DXR is a code search and navigation tool aimed at making sense of large projects. It supports full-text and regex searches as well as structural queries.

Header

Mercurial (777e60ca8853)

VCS Links

Line Code
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011
/* This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */

#include "xpcom-private.h"

//-----------------------------------------------------------------------------
// XP_MACOSX or ANDROID
//-----------------------------------------------------------------------------
#if defined(XP_MACOSX) || defined(ANDROID)

#include "nsAString.h"
#include "nsReadableUtils.h"
#include "nsString.h"

nsresult
NS_CopyNativeToUnicode(const nsACString &input, nsAString  &output)
{
    CopyUTF8toUTF16(input, output);
    return NS_OK;
}

nsresult
NS_CopyUnicodeToNative(const nsAString  &input, nsACString &output)
{
    CopyUTF16toUTF8(input, output);
    return NS_OK;
}

void
NS_StartupNativeCharsetUtils()
{
}

void
NS_ShutdownNativeCharsetUtils()
{
}


//-----------------------------------------------------------------------------
// XP_UNIX
//-----------------------------------------------------------------------------
#elif defined(XP_UNIX)

#include <stdlib.h>   // mbtowc, wctomb
#include <locale.h>   // setlocale
#include "mozilla/Mutex.h"
#include "nscore.h"
#include "nsAString.h"
#include "nsReadableUtils.h"

using namespace mozilla;

//
// choose a conversion library.  we used to use mbrtowc/wcrtomb under Linux,
// but that doesn't work for non-BMP characters whether we use '-fshort-wchar'
// or not (see bug 206811 and 
// news://news.mozilla.org:119/bajml3$fvr1@ripley.netscape.com). we now use
// iconv for all platforms where nltypes.h and nllanginfo.h are present 
// along with iconv.
//
#if defined(HAVE_ICONV) && defined(HAVE_NL_TYPES_H) && defined(HAVE_LANGINFO_CODESET)
#define USE_ICONV 1
#else
#define USE_STDCONV 1
#endif

static void
isolatin1_to_utf16(const char **input, uint32_t *inputLeft, char16_t **output, uint32_t *outputLeft)
{
    while (*inputLeft && *outputLeft) {
        **output = (unsigned char) **input;
        (*input)++;
        (*inputLeft)--;
        (*output)++;
        (*outputLeft)--;
    }
}

static void
utf16_to_isolatin1(const char16_t **input, uint32_t *inputLeft, char **output, uint32_t *outputLeft)
{
    while (*inputLeft && *outputLeft) {
        **output = (unsigned char) **input;
        (*input)++;
        (*inputLeft)--;
        (*output)++;
        (*outputLeft)--;
    }
}

//-----------------------------------------------------------------------------
// conversion using iconv
//-----------------------------------------------------------------------------
#if defined(USE_ICONV)
#include <nl_types.h> // CODESET
#include <langinfo.h> // nl_langinfo
#include <iconv.h>    // iconv_open, iconv, iconv_close
#include <errno.h>
#include "plstr.h"

#if defined(HAVE_ICONV_WITH_CONST_INPUT)
#define ICONV_INPUT(x) (x)
#else
#define ICONV_INPUT(x) ((char **)x)
#endif

// solaris definitely needs this, but we'll enable it by default
// just in case... but we know for sure that iconv(3) in glibc
// doesn't need this.
#if !defined(__GLIBC__)
#define ENABLE_UTF8_FALLBACK_SUPPORT
#endif

#define INVALID_ICONV_T ((iconv_t) -1)

static inline size_t
xp_iconv(iconv_t converter,
         const char **input,
         size_t      *inputLeft,
         char       **output,
         size_t      *outputLeft)
{
    size_t res, outputAvail = outputLeft ? *outputLeft : 0;
    res = iconv(converter, ICONV_INPUT(input), inputLeft, output, outputLeft);
    if (res == (size_t) -1) {
        // on some platforms (e.g., linux) iconv will fail with
        // E2BIG if it cannot convert _all_ of its input.  it'll
        // still adjust all of the in/out params correctly, so we
        // can ignore this error.  the assumption is that we will
        // be called again to complete the conversion.
        if ((errno == E2BIG) && (*outputLeft < outputAvail))
            res = 0;
    }
    return res;
}

