Hello,
I attach a patch that adds a configure option "--enable-utf8-mode" that
can be used to enable utf8 characters on systems that this is not
supported by its libc (embedded) systems. It is based on the Xutf8
implementation by Jean-Marc Lienher.
regards,
Nikos
Index: src/ntfs-3g.c
===================================================================
--- a/src/ntfs-3g.c (revision 733)
+++ b/src/ntfs-3g.c (working copy)
@@ -160,8 +160,13 @@
static const char *usage_msg =
"\n"
-"%s %s %s %d - Third Generation NTFS Driver\n"
+"%s %s %s %d - Third Generation NTFS Driver"
+#ifdef NTFS_UTF8
+" (UTF8 mode)\n"
+#else
"\n"
+#endif
+"\n"
"Copyright (C) 2006-2008 Szabolcs Szakacsits\n"
"Copyright (C) 2005-2007 Yura Pakhuchiy\n"
"\n"
@@ -358,7 +363,9 @@
stream_name_len = ntfs_fuse_parse_path(org_path, &path, &stream_name);
if (stream_name_len < 0)
return stream_name_len;
+
memset(stbuf, 0, sizeof(struct stat));
+
ni = ntfs_pathname_to_inode(ctx->vol, NULL, path);
if (!ni) {
res = -errno;
Index: config.h.in
===================================================================
--- a/config.h.in (revision 733)
+++ b/config.h.in (working copy)
@@ -270,6 +270,9 @@
/* Don't use default IO ops */
#undef NO_NTFS_DEVICE_DEFAULT_IO_OPS
+/* Define to 1 if utf8 mode enabled */
+#undef NTFS_UTF8
+
/* Name of package */
#undef PACKAGE
Index: configure.ac
===================================================================
--- a/configure.ac (revision 733)
+++ b/configure.ac (working copy)
@@ -53,6 +53,13 @@
)
AC_ARG_ENABLE(
+ [utf8-mode],
+ [AS_HELP_STRING([--enable-utf8-mode],[enable UTF8 mode of operation (avoid using locale)])],
+ [utf8_mode="yes"],
+ [utf8_mode="no"]
+)
+
+AC_ARG_ENABLE(
[pedantic],
[AS_HELP_STRING([--enable-pedantic],[enable compile pedantic mode])],
,
@@ -302,6 +309,14 @@
)
fi
+if test "${utf8_mode}" = "yes"; then
+ AC_DEFINE(
+ [NTFS_UTF8],
+ [1],
+ [Define to 1 if utf8 mode enabled]
+ )
+fi
+
test "${enable_device_default_io_ops}" = "no" && AC_DEFINE(
[NO_NTFS_DEVICE_DEFAULT_IO_OPS],
[1],
Index: libntfs-3g/Xutf8.c
===================================================================
--- a/libntfs-3g/Xutf8.c (revision 0)
+++ b/libntfs-3g/Xutf8.c (revision 0)
@@ -0,0 +1,186 @@
+/*
+ * "$Id: $"
+ *
+ * Unicode to UTF-8 conversion functions.
+ *
+ * Copyright (c) 2000,2001 by O'ksi'D.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * O'KSI'D BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF
+ * OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ * Except as contained in this notice, the name of the O'ksi'D shall
+ * not be used in advertising or otherwise to promote the sale, use or other
+ * dealings in this Software without prior written authorization from O'ksi'D.
