bash-4.2 remove leftover and stray files

This commit is contained in:
Chet Ramey
2011-12-12 22:16:30 -05:00
parent e3af937042
commit 9519c7f69c
192 changed files with 0 additions and 230379 deletions
-162
View File
@@ -1,162 +0,0 @@
## -*- text -*- ####################################################
# #
# Makefile for the GNU Glob Library. #
# #
####################################################################
#
# Copyright (C) 1996-2009 Free Software Foundation, Inc.
# 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 3 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, see <http://www.gnu.org/licenses/>.
srcdir = @srcdir@
VPATH = .:@srcdir@
topdir = @top_srcdir@
BUILD_DIR = @BUILD_DIR@
INSTALL = @INSTALL@
INSTALL_PROGRAM = @INSTALL_PROGRAM@
INSTALL_DATA = @INSTALL_DATA@
CC = @CC@
RANLIB = @RANLIB@
AR = @AR@
ARFLAGS = @ARFLAGS@
RM = rm -f
CP = cp
MV = mv
SHELL = @MAKE_SHELL@
PROFILE_FLAGS = @PROFILE_FLAGS@
CFLAGS = @CFLAGS@
LOCAL_CFLAGS = @LOCAL_CFLAGS@
CPPFLAGS = @CPPFLAGS@
LDFLAGS = @LDFLAGS@ @LOCAL_LDFLAGS@
DEFS = @DEFS@
LOCAL_DEFS = @LOCAL_DEFS@
BASHINCDIR = ${topdir}/include
INCLUDES = -I. -I../.. -I$(topdir) -I$(BASHINCDIR) -I$(topdir)/lib
CCFLAGS = $(PROFILE_FLAGS) $(DEFS) $(LOCAL_DEFS) $(CPPFLAGS) ${INCLUDES} \
$(LOCAL_CFLAGS) $(CFLAGS)
# Here is a rule for making .o files from .c files that doesn't force
# the type of the machine (like -sun3) into the flags.
.c.o:
$(RM) $@
$(CC) -c $(CCFLAGS) $<
# The name of the library target.
LIBRARY_NAME = libglob.a
# The C code source files for this library.
CSOURCES = $(srcdir)/glob.c $(srcdir)/strmatch.c $(srcdir)/smatch.c \
$(srcdir)/xmbsrtowcs.c
# The header files for this library.
HSOURCES = $(srcdir)/strmatch.h
OBJECTS = glob.o strmatch.o smatch.o xmbsrtowcs.o
# The texinfo files which document this library.
DOCSOURCE = doc/glob.texi
DOCOBJECT = doc/glob.dvi
DOCSUPPORT = doc/Makefile
DOCUMENTATION = $(DOCSOURCE) $(DOCOBJECT) $(DOCSUPPORT)
SUPPORT = Makefile ChangeLog $(DOCSUPPORT)
SOURCES = $(CSOURCES) $(HSOURCES) $(DOCSOURCE)
THINGS_TO_TAR = $(SOURCES) $(SUPPORT)
######################################################################
all: $(LIBRARY_NAME)
$(LIBRARY_NAME): $(OBJECTS)
$(RM) -f $@
$(AR) $(ARFLAGS) $@ $(OBJECTS)
-test -n "$(RANLIB)" && $(RANLIB) $@
what-tar:
@for file in $(THINGS_TO_TAR); do \
echo $(selfdir)$$file; \
done
documentation: force
-(cd doc; $(MAKE) $(MFLAGS))
force:
