t **dest, const char *src)
{
    size_t src_len = strlen(src)+1;

    *dest = NULL;
    return convert_string_allocate(NULL, CH_UNIX, CH_UTF16LE, src, src_len, (void **)dest, True);
}

/**
 Copy a string from a char* src to a UTF-8 destination.
 Return the number of bytes occupied by the string in the destination
 Flags can have:
  STR_TERMINATE means include the null termination
  STR_UPPER     means uppercase in the destination
 dest_len is the maximum length allowed in the destination. If dest_len
 is -1 then no maxiumum is used.
**/

static size_t push_utf8(void *dest, const char *src, size_t dest_len, int flags)
{
    size_t src_len = strlen(src);
    pstring tmpbuf;

    /* treat a pstring as "unlimited" length */
    if (dest_len == (size_t)-1)
        dest_len = sizeof(pstring);

    if (flags & STR_UPPER) {
        pstrcpy(tmpbuf, src);
        strupper_m(tmpbuf);
        src = tmpbuf;
    }

    if (flags & STR_TERMINATE)
        src_len++;

    return convert_string(CH_UNIX, CH_UTF8, src, src_len, dest, dest_len, True);
}

size_t push_utf8_fstring(void *dest, const char *src)
{
    return push_utf8(dest, src, sizeof(fstring), STR_TERMINATE);
}

/**
 * Copy a string from a unix char* src to a UTF-8 destination, allocating a buffer using talloc
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t push_utf8_talloc(TALLOC_CTX *ctx, char **dest, const char *src)
{
    size_t src_len = strlen(src)+1;

    *dest = NULL;
    return convert_string_talloc(ctx, CH_UNIX, CH_UTF8, src, src_len, (void**)dest, True);
}

/**
 * Copy a string from a unix char* src to a UTF-8 destination, allocating a buffer
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t push_utf8_allocate(char **dest, const char *src)
{
    size_t src_len = strlen(src)+1;

    *dest = NULL;
    return convert_string_allocate(NULL, CH_UNIX, CH_UTF8, src, src_len, (void **)dest, True);
}

/**
 Copy a string from a ucs2 source to a unix char* destination.
 Flags can have:
  STR_TERMINATE means the string in src is null terminated.
  STR_NOALIGN   means don't try to align.
 if STR_TERMINATE is set then src_len is ignored if it is -1.
 src_len is the length of the source area in bytes
 Return the number of bytes occupied by the string in src.
 The resulting string in "dest" is always null terminated.
**/

size_t pull_ucs2(const void *base_ptr, char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
{
    size_t ret;

    if (dest_len == (size_t)-1)
        dest_len = sizeof(pstring);

    if (ucs2_align(base_ptr, src, flags)) {
        src = (const void *)((const char *)src + 1);
        if (src_len != (size_t)-1)
            src_len--;
    }

    if (flags & STR_TERMINATE) {
        /* src_len -1 is the default for null terminated strings. */
        if (src_len != (size_t)-1) {
            size_t len = strnlen_w((const smb_ucs2_t *)src,
                        src_len/2);
            if (len < src_len/2)
                len++;
            src_len = len*2;
        }
    }

    /* ucs2 is always a multiple of 2 bytes */
    if (src_len != (size_t)-1)
        src_len &= ~1;

    ret = convert_string(CH_UTF16LE, CH_UNIX, src, src_len, dest, dest_len, True);
    if (ret == (size_t)-1) {
        return 0;
    }

    if (src_len == (size_t)-1)
        src_len = ret*2;

    if (dest_len && ret) {
        /* Did we already process the terminating zero ? */
        if (dest[MIN(ret-1, dest_len-1)] != 0) {
            dest[MIN(ret, dest_len-1)] = 0;
        }
    } else {
        dest[0] = 0;
    }

    return src_len;
}

size_t pull_ucs2_pstring(char *dest, const void *src)
{
    return pull_ucs2(NULL, dest, src, sizeof(pstring), -1, STR_TERMINATE);
}

size_t pull_ucs2_fstring(char *dest, const void *src)
{
    return pull_ucs2(NULL, dest, src, sizeof(fstring), -1, STR_TERMINATE);
}