static inline void
xp_iconv_reset(iconv_t converter)
{
    // NOTE: the man pages on Solaris claim that you can pass nullptr
    // for all parameter to reset the converter, but beware the 
    // evil Solaris crash if you go down this route >:-)
    
    const char *zero_char_in_ptr  = nullptr;
    char       *zero_char_out_ptr = nullptr;
    size_t      zero_size_in      = 0,
                zero_size_out     = 0;

    xp_iconv(converter, &zero_char_in_ptr,
                        &zero_size_in,
                        &zero_char_out_ptr,
                        &zero_size_out);
}

static inline iconv_t
xp_iconv_open(const char **to_list, const char **from_list)
{
    iconv_t res;
    const char **from_name;
    const char **to_name;

    // try all possible combinations to locate a converter.
    to_name = to_list;
    while (*to_name) {
        if (**to_name) {
            from_name = from_list;
            while (*from_name) {
                if (**from_name) {
                    res = iconv_open(*to_name, *from_name);
                    if (res != INVALID_ICONV_T)
                        return res;
                }
                from_name++;
            }
        }
        to_name++;
    }

    return INVALID_ICONV_T;
}

/* 
 * char16_t[] is NOT a UCS-2 array BUT a UTF-16 string. Therefore, we
 * have to use UTF-16 with iconv(3) on platforms where it's supported.
 * However, the way UTF-16 and UCS-2 are interpreted varies across platforms 
 * and implementations of iconv(3). On Tru64, it also depends on the environment
 * variable. To avoid the trouble arising from byte-swapping 
 * (bug 208809), we have to try UTF-16LE/BE and UCS-2LE/BE before falling 
 * back to UTF-16 and UCS-2 and variants. We assume that UTF-16 and UCS-2 
 * on systems without UTF-16LE/BE and UCS-2LE/BE have the native endianness,
 * which isn't the case of glibc 2.1.x, for which we use 'UNICODELITTLE'
 * and 'UNICODEBIG'. It's also not true of Tru64 V4 when the environment
 * variable ICONV_BYTEORDER is set to 'big-endian', about which not much 
 * can be done other than adding a note in the release notes. (bug 206811)
 */
static const char *UTF_16_NAMES[] = {
#if defined(IS_LITTLE_ENDIAN)
    "UTF-16LE",
#if defined(__GLIBC__)
    "UNICODELITTLE",
#endif
    "UCS-2LE",
#else
    "UTF-16BE",
#if defined(__GLIBC__)
    "UNICODEBIG",
#endif
    "UCS-2BE",
#endif
    "UTF-16",
    "UCS-2",
    "UCS2",
    "UCS_2",
    "ucs-2",
    "ucs2",
    "ucs_2",
    nullptr
};

#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
static const char *UTF_8_NAMES[] = {
    "UTF-8",
    "UTF8",
    "UTF_8",
    "utf-8",
    "utf8",
    "utf_8",
    nullptr
};
#endif

static const char *ISO_8859_1_NAMES[] = {
    "ISO-8859-1",
#if !defined(__GLIBC__)
    "ISO8859-1",
    "ISO88591",
    "ISO_8859_1",
    "ISO8859_1",
    "iso-8859-1",
    "iso8859-1",
    "iso88591",
    "iso_8859_1",
    "iso8859_1",
#endif
    nullptr
};

class nsNativeCharsetConverter
{
public:
    nsNativeCharsetConverter();
   ~nsNativeCharsetConverter();

    nsresult NativeToUnicode(const char      **input , uint32_t *inputLeft,
                             char16_t       **output, uint32_t *outputLeft);
    nsresult UnicodeToNative(const char16_t **input , uint32_t *inputLeft,
                             char            **output, uint32_t *outputLeft);

    static void GlobalInit();
    static void GlobalShutdown();
    static bool IsNativeUTF8();

private:
    static iconv_t gNativeToUnicode;
    static iconv_t gUnicodeToNative;
#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
    static iconv_t gNativeToUTF8;
    static iconv_t gUTF8ToNative;
    static iconv_t gUnicodeToUTF8;
    static iconv_t gUTF8ToUnicode;
#endif
    static Mutex  *gLock;
    static bool    gInitialized;
    static bool    gIsNativeUTF8;