# The rule for 'includes' is written funny so that the if statement
# always returns TRUE unless there really was an error installing the
# include files.
install:
clean:
rm -f $(OBJECTS) $(LIBRARY_NAME)
-(cd doc && $(MAKE) $(MFLAGS) $@ )
realclean distclean maintainer-clean: clean
-( cd doc && $(MAKE) $(MFLAGS) $@ )
$(RM) -f Makefile
mostlyclean: clean
-( cd doc && $(MAKE) $(MFLAGS) $@ )
${BUILD_DIR}/pathnames.h: ${BUILD_DIR}/config.h ${BUILD_DIR}/Makefile Makefile
-( cd ${BUILD_DIR} && ${MAKE} ${MFLAGS} pathnames.h )
######################################################################
# #
# Dependencies for the object files which make up this library. #
# #
######################################################################
smatch.o: strmatch.h
smatch.o: $(BUILD_DIR)/config.h
smatch.o: $(BASHINCDIR)/chartypes.h
smatch.o: $(BASHINCDIR)/ansi_stdlib.h $(topdir)/bashansi.h
smatch.o: $(BASHINCDIR)/shmbutil.h
smatch.o: $(topdir)/xmalloc.h
strmatch.o: strmatch.h
strmatch.o: $(BUILD_DIR)/config.h
strmatch.o: $(BASHINCDIR)/stdc.h
glob.o: $(BUILD_DIR)/config.h
glob.o: $(topdir)/shell.h $(BUILD_DIR)/pathnames.h
glob.o: $(topdir)/bashtypes.h $(BASHINCDIR)/ansi_stdlib.h $(topdir)/bashansi.h
glob.o: $(BASHINCDIR)/posixstat.h $(BASHINCDIR)/memalloc.h
glob.o: strmatch.h glob.h
glob.o: $(BASHINCDIR)/shmbutil.h
glob.o: $(topdir)/xmalloc.h
xmbsrtowcs.o: ${BUILD_DIR}/config.h
xmbsrtowcs.o: ${topdir}/bashansi.h ${BASHINCDIR}/ansi_stdlib.h
xmbsrtowcs.o: ${BASHINCDIR}/shmbutil.h
# Rules for deficient makes, like SunOS and Solaris
glob.o: glob.c
strmatch.o: strmatch.c
smatch.o: smatch.c
xmbsrtowcs.o: xmbsrtowcs.c
# dependencies for C files that include other C files
glob.o: glob_loop.c
smatch.o: sm_loop.c
-314
View File
@@ -1,314 +0,0 @@
/* gmisc.c -- miscellaneous pattern matching utility functions for Bash.
Copyright (C) 2010 Free Software Foundation, Inc.
This file is part of GNU Bash, the Bourne-Again SHell.
Bash 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 3 of the License, or
(at your option) any later version.
Bash 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 Bash. If not, see <http://www.gnu.org/licenses/>.
*/
#include <config.h>
#include "bashtypes.h"
#if defined (HAVE_UNISTD_H)
# include <unistd.h>
#endif
#include "bashansi.h"
#include "shmbutil.h"
#include "stdc.h"
#ifndef LPAREN
# define LPAREN '('
#endif
#ifndef RPAREN
# define RPAREN ')'
#endif
#if defined (HANDLE_MULTIBYTE)
#define WLPAREN L'('
#define WRPAREN L')'
/* Return 1 of the first character of WSTRING could match the first
character of pattern WPAT. Wide character version. */
int
match_pattern_wchar (wpat, wstring)
wchar_t *wpat, *wstring;
{
wchar_t wc;
if (*wstring == 0)
return (0);
switch (wc = *wpat++)
{
default:
return (*wstring == wc);
case L'\\':
return (*wstring == *wpat);
case L'?':
return (*wpat == WLPAREN ? 1 : (*wstring != L'\0'));
case L'*':
return (1);
case L'+':
case L'!':
case L'@':
return (*wpat == WLPAREN ? 1 : (*wstring == wc));
case L'[':
return (*wstring != L'\0');
}
}
int
wmatchlen (wpat, wmax)
wchar_t *wpat;
size_t wmax;
{
wchar_t wc, *wbrack;
int matlen, t, in_cclass, in_collsym, in_equiv;
if (*wpat == 0)
return (0);
matlen = 0;
while (wc = *wpat++)
{
switch (wc)
{
default:
matlen++;
break;
case L'\\':
if (*wpat == 0)
return ++matlen;
else
{
matlen++;
wpat++;
}
break;
case L'?':
if (*wpat == WLPAREN)
return (matlen = -1); /* XXX for now */