/**
 * Copy a string from a UCS2 src to a unix char * destination, allocating a buffer using talloc
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_ucs2_talloc(TALLOC_CTX *ctx, char **dest, const smb_ucs2_t *src)
{
    size_t src_len = (strlen_w(src)+1) * sizeof(smb_ucs2_t);
    *dest = NULL;
    return convert_string_talloc(ctx, CH_UTF16LE, CH_UNIX, src, src_len, (void **)dest, True);
}

/**
 * Copy a string from a UCS2 src to a unix char * destination, allocating a buffer
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_ucs2_allocate(char **dest, const smb_ucs2_t *src)
{
    size_t src_len = (strlen_w(src)+1) * sizeof(smb_ucs2_t);
    *dest = NULL;
    return convert_string_allocate(NULL, CH_UTF16LE, CH_UNIX, src, src_len, (void **)dest, True);
}

/**
 * Copy a string from a UTF-8 src to a unix char * destination, allocating a buffer using talloc
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_utf8_talloc(TALLOC_CTX *ctx, char **dest, const char *src)
{
    size_t src_len = strlen(src)+1;
    *dest = NULL;
    return convert_string_talloc(ctx, CH_UTF8, CH_UNIX, src, src_len, (void **)dest, True);
}

/**
 * Copy a string from a UTF-8 src to a unix char * destination, allocating a buffer
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_utf8_allocate(char **dest, const char *src)
{
    size_t src_len = strlen(src)+1;
    *dest = NULL;
    return convert_string_allocate(NULL, CH_UTF8, CH_UNIX, src, src_len, (void **)dest, True);
}

/**
 * Copy a string from a DOS src to a unix char * destination, allocating a buffer using talloc
 *
 * @param dest always set at least to NULL
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_ascii_talloc(TALLOC_CTX *ctx, char **dest, const char *src)
{
    size_t src_len = strlen(src)+1;
    *dest = NULL;
    return convert_string_talloc(ctx, CH_DOS, CH_UNIX, src, src_len, (void **)dest, True);
}

/**
 Copy a string from a char* src to a unicode or ascii
 dos codepage destination choosing unicode or ascii based on the
 flags in the SMB buffer starting at base_ptr.
 Return the number of bytes occupied by the string in the destination.
 flags can have:
  STR_TERMINATE means include the null termination.
  STR_UPPER     means uppercase in the destination.
  STR_ASCII     use ascii even with unicode packet.
  STR_NOALIGN   means don't do alignment.
 dest_len is the maximum length allowed in the destination. If dest_len
 is -1 then no maxiumum is used.
**/

size_t push_string_fn(const char *function, unsigned int line, const void *base_ptr, void *dest, const char *src, size_t dest_len, int flags)
{
#ifdef DEVELOPER
    /* We really need to zero fill here, not clobber
     * region, as we want to ensure that valgrind thinks
     * all of the outgoing buffer has been written to
     * so a send() or write() won't trap an error.
     * JRA.
     */
#if 0
    if (dest_len != (size_t)-1)
        clobber_region(function, line, dest, dest_len);
#else
    if (dest_len != (size_t)-1)
        memset(dest, '\0', dest_len);
#endif
#endif

    if (!(flags & STR_ASCII) && \
        ((flags & STR_UNICODE || \
          (SVAL(base_ptr, smb_flg2) & FLAGS2_UNICODE_STRINGS)))) {
        return push_ucs2(base_ptr, dest, src, dest_len, flags);
    }
    return push_ascii(dest, src, dest_len, flags);
}


