From 7cdd956e1c55930a020b61806d642b6047deffc7 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Sun, 13 Feb 2000 19:00:53 +0000 Subject: Update. 2000-02-13 Ulrich Drepper * iconvdata/Makefile (modules): Add UTF-16. (distribute): Add utf-16.c. * iconvdata/gconv-modules: Add entries for UTF-16, UTF-16BE, and UTF-16LE. * iconvdata/utf-16.c: New file. * iconv/gconv_builtin.h: Remove UTF-16 entries here. * iconv/gconv_simple.c: Remove conversion functions to and from UTF-16. * iconv/skeleton.c: Increment __invocation_coounter after every call to the loops. --- ChangeLog | 13 ++ iconv/gconv_builtin.h | 14 +- iconv/gconv_simple.c | 170 ------------------------- iconv/skeleton.c | 6 +- iconvdata/Makefile | 5 +- iconvdata/gconv-modules | 12 ++ iconvdata/utf-16.c | 330 ++++++++++++++++++++++++++++++++++++++++++++++++ 7 files changed, 362 insertions(+), 188 deletions(-) create mode 100644 iconvdata/utf-16.c diff --git a/ChangeLog b/ChangeLog index a859dc6..e109ee2 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,16 @@ +2000-02-13 Ulrich Drepper + + * iconvdata/Makefile (modules): Add UTF-16. + (distribute): Add utf-16.c. + * iconvdata/gconv-modules: Add entries for UTF-16, UTF-16BE, and + UTF-16LE. + * iconvdata/utf-16.c: New file. + * iconv/gconv_builtin.h: Remove UTF-16 entries here. + * iconv/gconv_simple.c: Remove conversion functions to and from UTF-16. + + * iconv/skeleton.c: Increment __invocation_coounter after every call + to the loops. + 2000-02-12 Andreas Jaeger * sysdeps/unix/sysv/linux/i386/msgctl.c (__new_msgctl): Move errno diff --git a/iconv/gconv_builtin.h b/iconv/gconv_builtin.h index fec1cf5..6e91a86 100644 --- a/iconv/gconv_builtin.h +++ b/iconv/gconv_builtin.h @@ -1,5 +1,5 @@ /* Builtin transformations. - Copyright (C) 1997, 1998, 1999 Free Software Foundation, Inc. + Copyright (C) 1997, 1998, 1999, 2000 Free Software Foundation, Inc. This file is part of the GNU C Library. Contributed by Ulrich Drepper , 1997. @@ -80,15 +80,3 @@ BUILTIN_TRANSFORMATION (NULL, "INTERNAL", 8, "UNICODELITTLE//", 1, "=INTERNAL->ucs2little", __gconv_transform_internal_ucs2little, NULL, NULL, 4, 4, 2, 2) - -BUILTIN_ALIAS ("UTF-16//", "UTF16//") - -BUILTIN_TRANSFORMATION (NULL, "INTERNAL", 8, "UTF16//", - 1, "=INTERNAL->utf16", - __gconv_transform_internal_utf16, NULL, NULL, - 4, 4, 2, 4) - -BUILTIN_TRANSFORMATION (NULL, "UTF16//", 7, "INTERNAL", - 1, "=utf16->INTERNAL", - __gconv_transform_utf16_internal, NULL, NULL, - 2, 4, 4, 4) diff --git a/iconv/gconv_simple.c b/iconv/gconv_simple.c index 6acdd5d..31e7970 100644 --- a/iconv/gconv_simple.c +++ b/iconv/gconv_simple.c @@ -462,173 +462,3 @@ internal_ucs4_loop (const unsigned char **inptrp, const unsigned char *inend, #endif #include #include - - -/* Convert from the internal (UCS4-like) format to UTF-16. */ -#define DEFINE_INIT 0 -#define DEFINE_FINI 0 -#define MIN_NEEDED_FROM 4 -#define MIN_NEEDED_TO 2 -#define MAX_NEEDED_TO 4 -#define FROM_DIRECTION 1 -#define FROM_LOOP internal_utf16_loop -#define TO_LOOP internal_utf16_loop /* This is not used. */ -#define FUNCTION_NAME __gconv_transform_internal_utf16 - -#define MIN_NEEDED_INPUT MIN_NEEDED_FROM -#define MIN_NEEDED_OUTPUT MIN_NEEDED_TO -#define MAX_NEEDED_OUTPUT MAX_NEEDED_TO -#define LOOPFCT FROM_LOOP -#if __BYTE_ORDER == __LITTLE_ENDIAN -# define