else
matlen++;
break;
case L'*':
return (matlen = -1);
case L'+':
case L'!':
case L'@':
if (*wpat == WLPAREN)
return (matlen = -1); /* XXX for now */
else
matlen++;
break;
case L'[':
/* scan for ending `]', skipping over embedded [:...:] */
wbrack = wpat;
wc = *wpat++;
do
{
if (wc == 0)
{
matlen += wpat - wbrack - 1; /* incremented below */
break;
}
else if (wc == L'\\')
{
wc = *wpat++;
if (*wpat == 0)
break;
}
else if (wc == L'[' && *wpat == L':') /* character class */
{
wpat++;
in_cclass = 1;
}
else if (in_cclass && wc == L':' && *wpat == L']')
{
wpat++;
in_cclass = 0;
}
else if (wc == L'[' && *wpat == L'.') /* collating symbol */
{
wpat++;
if (*wpat == L']') /* right bracket can appear as collating symbol */
wpat++;
in_collsym = 1;
}
else if (in_collsym && wc == L'.' && *wpat == L']')
{
wpat++;
in_collsym = 0;
}
else if (wc == L'[' && *wpat == L'=') /* equivalence class */
{
wpat++;
if (*wpat == L']') /* right bracket can appear as equivalence class */
wpat++;
in_equiv = 1;
}
else if (in_equiv && wc == L'=' && *wpat == L']')
{
wpat++;
in_equiv = 0;
}
}
while ((wc = *wpat++) != L']');
matlen++; /* bracket expression can only match one char */
break;
}
}
return matlen;
}
#endif
/* Return 1 of the first character of STRING could match the first
character of pattern PAT. Used to avoid n2 calls to strmatch(). */
int
match_pattern_char (pat, string)
char *pat, *string;
{
char c;
if (*string == 0)
return (0);
switch (c = *pat++)
{
default:
return (*string == c);
case '\\':
return (*string == *pat);
case '?':
return (*pat == LPAREN ? 1 : (*string != '\0'));
case '*':
return (1);
case '+':
case '!':
case '@':
return (*pat == LPAREN ? 1 : (*string == c));
case '[':
return (*string != '\0');
}
}
int
umatchlen (pat, max)
char *pat;
size_t max;
{
char c, *brack;
int matlen, t, in_cclass, in_collsym, in_equiv;
if (*pat == 0)
return (0);
matlen = 0;
while (c = *pat++)
{
switch (c)
{
default:
matlen++;
break;
case '\\':
if (*pat == 0)
return ++matlen;
else
{
matlen++;
pat++;
}
break;
case '?':
if (*pat == LPAREN)
return (matlen = -1); /* XXX for now */
else
matlen++;
break;
case '*':
return (matlen = -1);
case '+':
case '!':
case '@':
if (*pat == LPAREN)
return (matlen = -1); /* XXX for now */
else
matlen++;
break;
case '[':
/* scan for ending `]', skipping over embedded [:...:] */
brack = pat;
c = *pat++;
do
{
if (c == 0)
{
matlen += pat - brack - 1; /* incremented below */
break;
}
else if (c == '\\')
{
c = *pat++;
if (*pat == 0)
break;
}
else if (c == '[' && *pat == ':') /* character class */
{
pat++;
in_cclass = 1;
}
else if (in_cclass && c == ':' && *pat == ']')
{
pat++;
in_cclass = 0;
}
else if (c == '[' && *pat == '.') /* collating symbol */
{
pat++;
if (*pat == ']') /* right bracket can appear as collating symbol */
pat++;
in_collsym = 1;
}
else if (in_collsym && c == '.' && *pat == ']')
{
pat++;
in_collsym = 0;
}
else if (c == '[' && *pat == '=') /* equivalence class */
{
pat++;
if (*pat == ']') /* right bracket can appear as equivalence class */
pat++;
in_equiv = 1;
}
else if (in_equiv && c == '=' && *pat == ']')
{
pat++;
in_equiv = 0;
}
}
while ((c = *pat++) != ']');
matlen++; /* bracket expression can only match one char */
break;
}
}
return matlen;
}
-406
View File
@@ -1,406 +0,0 @@
/* strmatch.c -- ksh-like extended pattern matching for the shell and filename
globbing. */
/* Copyright (C) 1991-2011 Free Software Foundation, Inc.