    static void LazyInit();

    static void Lock()   { if (gLock) gLock->Lock();   }
    static void Unlock() { if (gLock) gLock->Unlock(); }
};

iconv_t nsNativeCharsetConverter::gNativeToUnicode = INVALID_ICONV_T;
iconv_t nsNativeCharsetConverter::gUnicodeToNative = INVALID_ICONV_T;
#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
iconv_t nsNativeCharsetConverter::gNativeToUTF8    = INVALID_ICONV_T;
iconv_t nsNativeCharsetConverter::gUTF8ToNative    = INVALID_ICONV_T;
iconv_t nsNativeCharsetConverter::gUnicodeToUTF8   = INVALID_ICONV_T;
iconv_t nsNativeCharsetConverter::gUTF8ToUnicode   = INVALID_ICONV_T;
#endif
Mutex  *nsNativeCharsetConverter::gLock            = nullptr;
bool    nsNativeCharsetConverter::gInitialized     = false;
bool    nsNativeCharsetConverter::gIsNativeUTF8    = false;

void
nsNativeCharsetConverter::LazyInit()
{
    // LazyInit may be called before NS_StartupNativeCharsetUtils, but
    // the setlocale it does has to be called before nl_langinfo. Like in
    // NS_StartupNativeCharsetUtils, assume we are called early enough that
    // we are the first to care about the locale's charset.
    if (!gLock)
      setlocale(LC_CTYPE, "");
    const char  *blank_list[] = { "", nullptr };
    const char **native_charset_list = blank_list;
    const char  *native_charset = nl_langinfo(CODESET);
    if (native_charset == nullptr) {
        NS_ERROR("native charset is unknown");
        // fallback to ISO-8859-1
        native_charset_list = ISO_8859_1_NAMES;
    }
    else
        native_charset_list[0] = native_charset;

    // Most, if not all, Unixen supporting UTF-8 and nl_langinfo(CODESET) 
    // return 'UTF-8' (or 'utf-8')
    if (!PL_strcasecmp(native_charset, "UTF-8"))
        gIsNativeUTF8 = true;

    gNativeToUnicode = xp_iconv_open(UTF_16_NAMES, native_charset_list);
    gUnicodeToNative = xp_iconv_open(native_charset_list, UTF_16_NAMES);

#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
    if (gNativeToUnicode == INVALID_ICONV_T) {
        gNativeToUTF8 = xp_iconv_open(UTF_8_NAMES, native_charset_list);
        gUTF8ToUnicode = xp_iconv_open(UTF_16_NAMES, UTF_8_NAMES);
        NS_ASSERTION(gNativeToUTF8 != INVALID_ICONV_T, "no native to utf-8 converter");
        NS_ASSERTION(gUTF8ToUnicode != INVALID_ICONV_T, "no utf-8 to utf-16 converter");
    }
    if (gUnicodeToNative == INVALID_ICONV_T) {
        gUnicodeToUTF8 = xp_iconv_open(UTF_8_NAMES, UTF_16_NAMES);
        gUTF8ToNative = xp_iconv_open(native_charset_list, UTF_8_NAMES);
        NS_ASSERTION(gUnicodeToUTF8 != INVALID_ICONV_T, "no utf-16 to utf-8 converter");
        NS_ASSERTION(gUTF8ToNative != INVALID_ICONV_T, "no utf-8 to native converter");
    }
#else
    NS_ASSERTION(gNativeToUnicode != INVALID_ICONV_T, "no native to utf-16 converter");
    NS_ASSERTION(gUnicodeToNative != INVALID_ICONV_T, "no utf-16 to native converter");
#endif