BODY \ - { \ - if (*((uint32_t *) inptr) >= 0x10000) \ - { \ - if (*((uint32_t *) inptr) >= 0x110000) \ - { \ - result = __GCONV_ILLEGAL_INPUT; \ - break; \ - } \ - \ - /* Generate a surrogate character. */ \ - if (NEED_LENGTH_TEST && outptr + 4 > outend) \ - { \ - /* Overflow in the output buffer. */ \ - result = __GCONV_FULL_OUTPUT; \ - break; \ - } \ - \ - *((uint16_t *) outptr)++ = bswap_16 (0xd7c0 \ - + (*((uint32_t *) inptr) >> 10));\ - *((uint16_t *) outptr)++ = bswap_16 (0xdc00 \ - + (*((uint32_t *) inptr) \ - & 0x3ff)); \ - } \ - else \ - /* Please note that we use the `uint32_t' from-pointer as an `uint16_t' \ - pointer which works since we are on a little endian machine. */ \ - *((uint16_t *) outptr)++ = bswap_16 (*((uint16_t *) inptr)); \ - inptr += 4; \ - } -#else -# define BODY \ - { \ - if (*((uint32_t *) inptr) >= 0x10000) \ - { \ - if (*((uint32_t *) inptr) >= 0x110000) \ - { \ - result = __GCONV_ILLEGAL_INPUT; \ - break; \ - } \ - \ - /* Generate a surrogate character. */ \ - if (NEED_LENGTH_TEST && outptr + 4 > outend) \ - { \ - /* Overflow in the output buffer. */ \ - result = __GCONV_FULL_OUTPUT; \ - break; \ - } \ - \ - *((uint16_t *) outptr)++ = 0xd7c0 + (*((uint32_t *) inptr) >> 10); \ - *((uint16_t *) outptr)++ = 0xdc00 + (*((uint32_t *) inptr) & 0x3ff); \ - } \ - else \ - *((uint16_t *) outptr)++ = *((uint32_t *) inptr)++; \ - } -#endif -#include -#include - - -/* Convert from UTF-16 to the internal (UCS4-like) format. */ -#define DEFINE_INIT 0 -#define DEFINE_FINI 0 -#define MIN_NEEDED_FROM 2 -#define MAX_NEEDED_FROM 4 -#define MIN_NEEDED_TO 4 -#define FROM_DIRECTION 1 -#define FROM_LOOP utf16_internal_loop -#define TO_LOOP utf16_internal_loop /* This is not used.*/ -#define FUNCTION_NAME __gconv_transform_utf16_internal - -#define MIN_NEEDED_INPUT MIN_NEEDED_FROM -#define MAX_NEEDED_INPUT MAX_NEEDED_FROM -#define MIN_NEEDED_OUTPUT MIN_NEEDED_TO -#define LOOPFCT FROM_LOOP -#if __BYTE_ORDER == __LITTLE_ENDIAN -# define BODY \ - { \ - uint16_t u1 = bswap_16 (*(uint16_t *) inptr); \ - \ - if (u1 < 0xd800 || u1 > 0xdfff) \ - { \ - /* No surrogate. */ \ - *((uint32_t *) outptr)++ = u1; \ - inptr += 2; \ - } \ - else \ - { \ - uint16_t u2; \ - \ - /* It's a surrogate character. At least the first word says \ - it is. */ \ - if (NEED_LENGTH_TEST && inptr + 4 > inend) \ - { \ - /* We don't have enough input for another complete input \ - character. */ \ - result = __GCONV_INCOMPLETE_INPUT; \ - break; \ - } \ - \ - u2 = bswap_16 (((uint16_t *) inptr)[1]); \ - if (u2 < 0xdc00 || u2 >= 0xdfff) \ - { \ - /* This is no valid second word for a surrogate. */ \ - result = __GCONV_ILLEGAL_INPUT; \ - break; \ - } \ - \ - *((uint32_t *) outptr)++ = ((u1 - 0xd7c0) << 10) + (u2 - 0xdc00); \ - inptr += 4; \ - } \ - } -#else -# define BODY \ - { \ - uint16_t u1 = *(uint16_t *) inptr; \ - \ - if (u1 < 0xd800 || u1 > 0xdfff) \ - { \ - /* No surrogate. */ \ - *((uint32_t *) outptr)++ = u1; \ - inptr += 2; \ - } \ - else \ - { \ - uint16_t u2; \ - \ - /* It's a surrogate character. At least the first word says \ - it is. */ \ - if (NEED_LENGTH_TEST && inptr + 4 > inend) \ - { \ - /* We don't have enough input for another complete input \ - character. */ \ - result = __GCONV_INCOMPLETE_INPUT; \ - break; \ - } \ - \ - u2 = ((uint16_t *) inptr)[1]; \ - if (u2 < 0xdc00 || u2 >= 0xdfff) \ - { \ - /* This is no valid second word for a surrogate. */ \ - result = __GCONV_ILLEGAL_INPUT; \ - break; \ - } \ - \ - *((uint32_t *) outptr)++ = ((u1 - 0xd7c0) << 10) + (u2 - 0xdc00); \ - inptr += 4; \ - } \ - } -#endif -#include -#include diff --git a/iconv/skeleton.c b/iconv/skeleton.c index da15758..fd58be8 100644 --- a/iconv/skeleton.c +++ b/iconv/skeleton.c @@ -332,15 +332,15 @@ FUNCTION_NAME (struct __gconv_step *step, struct __gconv_step_data *data, if (status == __GCONV_FULL_OUTPUT) status = __GCONV_OK; } + + /* We finished one use of the loops. */ + ++data->__invocation_counter; } while (status == __GCONV_OK); #ifdef END_LOOP END_LOOP #endif - - /* We finished one use of this step. */ - ++data->__invocation_counter; } return status; diff --git a/iconvdata/Makefile b/iconvdata/Makefile index f4aa5f3..dee418e 100644 --- a/iconvdata/Makefile +++ b/iconvdata/Makefile @@ -45,7 +45,7 @@ modules := ISO8859-1 ISO8859-2 ISO8859-3 ISO8859-4 ISO8859-5 \ INIS-CYRILLIC ISO_6937-2 ISO_2033 ISO_5427 ISO_5427-EXT \ ISO_5428 ISO_10367-BOX MAC-IS MAC-UK NATS-DANO NATS-SEFI \ SAMI-WS2 ISO-IR-197 TIS-620 KOI8-U GBK ISIRI-3342 GBGBK \ - ISO-2022-CN libISOIR165 + ISO-2022-CN libISOIR165 UTF-16 modules.so := $(addsuffix .so, $(modules)) @@ -117,7 +117,8 @@ distribute := gconv-modules extra-module.mk gap.awk gaptab.awk \ koi8-r.c ksc5601.c ksc5601.h latin-greek.c latin-greek-1.c \ macintosh.c mac-is.c mac-uk.c nats-dano.c nats-sefi.c sjis.c \ t.61.c uhc.c sami-ws2.c iso-ir-197.c tis-620.c koi8-u.c \ - isiri-3342.c gbgbk.c iso-2022-cn.c cns11643l2.h iso8859-16.c + isiri-3342.c gbgbk.c iso-2022-cn.c cns11643l2.h iso8859-16.c \ + utf-16.c # We build the transformation modules only when we build shared libs. ifeq (yes,$(build-shared)) diff --git a/iconvdata/gconv-modules b/iconvdata/gconv-modules index 9b4cf10..314578b 100644 --- a/iconvdata/gconv-modules +++ b/iconvdata/gconv-modules @@ -1172,3 +1172,15 @@ module INTERNAL KOI8-U// KOI8-U 1 # from to module cost module ISIRI-3342// INTERNAL ISIRI-3342 1 module INTERNAL ISIRI-3342// ISIRI-3342 1 + +# from to module cost +module UTF-16// INTERNAL UTF-16 1 +module INTERNAL UTF-16// UTF-16 1 + +# from to module cost +module UTF-16LE// INTERNAL UTF-16 1 +module INTERNAL UTF-16LE// UTF-16 1 + +# from to module cost +module UTF-16BE// INTERNAL UTF-16 1 +module INTERNAL UTF-16BE// UTF-16 1 diff --git a/iconvdata/utf-16.c b/iconvdata/utf-16.c new file mode 100644 index 0000000..8348ce2 --- /dev/null +++ b/iconvdata/utf-16.c @@ -0,0 +1,330 @@ +/* Conversion module for UTF-16. + Copyright (C) 1999, 2000 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Ulrich Drepper , 1999. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Library General Public License as + published by the Free Software Foundation; either version 2 of the + License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Library General Public License for more details. + + You should have received a copy of the GNU Library General Public + License along with the GNU C Library; see the file COPYING.LIB. If not, + write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, + Boston, MA 02111-1307, USA. */ + +#include +#include +#include +#include +#include +#include + +/* This is the Byte Order Mark character (BOM). */ +#define BOM 0xfeff + + +/* Definitions used in the body of the `gconv' function. */ +#define FROM_LOOP from_utf16_loop +#define TO_LOOP to_utf16_loop +#define DEFINE_INIT 0 +#define DEFINE_FINI 0 +#define MIN_NEEDED_FROM 2 +#define MAX_NEEDED_FROM 4 +#define MIN_NEEDED_TO 4 +#define FROM_DIRECTION (dir == from_utf16) +#define PREPARE_LOOP \ + enum direction dir = ((struct utf16_data *) step->__data)->dir; \ + enum variant var = ((struct utf16_data *) step->__data)->var; \ + if (!FROM_DIRECTION && var == UTF_16 && !data->__internal_use \ + && data->__invocation_counter == 0) \ + { \ + /* Emit the Byte Order Mark. */ \ + if (outbuf + 2 > outend) \ + return __GCONV_FULL_OUTPUT; \ + \ + *(uint16_t *) outbuf = BOM; \ + outbuf += 2; \ + } +#define EXTRA_LOOP_ARGS , var, data + + +/* Direction of the transformation. */ +enum direction +{ + illegal_dir, + to_utf16, + from_utf16 +}; + +enum variant +{ + illegal_var, + UTF_16, + UTF_16LE, + UTF_16BE +}; + +struct utf16_data +{ + enum direction dir; + enum variant var; +}; + + +int +gconv_init (struct __gconv_step *step) +{ + /* Determine which direction. */ + struct utf16_data *new_data; + enum direction dir = illegal_dir; + enum variant var = illegal_var; + int result; + + if (__strcasecmp (step->__from_name, "UTF-16") == 0) + { + dir = from_utf16; + var = UTF_16; + } + else if (__strcasecmp (step->__to_name, "UTF-16") == 0) + { + dir = to_utf16; + var = UTF_16; + } + else if (__strcasecmp (step->__from_name, "UTF-16BE") == 0) + { + dir = from_utf16; + var = UTF_16BE; + } + else if (__strcasecmp (step->__to_name, "UTF-16BE") == 0) + { + dir = to_utf16; + var = UTF_16BE; + } + else if (__strcasecmp (step->__from_name, "UTF-16LE") == 0) + { + dir = from_utf16; + var = UTF_16LE; + } + else if (__strcasecmp (step->__to_name, "UTF-16LE") == 0) + { + dir = to_utf16; + var = UTF_16LE; + } + + result = __GCONV_NOCONV; + if (dir != illegal_dir) + { + new_data = (struct utf16_data *) malloc (sizeof (struct utf16_data)); + + result = __GCONV_NOMEM; + if (new_data != NULL) + { + new_data->dir = dir; + new_data->var = var; + step->__data = new_data; + + if (var == from_utf16) + { + step->__min_needed_from = MIN_NEEDED_FROM; + step->__max_needed_from = MIN_NEEDED_FROM; + step->__min_needed_to = MIN_NEEDED_TO; + step->__max_needed_to = MIN_NEEDED_TO; + } + else + { + step->__min_needed_from = MIN_NEEDED_TO; + step->__max_needed_from = MIN_NEEDED_TO; + step->__min_needed_to = MIN_NEEDED_FROM; + step->__max_needed_to = MIN_NEEDED_FROM; + } + + step->__stateful = 0; + + result = __GCONV_OK; + } + } + + return result; +} + + +void +gconv_end (struct __gconv_step *data) +{ + free (data->__data); +} + + +/* Convert from the internal (UCS4-like) format to UTF-16. */ +#define MIN_NEEDED_INPUT MIN_NEEDED_TO +#define MIN_NEEDED_OUTPUT MIN_NEEDED_FROM +#define MAX_NEEDED_OUTPUT MAX_NEEDED_FROM +#define LOOPFCT TO_LOOP +#define BODY \ + { \ + uint32_t c = *((uint32_t *) inptr); \ + \ + if ((__BYTE_ORDER == __LITTLE_ENDIAN && var == UTF_16BE) \ + || (__BYTE_ORDER == __BIG_ENDIAN && var == UTF_16LE)) \ + { \ + if (c >= 0x10000) \ + { \ + if (c >= 0x110000) \ + { \ + result = __GCONV_ILLEGAL_INPUT; \ + break; \ + } \ + \ + /* Generate a surrogate character. */ \ + if (NEED_LENGTH_TEST && outptr + 4 > outend) \ + { \ + /* Overflow in