/**
 Copy a string from a unicode or ascii source (depending on
 the packet flags) to a char* destination.
 Flags can have:
  STR_TERMINATE means the string in src is null terminated.
  STR_UNICODE   means to force as unicode.
  STR_ASCII     use ascii even with unicode packet.
  STR_NOALIGN   means don't do alignment.
 if STR_TERMINATE is set then src_len is ignored is it is -1
 src_len is the length of the source area in bytes.
 Return the number of bytes occupied by the string in src.
 The resulting string in "dest" is always null terminated.
**/

size_t pull_string_fn(const char *function, unsigned int line, const void *base_ptr, char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
{
#ifdef DEVELOPER
    if (dest_len != (size_t)-1)
        clobber_region(function, line, dest, dest_len);
#endif

    if (!(flags & STR_ASCII) && \
        ((flags & STR_UNICODE || \
          (SVAL(base_ptr, smb_flg2) & FLAGS2_UNICODE_STRINGS)))) {
        return pull_ucs2(base_ptr, dest, src, dest_len, src_len, flags);
    }
    return pull_ascii(dest, src, dest_len, src_len, flags);
}

size_t align_string(const void *base_ptr, const char *p, int flags)
{
    if (!(flags & STR_ASCII) && \
        ((flags & STR_UNICODE || \
          (SVAL(base_ptr, smb_flg2) & FLAGS2_UNICODE_STRINGS)))) {
        return ucs2_align(base_ptr, p, flags);
    }
    return 0;
}

/*
  Return the unicode codepoint for the next multi-byte CH_UNIX character
  in the string. The unicode codepoint (codepoint_t) is an unsinged 32 bit value.

  Also return the number of bytes consumed (which tells the caller
  how many bytes to skip to get to the next CH_UNIX character).

  Return INVALID_CODEPOINT if the next character cannot be converted.
*/

codepoint_t next_codepoint(const char *str, size_t *size)
{
    /* It cannot occupy more than 4 bytes in UTF16 format */
    uint8_t buf[4];
    smb_iconv_t descriptor;
#ifdef __OS2__
    size_t ilen_max;
#endif
    size_t ilen_orig;
    size_t ilen;
    size_t olen_orig;
    size_t olen;
    const char *inbuf;
    char *outbuf;

#ifdef __OS2__
    *size = 1;
#endif

    if ((str[0] & 0x80) == 0) {
#ifndef __OS2__
        *size = 1;
#endif
        return (codepoint_t)str[0];
    }

    lazy_initialize_conv();

    descriptor = conv_handles[CH_UNIX][CH_UTF16LE];
    if (descriptor == (smb_iconv_t)-1 || descriptor == (smb_iconv_t)0) {
#ifndef __OS2__
        *size = 1;
#endif
        return INVALID_CODEPOINT;
    }
#ifdef __OS2__
    /* We assume that no multi-byte character can take
       more than 5 bytes. This is OK as we only
       support codepoints up to 1M */

    ilen_max = strnlen( str, 5 );
#else
    *size = 1;
#endif
    ilen_orig = 1;
    olen_orig = 2;
    while( 1 )
    {
        ilen = ilen_orig;
        olen = olen_orig;
        inbuf = str;
        outbuf = ( char * )buf;
        if( smb_iconv( descriptor, &inbuf, &ilen, &outbuf, &olen ) != ( size_t )-1 )
            break;

        switch( errno )
        {
            case E2BIG :
                if( olen_orig == 2 )
                    olen_orig = 4;
                else
                    return INVALID_CODEPOINT;
                break;

            case EINVAL :
#ifndef __OS2__
                /* We assume that no multi-byte character can take
                   more than 5 bytes. This is OK as we only
                   support codepoints up to 1M */
                if( ilen_orig < 5 )
#else
                if( ilen_orig < ilen_max )
#endif
                    ilen_orig++;
                else
                    return INVALID_CODEPOINT;
                break;

            case EILSEQ :
            default :
                return INVALID_CODEPOINT;
        }
    }

    olen = olen_orig - olen;

    *size = ilen_orig - ilen;

    if (olen == 2) {
        /* 2 byte, UTF16-LE encoded value. */
        return (codepoint_t)SVAL(buf, 0);
    }
    if (olen == 4) {
        /* Decode a 4 byte UTF16-LE character manually.
           See RFC2871 for the encoding machanism.
        */
        codepoint_t w1 = SVAL(buf,0) & ~0xD800;
        codepoint_t w2 = SVAL(buf,2) & ~0xDC00;

        return (codepoint_t)0x10000 +
                (w1 << 10) + w2;
    }

    /* no other length is valid */
    return INVALID_CODEPOINT;
}