    /*
     * On Solaris 8 (and newer?), the iconv modules converting to UCS-2
     * prepend a byte order mark unicode character (BOM, u+FEFF) during
     * the first use of the iconv converter. The same is the case of 
     * glibc 2.2.9x and Tru64 V5 (see bug 208809) when 'UTF-16' is used. 
     * However, we use 'UTF-16LE/BE' in both cases, instead so that we 
     * should be safe. But just in case...
     *
     * This dummy conversion gets rid of the BOMs and fixes bug 153562.
     */
    char dummy_input[1] = { ' ' };
    char dummy_output[4];

    if (gNativeToUnicode != INVALID_ICONV_T) {
	const char *input = dummy_input;
	size_t input_left = sizeof(dummy_input);
	char *output = dummy_output;
	size_t output_left = sizeof(dummy_output);

	xp_iconv(gNativeToUnicode, &input, &input_left, &output, &output_left);
    }
#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
    if (gUTF8ToUnicode != INVALID_ICONV_T) {
	const char *input = dummy_input;
	size_t input_left = sizeof(dummy_input);
	char *output = dummy_output;
	size_t output_left = sizeof(dummy_output);

	xp_iconv(gUTF8ToUnicode, &input, &input_left, &output, &output_left);
    }
#endif

    gInitialized = true;
}

void
nsNativeCharsetConverter::GlobalInit()
{
    gLock = new Mutex("nsNativeCharsetConverter.gLock");
}

void
nsNativeCharsetConverter::GlobalShutdown()
{
    if (gLock) {
        delete gLock;
        gLock = nullptr;
    }

    if (gNativeToUnicode != INVALID_ICONV_T) {
        iconv_close(gNativeToUnicode);
        gNativeToUnicode = INVALID_ICONV_T;
    }

    if (gUnicodeToNative != INVALID_ICONV_T) {
        iconv_close(gUnicodeToNative);
        gUnicodeToNative = INVALID_ICONV_T;
    }

#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
    if (gNativeToUTF8 != INVALID_ICONV_T) {
        iconv_close(gNativeToUTF8);
        gNativeToUTF8 = INVALID_ICONV_T;
    }
    if (gUTF8ToNative != INVALID_ICONV_T) {
        iconv_close(gUTF8ToNative);
        gUTF8ToNative = INVALID_ICONV_T;
    }
    if (gUnicodeToUTF8 != INVALID_ICONV_T) {
        iconv_close(gUnicodeToUTF8);
        gUnicodeToUTF8 = INVALID_ICONV_T;
    }
    if (gUTF8ToUnicode != INVALID_ICONV_T) {
        iconv_close(gUTF8ToUnicode);
        gUTF8ToUnicode = INVALID_ICONV_T;
    }
#endif

    gInitialized = false;
}

nsNativeCharsetConverter::nsNativeCharsetConverter()
{
    Lock();
    if (!gInitialized)
        LazyInit();
}

nsNativeCharsetConverter::~nsNativeCharsetConverter()
{
    // reset converters for next time
    if (gNativeToUnicode != INVALID_ICONV_T)
        xp_iconv_reset(gNativeToUnicode);
    if (gUnicodeToNative != INVALID_ICONV_T)
        xp_iconv_reset(gUnicodeToNative);
#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
    if (gNativeToUTF8 != INVALID_ICONV_T)
        xp_iconv_reset(gNativeToUTF8);
    if (gUTF8ToNative != INVALID_ICONV_T)
        xp_iconv_reset(gUTF8ToNative);
    if (gUnicodeToUTF8 != INVALID_ICONV_T)
        xp_iconv_reset(gUnicodeToUTF8);
    if (gUTF8ToUnicode != INVALID_ICONV_T)
        xp_iconv_reset(gUTF8ToUnicode);
#endif
    Unlock();
}

nsresult
nsNativeCharsetConverter::NativeToUnicode(const char **input,
                                          uint32_t    *inputLeft,
                                          char16_t  **output,
                                          uint32_t    *outputLeft)
{
    size_t res = 0;
    size_t inLeft = (size_t) *inputLeft;
    size_t outLeft = (size_t) *outputLeft * 2;

    if (gNativeToUnicode != INVALID_ICONV_T) {

        res = xp_iconv(gNativeToUnicode, input, &inLeft, (char **) output, &outLeft);

        *inputLeft = inLeft;
        *outputLeft = outLeft / 2;
        if (res != (size_t) -1) 
            return NS_OK;

