UTF-32 conversion module
Bruno Haible
haible@ilog.fr
Thu Apr 12 05:46:00 GMT 2001
Unicode-3.1 contains (as normative part) in Standard Annex 19 a definition
of three encodings UTF-32, UTF-32LE, UTF-32BE, that behave like UTF-16[LE/BE]
except there are no surrogates.
Here is a conversion module for this encoding (derived from utf-16.c).
2001-04-11 Bruno Haible <haible@clisp.cons.org>
* iconvdata/utf-32.c: New file.
* iconvdata/gconv-modules: Add entries for UTF-32, UTF-32LE, UTF-32BE.
* iconvdata/Makefile (modules): Add UTF-32.
(distribute): Add utf-32.c.
*** glibc-20010315/iconvdata/utf-32.c.bak Wed Apr 11 12:12:44 2001
--- glibc-20010315/iconvdata/utf-32.c Thu Apr 12 02:26:18 2001
***************
*** 0 ****
--- 1,270 ----
+ /* Conversion module for UTF-32.
+ Copyright (C) 1999, 2000, 2001 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ 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 <byteswap.h>
+ #include <dlfcn.h>
+ #include <gconv.h>
+ #include <stddef.h>
+ #include <stdint.h>
+ #include <stdlib.h>
+ #include <string.h>
+
+ /* This is the Byte Order Mark character (BOM). */
+ #define BOM 0x0000feffu
+ /* And in the other byte order. */
+ #define BOM_OE 0xfffe0000u
+
+
+ /* Definitions used in the body of the `gconv' function. */
+ #define FROM_LOOP from_utf32_loop
+ #define TO_LOOP to_utf32_loop
+ #define DEFINE_INIT 0
+ #define DEFINE_FINI 0
+ #define MIN_NEEDED_FROM 4
+ #define MIN_NEEDED_TO 4
+ #define FROM_DIRECTION (dir == from_utf32)
+ #define PREPARE_LOOP \
+ enum direction dir = ((struct utf32_data *) step->__data)->dir; \
+ enum variant var = ((struct utf32_data *) step->__data)->var; \
+ int swap; \
+ if (FROM_DIRECTION && var == UTF_32) \
+ { \
+ if (data->__invocation_counter == 0) \
+ { \
+ /* We have to find out which byte order the file is encoded in. */ \
+ if (inptr + 4 > inend) \
+ return __GCONV_EMPTY_INPUT; \
+ \
+ if (get32u (inptr) == BOM) \
+ /* Simply ignore the BOM character. */ \
+ *inptrp = inptr += 4; \
+ else if (get32u (inptr) == BOM_OE) \
+ { \
+ ((struct utf32_data *) step->__data)->swap = 1; \
+ *inptrp = inptr += 4; \
+ } \
+ } \
+ } \
+ else if (!FROM_DIRECTION && var == UTF_32 && !data->__internal_use \
+ && data->__invocation_counter == 0) \
+ { \
+ /* Emit the Byte Order Mark. */ \
+ if (__builtin_expect (outbuf + 4 > outend, 0)) \
+ return __GCONV_FULL_OUTPUT; \
+ \
+ put32u (outbuf, BOM); \
+ outbuf += 4; \
+ } \
+ swap = ((struct utf32_data *) step->__data)->swap;
+ #define EXTRA_LOOP_ARGS , var, swap
+
+
+ /* Direction of the transformation. */
+ enum direction
+ {
+ illegal_dir,
+ to_utf32,
+ from_utf32
+ };
+
+ enum variant
+ {
+ illegal_var,
+ UTF_32,
+ UTF_32LE,
+ UTF_32BE
+ };
+
+ struct utf32_data
+ {
+ enum direction dir;
+ enum variant var;
+ int swap;
+ };
+
+
+ extern int gconv_init (struct __gconv_step *step);
+ int
+ gconv_init (struct __gconv_step *step)
+ {
+ /* Determine which direction. */
+ struct utf32_data *new_data;
+ enum direction dir = illegal_dir;
+ enum variant var = illegal_var;
+ int result;
+
+ if (__strcasecmp (step->__from_name, "UTF-32//") == 0)
+ {
+ dir = from_utf32;
+ var = UTF_32;
+ }
+ else if (__strcasecmp (step->__to_name, "UTF-32//") == 0)
+ {
+ dir = to_utf32;
+ var = UTF_32;
+ }
+ else if (__strcasecmp (step->__from_name, "UTF-32BE//") == 0)
+ {
+ dir = from_utf32;
+ var = UTF_32BE;
+ }
+ else if (__strcasecmp (step->__to_name, "UTF-32BE//") == 0)
+ {
+ dir = to_utf32;
+ var = UTF_32BE;
+ }
+ else if (__strcasecmp (step->__from_name, "UTF-32LE//") == 0)
+ {
+ dir = from_utf32;
+ var = UTF_32LE;
+ }
+ else if (__strcasecmp (step->__to_name, "UTF-32LE//") == 0)
+ {
+ dir = to_utf32;
+ var = UTF_32LE;
+ }
+
+ result = __GCONV_NOCONV;
+ if (__builtin_expect (dir, to_utf32) != illegal_dir)
+ {
+ new_data = (struct utf32_data *) malloc (sizeof (struct utf32_data));
+
+ result = __GCONV_NOMEM;
+ if (new_data != NULL)
+ {
+ new_data->dir = dir;
+ new_data->var = var;
+ new_data->swap = ((var == UTF_32LE && BYTE_ORDER == BIG_ENDIAN)
+ || (var == UTF_32BE
+ && BYTE_ORDER == LITTLE_ENDIAN));
+ step->__data = new_data;
+
+ if (dir == from_utf32)
+ {
+ 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;
+ }
+
+
+ extern void gconv_end (struct __gconv_step *data);
+ void
+ gconv_end (struct __gconv_step *data)
+ {
+ free (data->__data);
+ }
+
+
+ /* Convert from the internal (UCS4-like) format to UTF-32. */
+ #define MIN_NEEDED_INPUT MIN_NEEDED_TO
+ #define MIN_NEEDED_OUTPUT MIN_NEEDED_FROM
+ #define LOOPFCT TO_LOOP
+ #define BODY \
+ { \
+ uint32_t c = get32 (inptr); \
+ \
+ if (__builtin_expect (c >= 0x110000, 0)) \
+ { \
+ STANDARD_ERR_HANDLER (4); \
+ } \
+ else if (__builtin_expect (c >= 0xd800 && c < 0xe000, 0)) \
+ { \
+ /* Surrogate characters in UCS-4 input are not valid. \
+ We must catch this. If we let surrogates pass through, \
+ attackers could make a security hole exploit by \
+ generating "irregular UTF-32" sequences. */ \
+ if (! ignore_errors_p ()) \
+ { \
+ result = __GCONV_ILLEGAL_INPUT; \
+ break; \
+ } \
+ inptr += 4; \
+ ++*irreversible; \
+ continue; \
+ } \
+ \
+ if (swap) \
+ put32 (outptr, bswap_32 (c)); \
+ else \
+ put32 (outptr, c); \
+ \
+ outptr += 4; \
+ inptr += 4; \
+ }
+ #define LOOP_NEED_FLAGS
+ #define EXTRA_LOOP_DECLS \
+ , enum variant var, int swap
+ #include <iconv/loop.c>
+
+
+ /* Convert from UTF-32 to the internal (UCS4-like) format. */
+ #define MIN_NEEDED_INPUT MIN_NEEDED_FROM
+ #define MIN_NEEDED_OUTPUT MIN_NEEDED_TO
+ #define LOOPFCT FROM_LOOP
+ #define BODY \
+ { \
+ uint32_t u1 = get32 (inptr); \
+ \
+ if (swap) \
+ u1 = bswap_32 (u1); \
+ \
+ if (__builtin_expect (u1 >= 0x110000, 0)) \
+ { \
+ /* This is illegal. */ \
+ if (! ignore_errors_p ()) \
+ { \
+ /* This is an illegal character. */ \
+ result = __GCONV_ILLEGAL_INPUT; \
+ break; \
+ } \
+ \
+ inptr += 4; \
+ ++*irreversible; \
+ continue; \
+ } \
+ \
+ put32 (outptr, u1); \
+ inptr += 4; \
+ outptr += 4; \
+ }
+ #define LOOP_NEED_FLAGS
+ #define EXTRA_LOOP_DECLS \
+ , enum variant var, int swap
+ #include <iconv/loop.c>
+
+
+ /* Now define the toplevel functions. */
+ #include <iconv/skeleton.c>
*** glibc-20010315/iconvdata/gconv-modules.bak Wed Mar 14 00:02:38 2001
--- glibc-20010315/iconvdata/gconv-modules Wed Apr 11 12:15:50 2001
***************
*** 1319,1324 ****
--- 1319,1336 ----
module INTERNAL UNICODE// UNICODE 1
# from to module cost
+ module UTF-32// INTERNAL UTF-32 1
+ module INTERNAL UTF-32// UTF-32 1
+
+ # from to module cost
+ module UTF-32LE// INTERNAL UTF-32 1
+ module INTERNAL UTF-32LE// UTF-32 1
+
+ # from to module cost
+ module UTF-32BE// INTERNAL UTF-32 1
+ module INTERNAL UTF-32BE// UTF-32 1
+
+ # from to module cost
module UTF-7// INTERNAL UTF-7 1
module INTERNAL UTF-7// UTF-7 1
*** glibc-20010315/iconvdata/Makefile.bak Tue Feb 27 16:54:34 2001
--- glibc-20010315/iconvdata/Makefile Wed Apr 11 12:20:26 2001
***************
*** 46,52 ****
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 UTF-16 UNICODE UTF-7 BIG5HKSCS \
GB18030 ISO-2022-CN-EXT VISCII GBBIG5
modules.so := $(addsuffix .so, $(modules))
--- 46,52 ----
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 UTF-16 UNICODE UTF-32 UTF-7 BIG5HKSCS \
GB18030 ISO-2022-CN-EXT VISCII GBBIG5
modules.so := $(addsuffix .so, $(modules))
***************
*** 134,140 ****
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 isiri-3342.h gbgbk.c iso-2022-cn.c cns11643l2.h \
! iso8859-16.c utf-16.c unicode.c utf-7.c big5hkscs.c \
iso-ir-165.c iso-ir-165.h gb18030.c iso-2022-cn-ext.c \
ibm932.c ibm932.h ibm943.c ibm943.h gbbig5.c
--- 134,140 ----
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 isiri-3342.h gbgbk.c iso-2022-cn.c cns11643l2.h \
! iso8859-16.c utf-16.c unicode.c utf-32.c utf-7.c big5hkscs.c \
iso-ir-165.c iso-ir-165.h gb18030.c iso-2022-cn-ext.c \
ibm932.c ibm932.h ibm943.c ibm943.h gbbig5.c
More information about the Libc-alpha
mailing list