the output buffer. */ \ + result = __GCONV_FULL_OUTPUT; \ + break; \ + } \ + \ + *((uint16_t *) outptr) = bswap_16 (0xd7c0 + (c >> 10)); \ + outptr += 2; \ + *((uint16_t *) outptr) = bswap_16 (0xdc00 + (c & 0x3ff)); \ + } \ + else \ + *((uint16_t *) outptr) = bswap_16 (c); \ + } \ + else \ + { \ + if (c >= 0x10000) \ + { \ + if (c >= 0x110000) \ + { \ + result = __GCONV_ILLEGAL_INPUT; \ + break; \ + } \ + \ + /* Generate a surrogate character. */ \ + if (NEED_LENGTH_TEST && outptr + 4 > outend) \ + { \ + /* Overflow in the output buffer. */ \ + result = __GCONV_FULL_OUTPUT; \ + break; \ + } \ + \ + *((uint16_t *) outptr) = 0xd7c0 + (c >> 10); \ + outptr += 2; \ + *((uint16_t *) outptr) = 0xdc00 + (c & 0x3ff); \ + } \ + else \ + *((uint16_t *) outptr) = c; \ + } \ + outptr += 2; \ + inptr += 4; \ + } +#define EXTRA_LOOP_DECLS \ + , enum variant var, struct __gconv_step_data *step_data +#include + + +/* Convert from UTF-16 to the internal (UCS4-like) format. */ +#define MIN_NEEDED_INPUT MIN_NEEDED_FROM +#define MAX_NEEDED_INPUT MAX_NEEDED_FROM +#define MIN_NEEDED_OUTPUT MIN_NEEDED_TO +#define LOOPFCT FROM_LOOP +#define BODY \ + { \ + uint16_t u1 = *(uint16_t *) inptr; \ + \ + if ((__BYTE_ORDER == __LITTLE_ENDIAN && var == UTF_16BE) \ + || (__BYTE_ORDER == __BIG_ENDIAN && var == UTF_16LE)) \ + { \ + u1 = bswap_16 (u1); \ + \ + if (u1 < 0xd800 || u1 > 0xdfff) \ + { \ + /* No surrogate. */ \ + *((uint32_t *) outptr) = u1; \ + inptr += 2; \ + } \ + else \ + { \ + uint16_t u2; \ + \ + /* It's a surrogate character. At least the first word says \ + it is. */ \ + if (NEED_LENGTH_TEST && inptr + 4 > inend) \ + { \ + /* We don't have enough input for another complete input \ + character. */ \ + result = __GCONV_INCOMPLETE_INPUT; \ + break; \ + } \ + \ + u2 = bswap_16 (((uint16_t *) inptr)[1]); \ + if (u2 < 0xdc00 || u2 >= 0xdfff) \ + { \ + /* This is no valid second word for a surrogate. */ \ + result = __GCONV_ILLEGAL_INPUT; \ + break; \ + } \ + \ + *((uint32_t *) outptr) = ((u1 - 0xd7c0) << 10) + (u2 - 0xdc00); \ + inptr += 4; \ + } \ + } \ + else \ + { \ + if (u1 == BOM && var == UTF_16 && !step_data->__internal_use \ + && step_data->__invocation_counter == 0 && inptr == *inptrp) \ + { \ + /* This is the first word in the file and it is the BOM and \ + we are converting a file without specified byte order. \ + Simply sack the BOM. */ \ + inptr += 2; \ + continue; \ + } \ + \ + if (u1 < 0xd800 || u1 > 0xdfff) \ + { \ + /* No surrogate. */ \ + *((uint32_t *) outptr) = u1; \ + inptr += 2; \ + } \ + else \ + { \ + uint16_t u2; \ + \ + /* It's a surrogate character. At least the first word says \ + it is. */ \ + if (NEED_LENGTH_TEST && inptr + 4 > inend) \ + { \ + /* We don't have enough input for another complete input \ + character. */ \ + result = __GCONV_INCOMPLETE_INPUT; \ + break; \ + } \ + \ + u2 = ((uint16_t *) inptr)[1]; \ + if (u2 < 0xdc00 || u2 >= 0xdfff) \ + { \ + /* This is no valid second word for a surrogate. */ \ + result = __GCONV_ILLEGAL_INPUT; \ + break; \ + } \ + \ + *((uint32_t *) outptr) = ((u1 - 0xd7c0) << 10) + (u2 - 0xdc00); \ + inptr += 4; \ + } \ + } \ + outptr += 4; \ + } +#define EXTRA_LOOP_DECLS \ + , enum variant var, struct __gconv_step_data *step_data +#include + + +/* Now define the toplevel functions. */ +#include -- cgit v1.1