        NS_WARNING("conversion from native to utf-16 failed");

        // reset converter
        xp_iconv_reset(gNativeToUnicode);
    }
#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
    else if ((gNativeToUTF8 != INVALID_ICONV_T) &&
             (gUTF8ToUnicode != INVALID_ICONV_T)) {
        // convert first to UTF8, then from UTF8 to UCS2
        const char *in = *input;

        char ubuf[1024];

        // we assume we're always called with enough space in |output|,
        // so convert many chars at a time...
        while (inLeft) {
            char *p = ubuf;
            size_t n = sizeof(ubuf);
            res = xp_iconv(gNativeToUTF8, &in, &inLeft, &p, &n);
            if (res == (size_t) -1) {
                NS_ERROR("conversion from native to utf-8 failed");
                break;
            }
            NS_ASSERTION(outLeft > 0, "bad assumption");
            p = ubuf;
            n = sizeof(ubuf) - n;
            res = xp_iconv(gUTF8ToUnicode, (const char **) &p, &n, (char **) output, &outLeft);
            if (res == (size_t) -1) {
                NS_ERROR("conversion from utf-8 to utf-16 failed");
                break;
            }
        }

        (*input) += (*inputLeft - inLeft);
        *inputLeft = inLeft;
        *outputLeft = outLeft / 2;

        if (res != (size_t) -1) 
            return NS_OK;

        // reset converters
        xp_iconv_reset(gNativeToUTF8);
        xp_iconv_reset(gUTF8ToUnicode);
    }
#endif

    // fallback: zero-pad and hope for the best
    // XXX This is lame and we have to do better.
    isolatin1_to_utf16(input, inputLeft, output, outputLeft);

    return NS_OK;
}

nsresult
nsNativeCharsetConverter::UnicodeToNative(const char16_t **input,
                                          uint32_t         *inputLeft,
                                          char            **output,
                                          uint32_t         *outputLeft)
{
    size_t res = 0;
    size_t inLeft = (size_t) *inputLeft * 2;
    size_t outLeft = (size_t) *outputLeft;

    if (gUnicodeToNative != INVALID_ICONV_T) {
        res = xp_iconv(gUnicodeToNative, (const char **) input, &inLeft, output, &outLeft);

        *inputLeft = inLeft / 2;
        *outputLeft = outLeft;
        if (res != (size_t) -1) {
            return NS_OK;
        }

        NS_ERROR("iconv failed");

        // reset converter
        xp_iconv_reset(gUnicodeToNative);
    }
#if defined(ENABLE_UTF8_FALLBACK_SUPPORT)
    else if ((gUnicodeToUTF8 != INVALID_ICONV_T) &&
             (gUTF8ToNative != INVALID_ICONV_T)) {
        const char *in = (const char *) *input;

        char ubuf[6]; // max utf-8 char length (really only needs to be 4 bytes)

        // convert one uchar at a time...
        while (inLeft && outLeft) {
            char *p = ubuf;
            size_t n = sizeof(ubuf), one_uchar = sizeof(char16_t);
            res = xp_iconv(gUnicodeToUTF8, &in, &one_uchar, &p, &n);
            if (res == (size_t) -1) {
                NS_ERROR("conversion from utf-16 to utf-8 failed");
                break;
            }
            p = ubuf;
            n = sizeof(ubuf) - n;
            res = xp_iconv(gUTF8ToNative, (const char **) &p, &n, output, &outLeft);
            if (res == (size_t) -1) {
                if (errno == E2BIG) {
                    // not enough room for last uchar... back up and return.
                    in -= sizeof(char16_t);
                    res = 0;
                }
                else
                    NS_ERROR("conversion from utf-8 to native failed");
                break;
            }
            inLeft -= sizeof(char16_t);
        }

        (*input) += (*inputLeft - inLeft / 2);
        *inputLeft = inLeft / 2;
        *outputLeft = outLeft;
        if (res != (size_t) -1) {
            return NS_OK;
        }

        // reset converters
        xp_iconv_reset(gUnicodeToUTF8);
        xp_iconv_reset(gUTF8ToNative);
    }
#endif