This file is part of GNU Bash, the Bourne Again SHell.
Bash 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 3 of the License, or
(at your option) any later version.
Bash 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 Bash. If not, see <http://www.gnu.org/licenses/>.
*/
#include <config.h>
#include <stdio.h> /* for debugging */
#include "strmatch.h"
#include <chartypes.h>
#include "bashansi.h"
#include "shmbutil.h"
#include "xmalloc.h"
#include "externs.h"
/* First, compile `sm_loop.c' for single-byte characters. */
#define CHAR unsigned char
#define U_CHAR unsigned char
#define XCHAR char
#define INT int
#define L(CS) CS
#define INVALID -1
#undef STREQ
#undef STREQN
#define STREQ(a, b) ((a)[0] == (b)[0] && strcmp(a, b) == 0)
#define STREQN(a, b, n) ((a)[0] == (b)[0] && strncmp(a, b, n) == 0)
/* We use strcoll(3) for range comparisons in bracket expressions,
even though it can have unwanted side effects in locales
other than POSIX or US. For instance, in the de locale, [A-Z] matches
all characters. */
#if defined (HAVE_STRCOLL)
/* Helper function for collating symbol equivalence. */
static int rangecmp (c1, c2)
int c1, c2;
{
static char s1[2] = { ' ', '\0' };
static char s2[2] = { ' ', '\0' };
int ret;
/* Eight bits only. Period. */
c1 &= 0xFF;
c2 &= 0xFF;
if (c1 == c2)
return (0);
s1[0] = c1;
s2[0] = c2;
if ((ret = strcoll (s1, s2)) != 0)
return ret;
return (c1 - c2);
}
#else /* !HAVE_STRCOLL */
# define rangecmp(c1, c2) ((int)(c1) - (int)(c2))
#endif /* !HAVE_STRCOLL */
#if defined (HAVE_STRCOLL)
static int
collequiv (c1, c2)
int c1, c2;
{
return (rangecmp (c1, c2) == 0);
}
#else
# define collequiv(c1, c2) ((c1) == (c2))
#endif
#define _COLLSYM _collsym
#define __COLLSYM __collsym
#define POSIXCOLL posix_collsyms
#include "collsyms.h"
static int
collsym (s, len)
CHAR *s;
int len;
{
register struct _collsym *csp;
char *x;
x = (char *)s;
for (csp = posix_collsyms; csp->name; csp++)
{
if (STREQN(csp->name, x, len) && csp->name[len] == '\0')
return (csp->code);
}
if (len == 1)
return s[0];
return INVALID;
}
/* unibyte character classification */
#if !defined (isascii) && !defined (HAVE_ISASCII)
# define isascii(c) ((unsigned int)(c) <= 0177)
#endif
enum char_class
{
CC_NO_CLASS = 0,
CC_ASCII, CC_ALNUM, CC_ALPHA, CC_BLANK, CC_CNTRL, CC_DIGIT, CC_GRAPH,
CC_LOWER, CC_PRINT, CC_PUNCT, CC_SPACE, CC_UPPER, CC_WORD, CC_XDIGIT
};
static char const *const cclass_name[] =
{
"",
"ascii", "alnum", "alpha", "blank", "cntrl", "digit", "graph",
"lower", "print", "punct", "space", "upper", "word", "xdigit"
};
#define N_CHAR_CLASS (sizeof(cclass_name) / sizeof (cclass_name[0]))
static int
is_cclass (c, name)
int c;
const char *name;
{
enum char_class char_class = CC_NO_CLASS;
int i, result;
for (i = 1; i < N_CHAR_CLASS; i++)
{
if (STREQ (name, cclass_name[i]))
{
char_class = (enum char_class)i;
break;
}
}
if (char_class == 0)
return -1;
switch (char_class)
{
case CC_ASCII:
result = isascii (c);
break;
case CC_ALNUM:
result = ISALNUM (c);
break;
case CC_ALPHA:
result = ISALPHA (c);
break;
case CC_BLANK:
result = ISBLANK (c);
break;
case CC_CNTRL:
result = ISCNTRL (c);
break;
case CC_DIGIT:
result = ISDIGIT (c);
break;
case CC_GRAPH:
result = ISGRAPH (c);
break;
case CC_LOWER:
result = ISLOWER (c);
break;
case CC_PRINT:
result = ISPRINT (c);
break;
case CC_PUNCT:
result = ISPUNCT (c);
break;
case CC_SPACE:
result = ISSPACE (c);
break;
case CC_UPPER:
result = ISUPPER (c);
break;
case CC_WORD:
result = (ISALNUM (c) || c == '_');
break;
case CC_XDIGIT:
result = ISXDIGIT (c);
break;
default:
result = -1;
break;
}
return result;
}
/* Now include `sm_loop.c' for single-byte characters. */
/* The result of FOLD is an `unsigned char' */
# define FOLD(c) ((flags & FNM_CASEFOLD) \
? TOLOWER ((unsigned char)c) \
: ((unsigned char)c))
#define FCT internal_strmatch
#define GMATCH gmatch
#define COLLSYM collsym
#define PARSE_COLLSYM parse_collsym
#define BRACKMATCH brackmatch
#define PATSCAN patscan
#define STRCOMPARE strcompare
#define EXTMATCH extmatch
#define STRCHR(S, C) strchr((S), (C))
#define STRCOLL(S1, S2) strcoll((S1), (S2))
#define STRLEN(S) strlen(S)
#define STRCMP(S1, S2) strcmp((S1), (S2))
#define RANGECMP(C1, C2) rangecmp((C1), (C2))
#define COLLEQUIV(C1, C2) collequiv((C1), (C2))
#define CTYPE_T enum char_class
#define IS_CCLASS(C, S) is_cclass((C), (S))
#include "sm_loop.c"
#if HANDLE_MULTIBYTE
# define CHAR wchar_t
# define U_CHAR wint_t
# define XCHAR wchar_t
# define INT wint_t
# define L(CS) L##CS
# define INVALID WEOF
# undef STREQ
# undef STREQN
# define STREQ(s1, s2) ((wcscmp (s1, s2) == 0))
# define STREQN(a, b, n) ((a)[0] == (b)[0] && wcsncmp(a, b, n) == 0)
extern char *mbsmbchar __P((const char *));
static int
rangecmp_wc (c1, c2)
wint_t c1, c2;
{
static wchar_t s1[2] = { L' ', L'\0' };
static wchar_t s2[2] = { L' ', L'\0' };
if (c1 == c2)
return 0;
s1[0] = c1;
s2[0] = c2;
return (wcscoll (s1, s2));
}
static int
collequiv_wc (c, equiv)
wint_t c, equiv;
{
return (!(c - equiv));
}
/* Helper function for collating symbol. */
# define _COLLSYM _collwcsym
# define __COLLSYM __collwcsym
# define POSIXCOLL posix_collwcsyms
# include "collsyms.h"
static wint_t
collwcsym (s, len)
wchar_t *s;
int len;
{
register struct _collwcsym *csp;
for (csp = posix_collwcsyms; csp->name; csp++)
{
if (STREQN(csp->name, s, len) && csp->name[len] == L'\0')
return (csp->code);
}
if (len == 1)
return s[0];
return INVALID;
}
static int
is_wcclass (wc, name)
wint_t wc;
wchar_t *name;
{
char *mbs;
mbstate_t state;
size_t mbslength;
wctype_t desc;
int want_word;
if ((wctype ("ascii") == (wctype_t)0) && (wcscmp (name, L"ascii") == 0))
{
int c;
if ((c = wctob (wc)) == EOF)
return 0;
else
return (c <= 0x7F);
}
want_word = (wcscmp (name, L"word") == 0);
if (want_word)
name = L"alnum";
memset (&state, '\0', sizeof (mbstate_t));
mbs = (char *) malloc (wcslen(name) * MB_CUR_MAX + 1);
mbslength = wcsrtombs (mbs, (const wchar_t **)&name, (wcslen(name) * MB_CUR_MAX + 1), &state);
if (mbslength == (size_t)-1 || mbslength == (size_t)-2)
{
free (mbs);
return -1;
}
desc = wctype (mbs);
free (mbs);
if (desc == (wctype_t)0)