+ *
+ *
+ * Author: Jean-Marc Lienher ( http://oksid.ch )
+ */
+
+#include "Xutf8.h"
+
+/*** NOTE : all functions are LIMITED to 24 bits Unicode values !!! ***/
+
+/*
+ * Converts the first char of the UTF-8 string to an Unicode value
+ * Returns the byte length of the converted UTF-8 char
+ * Returns -1 if the UTF-8 string is not valid
+ */
+int
+XConvertUtf8ToUcs(
+ const unsigned char *buf,
+ int len,
+ unsigned int *ucs)
+{
+ if (buf == 0 || ucs == 0) return 0;
+
+ if (buf[0] & 0x80) {
+ if (buf[0] & 0x40) {
+ if (buf[0] & 0x20) {
+ if (buf[0] & 0x10) {
+ if (buf[0] & 0x08) {
+ if (buf[0] & 0x04) {
+ if (buf[0] & 0x02) {
+ /* bad UTF-8 string */
+ } else {
+ /* 0x04000000 - 0x7FFFFFFF */
+ }
+ } else if (len > 4
+ && (buf[1] & 0xC0) == 0x80
+ && (buf[2] & 0xC0) == 0x80
+ && (buf[3] & 0xC0) == 0x80
+ && (buf[4] & 0xC0) == 0x80)
+ {
+ /* 0x00200000 - 0x03FFFFFF */
+ *ucs = ((buf[0] & ~0xF8) << 24) +
+ ((buf[1] & ~0x80) << 18) +
+ ((buf[2] & ~0x80) << 12) +
+ ((buf[3] & ~0x80) << 6) +
+ (buf[4] & ~0x80);
+ if (*ucs > 0x001FFFFF && *ucs < 0x01000000) return 5;
+ }
+ } else if (len > 3
+ && (buf[1] & 0xC0) == 0x80
+ && (buf[2] & 0xC0) == 0x80
+ && (buf[3] & 0xC0) == 0x80)
+ {
+ /* 0x00010000 - 0x001FFFFF */
+ *ucs = ((buf[0] & ~0xF0) << 18) +
+ ((buf[1] & ~0x80) << 12) +
+ ((buf[2] & ~0x80) << 6) +
+ (buf[3] & ~0x80);
+ if (*ucs > 0x0000FFFF) return 4;
+ }
+ } else if (len > 2 &&
+ (buf[1] & 0xC0) == 0x80 &&
+ (buf[2] & 0xC0) == 0x80)
+ {
+ /* 0x00000800 - 0x0000FFFF */
+ *ucs = ((buf[0] & ~0xE0) << 12) +
+ ((buf[1] & ~0x80) << 6) +
+ (buf[2] & ~0x80);
+ if (*ucs > 0x000007FF) return 3;
+ }
+ } else if (len > 1 && (buf[1] & 0xC0) == 0x80) {
+ /* 0x00000080 - 0x000007FF */
+ *ucs = ((buf[0] & ~0xC0) << 6) +
+ (buf[1] & ~0x80);
+ if (*ucs > 0x0000007F) return 2;
+ }
+ }
+ } else if (len > 0) {
+ /* 0x00000000 - 0x0000007F */
+ *ucs = buf[0];
+ return 1;
+ }
+
+ *ucs = (unsigned int) '?'; /* bad utf-8 string */
+ return -1;
+}
+
+/*
+ * Converts an Unicode value to an UTF-8 string
+ * NOTE : the buffer (buf) must be at least 5 bytes long !!!
+ */
+int
+XConvertUcsToUtf8(
+ unsigned int ucs,
+ char *buf)
+{
+ if (buf == 0) return 0;
+
+ if (ucs < 0x000080) {
+ buf[0] = ucs;
+ return 1;
+ } else if (ucs < 0x000800) {
+ buf[0] = 0xC0 | (ucs >> 6);
+ buf[1] = 0x80 | (ucs & 0x3F);
+ return 2;
+ } else if (ucs < 0x010000) {
+ buf[0] = 0xE0 | (ucs >> 12);
+ buf[1] = 0x80 | ((ucs >> 6) & 0x3F);
+ buf[2] = 0x80 | (ucs & 0x3F);
+ return 3;
+ } else if (ucs < 0x00200000) {
+ buf[0] = 0xF0 | (ucs >> 18);
+ buf[1] = 0x80 | ((ucs >> 12) & 0x3F);
+ buf[2] = 0x80 | ((ucs >> 6) & 0x3F);
+ buf[3] = 0x80 | (ucs & 0x3F);
+ return 4;
+ } else if (ucs < 0x01000000) {
+ buf[0] = 0xF8 | (ucs >> 24);
+ buf[1] = 0x80 | ((ucs >> 18) & 0x3F);
+ buf[2] = 0x80 | ((ucs >> 12) & 0x3F);
+ buf[3] = 0x80 | ((ucs >> 6) & 0x3F);
+ buf[4] = 0x80 | (ucs & 0x3F);
+ return 5;
+ }
+ buf[0] = '?';
+ return -1;
+}
+
+/*
+ * returns the byte length of the first UTF-8 char
+ * (returns -1 if not valid)
+ */
+int
+ XUtf8CharByteLen(
+ const unsigned char *buf,
+ int len)
+{
+ unsigned int ucs;