    // fallback: truncate and hope for the best
    // XXX This is lame and we have to do better.
    utf16_to_isolatin1(input, inputLeft, output, outputLeft);

    return NS_OK;
}

bool
nsNativeCharsetConverter::IsNativeUTF8()
{
    if (!gInitialized) {
        Lock();
        if (!gInitialized)
           LazyInit();
        Unlock();
    }
    return gIsNativeUTF8; 
}

#endif // USE_ICONV

//-----------------------------------------------------------------------------
// conversion using mb[r]towc/wc[r]tomb
//-----------------------------------------------------------------------------
#if defined(USE_STDCONV)
#if defined(HAVE_WCRTOMB) || defined(HAVE_MBRTOWC)
#include <wchar.h>    // mbrtowc, wcrtomb
#endif

class nsNativeCharsetConverter
{
public:
    nsNativeCharsetConverter();

    nsresult NativeToUnicode(const char      **input , uint32_t *inputLeft,
                             char16_t       **output, uint32_t *outputLeft);
    nsresult UnicodeToNative(const char16_t **input , uint32_t *inputLeft,
                             char            **output, uint32_t *outputLeft);

    static void GlobalInit();
    static void GlobalShutdown() { }
    static bool IsNativeUTF8();

private:
    static bool gWCharIsUnicode;

#if defined(HAVE_WCRTOMB) || defined(HAVE_MBRTOWC)
    mbstate_t ps;
#endif
};

bool nsNativeCharsetConverter::gWCharIsUnicode = false;

nsNativeCharsetConverter::nsNativeCharsetConverter()
{
#if defined(HAVE_WCRTOMB) || defined(HAVE_MBRTOWC)
    memset(&ps, 0, sizeof(ps));
#endif
}

void
nsNativeCharsetConverter::GlobalInit()
{
    // verify that wchar_t for the current locale is actually unicode.
    // if it is not, then we should avoid calling mbtowc/wctomb and
    // just fallback on zero-pad/truncation conversion.
    //
    // this test cannot be done at build time because the encoding of
    // wchar_t may depend on the runtime locale.  sad, but true!!
    //
    // so, if wchar_t is unicode then converting an ASCII character
    // to wchar_t should not change its numeric value.  we'll just
    // check what happens with the ASCII 'a' character.
    //
    // this test is not perfect... obviously, it could yield false
    // positives, but then at least ASCII text would be converted
    // properly (or maybe just the 'a' character) -- oh well :(

    char a = 'a';
    unsigned int w = 0;

    int res = mbtowc((wchar_t *) &w, &a, 1);

    gWCharIsUnicode = (res != -1 && w == 'a');

#ifdef DEBUG
    if (!gWCharIsUnicode)
        NS_WARNING("wchar_t is not unicode (unicode conversion will be lossy)");
#endif
}

nsresult
nsNativeCharsetConverter::NativeToUnicode(const char **input,
                                          uint32_t    *inputLeft,
                                          char16_t  **output,
                                          uint32_t    *outputLeft)
{
    if (gWCharIsUnicode) {
        int incr;

        // cannot use wchar_t here since it may have been redefined (e.g.,
        // via -fshort-wchar).  hopefully, sizeof(tmp) is sufficient XP.
        unsigned int tmp = 0;
        while (*inputLeft && *outputLeft) {
#ifdef HAVE_MBRTOWC
            incr = (int) mbrtowc((wchar_t *) &tmp, *input, *inputLeft, &ps);
#else
            // XXX is this thread-safe?
            incr = (int) mbtowc((wchar_t *) &tmp, *input, *inputLeft);
#endif
            if (incr < 0) {
                NS_WARNING("mbtowc failed: possible charset mismatch");
                // zero-pad and hope for the best
                tmp = (unsigned char) **input;
                incr = 1;
            }
            **output = (char16_t) tmp;
            (*input) += incr;
            (*inputLeft) -= incr;
            (*output)++;
            (*outputLeft)--;
        }
    }
    else {
        // wchar_t isn't unicode, so the best we can do is treat the
        // input as if it is isolatin1 :(
        isolatin1_to_utf16(input, inputLeft, output, outputLeft);
    }

    return NS_OK;
}

nsresult
nsNativeCharsetConverter::UnicodeToNative(const char16_t **input,
                                          uint32_t         *inputLeft,
                                          char            **output,
                                          uint32_t         *outputLeft)
{
    if (gWCharIsUnicode) {
        int incr;

        while (*inputLeft && *outputLeft >= MB_CUR_MAX) {
#ifdef HAVE_WCRTOMB
            incr = (int) wcrtomb(*output, (wchar_t) **input, &ps);
#else
            // XXX is this thread-safe?
            incr = (int) wctomb(*output, (wchar_t) **input);
#endif
            if (incr < 0) {
                NS_WARNING("mbtowc failed: possible charset mismatch");
                **output = (unsigned char) **input; // truncate
                incr = 1;
            }
            // most likely we're dead anyways if this assertion should fire
            NS_ASSERTION(uint32_t(incr) <= *outputLeft, "wrote beyond end of string");
            (*output) += incr;
            (*outputLeft) -= incr;
            (*input)++;
            (*inputLeft)--;
        }
    }
    else {
        // wchar_t isn't unicode, so the best we can do is treat the
        // input as if it is isolatin1 :(
        utf16_to_isolatin1(input, inputLeft, output, outputLeft);
    }

    return NS_OK;
}

// XXX : for now, return false
bool
nsNativeCharsetConverter::IsNativeUTF8()
{
    return false;
}

#endif // USE_STDCONV

//-----------------------------------------------------------------------------
// API implementation
//-----------------------------------------------------------------------------

nsresult
NS_CopyNativeToUnicode(const nsACString &input, nsAString &output)
{
    output.Truncate();

    uint32_t inputLen = input.Length();

    nsACString::const_iterator iter;
    input.BeginReading(iter);

    //
    // OPTIMIZATION: preallocate space for largest possible result; convert
    // directly into the result buffer to avoid intermediate buffer copy.
    //
    // this will generally result in a larger allocation, but that seems
    // better than an extra buffer copy.
    //
    if (!output.SetLength(inputLen, fallible_t()))
        return NS_ERROR_OUT_OF_MEMORY;
    nsAString::iterator out_iter;
    output.BeginWriting(out_iter);

    char16_t *result = out_iter.get();
    uint32_t resultLeft = inputLen;

    const char *buf = iter.get();
    uint32_t bufLeft = inputLen;

    nsNativeCharsetConverter conv;
    nsresult rv = conv.NativeToUnicode(&buf, &bufLeft, &result, &resultLeft);
    if (NS_SUCCEEDED(rv)) {
        NS_ASSERTION(bufLeft == 0, "did not consume entire input buffer");
        output.SetLength(inputLen - resultLeft);
    }
    return rv;
}

nsresult
NS_CopyUnicodeToNative(const nsAString &input, nsACString &output)
{
    output.Truncate();

    nsAString::const_iterator iter, end;
    input.BeginReading(iter);
    input.EndReading(end);

    // cannot easily avoid intermediate buffer copy.
    char temp[4096];

    nsNativeCharsetConverter conv;

    const char16_t *buf = iter.get();
    uint32_t bufLeft = Distance(iter, end);
    while (bufLeft) {
        char *p = temp;
        uint32_t tempLeft = sizeof(temp);

        nsresult rv = conv.UnicodeToNative(&buf, &bufLeft, &p, &tempLeft);
        if (NS_FAILED(rv)) return rv;

        if (tempLeft < sizeof(temp))
            output.Append(temp, sizeof(temp) - tempLeft);
    }
    return NS_OK;
}

bool
NS_IsNativeUTF8()
{
    return nsNativeCharsetConverter::IsNativeUTF8();
}

void
NS_StartupNativeCharsetUtils()
{
    //
    // need to initialize the locale or else charset conversion will fail.
    // better not delay this in case some other component alters the locale
    // settings.
    //
    // XXX we assume that we are called early enough that we should
    // always be the first to care about the locale's charset.
    //
    setlocale(LC_CTYPE, "");

    nsNativeCharsetConverter::GlobalInit();
}

void
NS_ShutdownNativeCharsetUtils()
{
    nsNativeCharsetConverter::GlobalShutdown();
}

//-----------------------------------------------------------------------------
// XP_WIN
//-----------------------------------------------------------------------------
#elif defined(XP_WIN)

#include <windows.h>
#include "nsString.h"
#include "nsAString.h"
#include "nsReadableUtils.h"

using namespace mozilla;

nsresult
NS_CopyNativeToUnicode(const nsACString &input, nsAString &output)
{
    uint32_t inputLen = input.Length();

    nsACString::const_iterator iter;
    input.BeginReading(iter);

    const char *buf = iter.get();

    // determine length of result
    uint32_t resultLen = 0;
    int n = ::MultiByteToWideChar(CP_ACP, 0, buf, inputLen, nullptr, 0);
    if (n > 0)
        resultLen += n;

    // allocate sufficient space
    if (!output.SetLength(resultLen, fallible_t()))
        return NS_ERROR_OUT_OF_MEMORY;
    if (resultLen > 0) {
        nsAString::iterator out_iter;
        output.BeginWriting(out_iter);

        char16_t *result = out_iter.get();

        ::MultiByteToWideChar(CP_ACP, 0, buf, inputLen, wwc(result), resultLen);
    }
    return NS_OK;
}

nsresult
NS_CopyUnicodeToNative(const nsAString  &input, nsACString &output)
{
    uint32_t inputLen = input.Length();

    nsAString::const_iterator iter;
    input.BeginReading(iter);

    char16ptr_t buf = iter.get();

    // determine length of result
    uint32_t resultLen = 0;

    int n = ::WideCharToMultiByte(CP_ACP, 0, buf, inputLen, nullptr, 0,
                                  nullptr, nullptr);
    if (n > 0)
        resultLen += n;

    // allocate sufficient space
    if (!output.SetLength(resultLen, fallible_t()))
        return NS_ERROR_OUT_OF_MEMORY;
    if (resultLen > 0) {
        nsACString::iterator out_iter;
        output.BeginWriting(out_iter);

        // default "defaultChar" is '?', which is an illegal character on windows
        // file system.  That will cause file uncreatable. Change it to '_'
        const char defaultChar = '_';

        char *result = out_iter.get();

        ::WideCharToMultiByte(CP_ACP, 0, buf, inputLen, result, resultLen,
                              &defaultChar, nullptr);
    }
    return NS_OK;
}

// moved from widget/windows/nsToolkit.cpp
int32_t 
NS_ConvertAtoW(const char *aStrInA, int aBufferSize, char16_t *aStrOutW)
{
    return MultiByteToWideChar(CP_ACP, 0, aStrInA, -1, wwc(aStrOutW), aBufferSize);
}

int32_t 
NS_ConvertWtoA(const char16_t *aStrInW, int aBufferSizeOut,
               char *aStrOutA, const char *aDefault)
{
    if ((!aStrInW) || (!aStrOutA) || (aBufferSizeOut <= 0))
        return 0;

    int numCharsConverted = WideCharToMultiByte(CP_ACP, 0, char16ptr_t(aStrInW), -1,
                                                aStrOutA, aBufferSizeOut,
                                                aDefault, nullptr);

    if (!numCharsConverted) {
        if (GetLastError() == ERROR_INSUFFICIENT_BUFFER) {
            // Overflow, add missing null termination but return 0
            aStrOutA[aBufferSizeOut-1] = '\0';
        }
        else {
            // Other error, clear string and return 0
            aStrOutA[0] = '\0';
        }
    }
    else if (numCharsConverted < aBufferSizeOut) {
        // Add 2nd null (really necessary?)
        aStrOutA[numCharsConverted] = '\0';
    }

    return numCharsConverted;
}

#else

#include "nsReadableUtils.h"

nsresult
NS_CopyNativeToUnicode(const nsACString &input, nsAString  &output)
{
    CopyASCIItoUTF16(input, output);
    return NS_OK;
}

nsresult
NS_CopyUnicodeToNative(const nsAString  &input, nsACString &output)
{
    LossyCopyUTF16toASCII(input, output);
    return NS_OK;
}

void
NS_StartupNativeCharsetUtils()
{
}

void
NS_ShutdownNativeCharsetUtils()
{
}

#endif