return -1;
if (want_word)
return (iswctype (wc, desc) || wc == L'_');
else
return (iswctype (wc, desc));
}
/* Now include `sm_loop.c' for multibyte characters. */
#define FOLD(c) ((flags & FNM_CASEFOLD) && iswupper (c) ? towlower (c) : (c))
#define FCT internal_wstrmatch
#define GMATCH gmatch_wc
#define COLLSYM collwcsym
#define PARSE_COLLSYM parse_collwcsym
#define BRACKMATCH brackmatch_wc
#define PATSCAN patscan_wc
#define STRCOMPARE wscompare
#define EXTMATCH extmatch_wc
#define STRCHR(S, C) wcschr((S), (C))
#define STRCOLL(S1, S2) wcscoll((S1), (S2))
#define STRLEN(S) wcslen(S)
#define STRCMP(S1, S2) wcscmp((S1), (S2))
#define RANGECMP(C1, C2) rangecmp_wc((C1), (C2))
#define COLLEQUIV(C1, C2) collequiv_wc((C1), (C2))
#define CTYPE_T enum char_class
#define IS_CCLASS(C, S) is_wcclass((C), (S))
#include "sm_loop.c"
#endif /* HAVE_MULTIBYTE */
int
xstrmatch (pattern, string, flags)
char *pattern;
char *string;
int flags;
{
#if HANDLE_MULTIBYTE
int ret;
size_t n;
wchar_t *wpattern, *wstring;
size_t plen, slen, mplen, mslen;
#if 0
plen = strlen (pattern);
mplen = mbstrlen (pattern);
if (plen == mplen && strlen (string) == mbstrlen (string))
#else
if (mbsmbchar (string) == 0 && mbsmbchar (pattern) == 0)
#endif
return (internal_strmatch ((unsigned char *)pattern, (unsigned char *)string, flags));
if (MB_CUR_MAX == 1)
return (internal_strmatch ((unsigned char *)pattern, (unsigned char *)string, flags));
n = xdupmbstowcs (&wpattern, NULL, pattern);
if (n == (size_t)-1 || n == (size_t)-2)
return (internal_strmatch ((unsigned char *)pattern, (unsigned char *)string, flags));
n = xdupmbstowcs (&wstring, NULL, string);
if (n == (size_t)-1 || n == (size_t)-2)
{
free (wpattern);
return (internal_strmatch ((unsigned char *)pattern, (unsigned char *)string, flags));
}
ret = internal_wstrmatch (wpattern, wstring, flags);
free (wpattern);
free (wstring);
return ret;
#else
return (internal_strmatch ((unsigned char *)pattern, (unsigned char *)string, flags));
#endif /* !HANDLE_MULTIBYTE */
}
-366
View File
@@ -1,366 +0,0 @@
/* xmbsrtowcs.c -- replacement function for mbsrtowcs */
/* Copyright (C) 2002-2004 Free Software Foundation, Inc.
This file is part of GNU Bash, the Bourne Again SHell.
Bash 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 3 of the License, or
(at your option) any later version.
Bash 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 Bash. If not, see <http://www.gnu.org/licenses/>.
*/
/* Ask for GNU extensions to get extern declaration for mbsnrtowcs if
available via glibc. */
#ifndef _GNU_SOURCE
# define _GNU_SOURCE 1
#endif
#include <config.h>
#include <bashansi.h>
/* <wchar.h>, <wctype.h> and <stdlib.h> are included in "shmbutil.h".
If <wchar.h>, <wctype.h>, mbsrtowcs(), exist, HANDLE_MULTIBYTE
is defined as 1. */
#include <shmbutil.h>
#if HANDLE_MULTIBYTE
#ifndef FREE
# define FREE(x) do { if (x) free (x); } while (0)
#endif
#if ! HAVE_STRCHRNUL
extern char *strchrnul __P((const char *, int));
#endif
/* On some locales (ex. ja_JP.sjis), mbsrtowc doesn't convert 0x5c to U<0x5c>.
So, this function is made for converting 0x5c to U<0x5c>. */
static mbstate_t local_state;
static int local_state_use = 0;
size_t
xmbsrtowcs (dest, src, len, pstate)
wchar_t *dest;
const char **src;
size_t len;
mbstate_t *pstate;
{
mbstate_t *ps;
size_t mblength, wclength, n;
ps = pstate;
if (pstate == NULL)
{
if (!local_state_use)
{
memset (&local_state, '\0', sizeof(mbstate_t));
local_state_use = 1;
}
ps = &local_state;
}
n = strlen (*src);
if (dest == NULL)
{
wchar_t *wsbuf;
const char *mbs;
mbstate_t psbuf;
/* It doesn't matter if malloc fails here, since mbsrtowcs should do
the right thing with a NULL first argument. */
wsbuf = (wchar_t *) malloc ((n + 1) * sizeof(wchar_t));
mbs = *src;
psbuf = *ps;
wclength = mbsrtowcs (wsbuf, &mbs, n, &psbuf);
if (wsbuf)
free (wsbuf);
return wclength;
}
for (wclength = 0; wclength < len; wclength++, dest++)
{
if (mbsinit(ps))
{
if (**src == '\0')
{
*dest = L'\0';
*src = NULL;
return (wclength);
}
else if (**src == '\\')
{
*dest = L'\\';
mblength = 1;
}
else
mblength = mbrtowc(dest, *src, n, ps);
}
else
mblength = mbrtowc(dest, *src, n, ps);
/* Cannot convert multibyte character to wide character. */
if (mblength == (size_t)-1 || mblength == (size_t)-2)
return (size_t)-1;
*src += mblength;
n -= mblength;
/* The multibyte string has been completely converted,
including the terminating '\0'. */
if (*dest == L'\0')
{
*src = NULL;
break;
}
}
return (wclength);
}
#if HAVE_MBSNRTOWCS
/* Convert a multibyte string to a wide character string. Memory for the
new wide character string is obtained with malloc.
Fast multiple-character version of xdupmbstowcs used when the indices are
not required and mbsnrtowcs is available. */
static size_t
xdupmbstowcs2 (destp, src)
wchar_t **destp; /* Store the pointer to the wide character string */
const char *src; /* Multibyte character string */
{
const char *p; /* Conversion start position of src */
wchar_t *wsbuf; /* Buffer for wide characters. */
size_t wsbuf_size; /* Size of WSBUF */
size_t wcnum; /* Number of wide characters in WSBUF */
mbstate_t state; /* Conversion State */
size_t wcslength; /* Number of wide characters produced by the conversion. */
const char *end_or_backslash;
size_t nms; /* Number of multibyte characters to convert at one time. */
mbstate_t tmp_state;
const char *tmp_p;
memset (&state, '\0', sizeof(mbstate_t));
wsbuf_size = 0;
wsbuf = NULL;
p = src;
wcnum = 0;
do
{
end_or_backslash = strchrnul(p, '\\');
nms = (end_or_backslash - p);
if (*end_or_backslash == '\0')
nms++;
/* Compute the number of produced wide-characters. */
tmp_p = p;
tmp_state = state;
wcslength = mbsnrtowcs(NULL, &tmp_p, nms, 0, &tmp_state);
/* Conversion failed. */
if (wcslength == (size_t)-1)
{
free (wsbuf);
*destp = NULL;
return (size_t)-1;
}
/* Resize the buffer if it is not large enough. */
if (wsbuf_size < wcnum+wcslength+1) /* 1 for the L'\0' or the potential L'\\' */
{
wchar_t *wstmp;
wsbuf_size = wcnum+wcslength+1; /* 1 for the L'\0' or the potential L'\\' */
wstmp = (wchar_t *) realloc (wsbuf, wsbuf_size * sizeof (wchar_t));
if (wstmp == NULL)
{
free (wsbuf);
*destp = NULL;
return (size_t)-1;
}
wsbuf = wstmp;
}
/* Perform the conversion. This is assumed to return 'wcslength'.
* It may set 'p' to NULL. */
mbsnrtowcs(wsbuf+wcnum, &p, nms, wsbuf_size-wcnum, &state);
wcnum += wcslength;
if (mbsinit (&state) && (p != NULL) && (*p == '\\'))
{
wsbuf[wcnum++] = L'\\';
p++;
}
}
while (p != NULL);
*destp = wsbuf;
/* Return the length of the wide character string, not including `\0'. */
return wcnum;
}
#endif /* HAVE_MBSNRTOWCS */
/* Convert a multibyte string to a wide character string. Memory for the
new wide character string is obtained with malloc.
The return value is the length of the wide character string. Returns a
pointer to the wide character string in DESTP. If INDICESP is not NULL,
INDICESP stores the pointer to the pointer array. Each pointer is to
the first byte of each multibyte character. Memory for the pointer array
is obtained with malloc, too.
If conversion is failed, the return value is (size_t)-1 and the values
of DESTP and INDICESP are NULL. */
#define WSBUF_INC 32
size_t
xdupmbstowcs (destp, indicesp, src)
wchar_t **destp; /* Store the pointer to the wide character string */
char ***indicesp; /* Store the pointer to the pointer array. */
const char *src; /* Multibyte character string */
{
const char *p; /* Conversion start position of src */
wchar_t wc; /* Created wide character by conversion */
wchar_t *wsbuf; /* Buffer for wide characters. */
char **indices; /* Buffer for indices. */
size_t wsbuf_size; /* Size of WSBUF */
size_t wcnum; /* Number of wide characters in WSBUF */
mbstate_t state; /* Conversion State */
/* In case SRC or DESP is NULL, conversion doesn't take place. */
if (src == NULL || destp == NULL)
{
if (destp)
*destp = NULL;
return (size_t)-1;
}
#if HAVE_MBSNRTOWCS
if (indicesp == NULL)
return (xdupmbstowcs2 (destp, src));
#endif
memset (&state, '\0', sizeof(mbstate_t));
wsbuf_size = WSBUF_INC;
wsbuf = (wchar_t *) malloc (wsbuf_size * sizeof(wchar_t));
if (wsbuf == NULL)
{
*destp = NULL;
return (size_t)-1;
}
indices = NULL;
if (indicesp)
{
indices = (char **) malloc (wsbuf_size * sizeof(char *));
if (indices == NULL)
{
free (wsbuf);
*destp = NULL;
return (size_t)-1;
}
}
p = src;
wcnum = 0;
do
{
size_t mblength; /* Byte length of one multibyte character. */
if (mbsinit (&state))
{
if (*p == '\0')
{
wc = L'\0';
mblength = 1;
}
else if (*p == '\\')
{
wc = L'\\';
mblength = 1;
}
else
mblength = mbrtowc(&wc, p, MB_LEN_MAX, &state);
}
else
mblength = mbrtowc(&wc, p, MB_LEN_MAX, &state);
/* Conversion failed. */
if (MB_INVALIDCH (mblength))
{
free (wsbuf);
FREE (indices);
*destp = NULL;
return (size_t)-1;
}
++wcnum;
/* Resize buffers when they are not large enough. */
if (wsbuf_size < wcnum)
{
wchar_t *wstmp;
char **idxtmp;
wsbuf_size += WSBUF_INC;
wstmp = (wchar_t *) realloc (wsbuf, wsbuf_size * sizeof (wchar_t));
if (wstmp == NULL)
{
free (wsbuf);
FREE (indices);
*destp = NULL;
return (size_t)-1;
}
wsbuf = wstmp;
if (indicesp)
{
idxtmp = (char **) realloc (indices, wsbuf_size * sizeof (char **));
if (idxtmp == NULL)
{
free (wsbuf);
free (indices);
*destp = NULL;
return (size_t)-1;
}
indices = idxtmp;
}
}
wsbuf[wcnum - 1] = wc;
if (indices)
indices[wcnum - 1] = (char *)p;
p += mblength;
}
while (MB_NULLWCH (wc) == 0);
/* Return the length of the wide character string, not including `\0'. */
*destp = wsbuf;
if (indicesp != NULL)
*indicesp = indices;
return (wcnum - 1);
}
#endif /* HANDLE_MULTIBYTE */