+ return XConvertUtf8ToUcs(buf, len, &ucs);
+}
+
+/*
+ * returns the quantity of Unicode chars in the UTF-8 string
+ */
+int
+ XCountUtf8Char(
+ const unsigned char *buf,
+ int len)
+{
+ int i = 0;
+ int nbc = 0;
+ while (i < len) {
+ int cl = XUtf8CharByteLen(buf + i, len - i);
+ if (cl < 1) cl = 1;
+ nbc++;
+ i += cl;
+ }
+ return nbc;
+}
Index: libntfs-3g/Xutf8.h
===================================================================
--- a/libntfs-3g/Xutf8.h (revision 0)
+++ b/libntfs-3g/Xutf8.h (revision 0)
@@ -0,0 +1,3 @@
+int XConvertUtf8ToUcs(const unsigned char *buf, int len, unsigned int *ucs);
+int XConvertUcsToUtf8(unsigned int ucs, char *buf);
+int XCountUtf8Char(const unsigned char *buf, int len);
Index: libntfs-3g/unistr.c
===================================================================
--- a/libntfs-3g/unistr.c (revision 733)
+++ b/libntfs-3g/unistr.c (working copy)
@@ -46,6 +46,7 @@
#include "debug.h"
#include "logging.h"
#include "misc.h"
+#include <assert.h>
/*
* IMPORTANT
@@ -373,6 +374,20 @@
err_val, ic, upcase, upcase_len);
}
+#ifdef NTFS_UTF8
+# undef HAVE_MBSINIT
+# define wchar_t unsigned int
+
+#include "Xutf8.h"
+
+#define mbtowc(a,b,c) XConvertUtf8ToUcs( b, c, a)
+#define wctomb(x,y) XConvertUcsToUtf8(y, x)
+#define mbstowcs(a, b, c) XCountUtf8Char(b, strlen(b))
+
+
+
+#endif /* NTFS_UTF8*/
+
/**
* ntfs_ucstombs - convert a little endian Unicode string to a multibyte string
* @ins: input Unicode string buffer
@@ -439,8 +454,9 @@
return -1;
}
tc = ntfs_malloc((mbs_len + 64) & ~63);
- if (!tc)
+ if (!tc) {
goto err_out;
+ }
memcpy(tc, mbs, mbs_len);
mbs_len = (mbs_len + 64) & ~63;
free(mbs);
@@ -454,10 +470,11 @@
#ifdef HAVE_MBSINIT
cnt = wcrtomb(mbs + o, wc, &mbstate);
#else
- cnt = wctomb(mbs + o, wc);
+ cnt = wctomb(mbs + o, wc);
#endif
- if (cnt == -1)
+ if (cnt == -1) {
goto err_out;
+ }
if (cnt <= 0) {
ntfs_log_debug("Eeek. cnt <= 0, cnt = %i\n", cnt);
errno = EINVAL;
@@ -539,7 +556,7 @@
}
#endif
#elif !defined(DJGPP)
- ins_len = mbstowcs(NULL, s, 0);
+ ins_len = mbstowcs(NULL, s, 0);
#else
/* Eeek!!! DJGPP has broken mbstowcs() implementation!!! */
ins_len = strlen(ins);
@@ -580,7 +597,7 @@
#ifdef HAVE_MBSINIT
cnt = mbrtowc(&wc, ins + i, ins_size - i, &mbstate);
#else
- cnt = mbtowc(&wc, ins + i, ins_size - i);
+ cnt = mbtowc(&wc, ins + i, ins_size - i);
#endif
if (!cnt)
break;
@@ -594,6 +611,7 @@
/* Make sure we are not overflowing the NTFS Unicode set. */
if ((unsigned long)wc >= (unsigned long)(1 <<
(8 * sizeof(ntfschar)))) {
+ ntfs_log_debug("Eeek. cnt = %d %x\n", wc, wc);
errno = EILSEQ;
goto err_out;
}
Index: libntfs-3g/Makefile.am
===================================================================
--- a/libntfs-3g/Makefile.am (revision 733)
+++ b/libntfs-3g/Makefile.am (working copy)
@@ -40,7 +40,8 @@
security.c \
unistr.c \
version.c \
- volume.c
+ volume.c \
+ Xutf8.c
if NTFS_DEVICE_DEFAULT_IO_OPS
if WINDOWS
------------------------------------------------------------------------------
_______________________________________________
ntfs-3g-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel
