Google Git
Sign in
goma / wine / refs/tags/wine-20030408 / . / ole / ole2nls.c
blob: a812f0be25ae095173304255bfa8a7ec40f3b592 [file] [log] [blame]
/*
* National Language Support library
*
* Copyright 1995 Martin von Loewis
* Copyright 1998 David Lee Lambert
* Copyright 2000 Julio César Gázquez
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This 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
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
#include "config.h"
#include "wine/port.h"
#include <string.h>
#include <stdio.h>
#include <ctype.h>
#include <stdlib.h>
#include <locale.h>
#include "windef.h"
#include "winbase.h"
#include "wingdi.h"
#include "winuser.h"
#include "winternl.h"
#include "wine/unicode.h"
#include "winver.h"
#include "winnls.h"
#include "winreg.h"
#include "winerror.h"
#include "wine/debug.h"
WINE_DEFAULT_DEBUG_CHANNEL(nls);
static const unsigned char LCM_Unicode_LUT[] = {
6 , 3, /* - 1 */
6 , 4, /* - 2 */
6 , 5, /* - 3 */
6 , 6, /* - 4 */
6 , 7, /* - 5 */
6 , 8, /* - 6 */
6 , 9, /* - 7 */
6 , 10, /* - 8 */
7 , 5, /* - 9 */
7 , 6, /* - 10 */
7 , 7, /* - 11 */
7 , 8, /* - 12 */
7 , 9, /* - 13 */
6 , 11, /* - 14 */
6 , 12, /* - 15 */
6 , 13, /* - 16 */
6 , 14, /* - 17 */
6 , 15, /* - 18 */
6 , 16, /* - 19 */
6 , 17, /* - 20 */
6 , 18, /* - 21 */
6 , 19, /* - 22 */
6 , 20, /* - 23 */
6 , 21, /* - 24 */
6 , 22, /* - 25 */
6 , 23, /* - 26 */
6 , 24, /* - 27 */
6 , 25, /* - 28 */
6 , 26, /* - 29 */
6 , 27, /* - 30 */
6 , 28, /* - 31 */
7 , 2, /* - 32 */
7 , 28, /* ! - 33 */
7 , 29, /* " - 34 */ /* " */
7 , 31, /* # - 35 */
7 , 33, /* $ - 36 */
7 , 35, /* % - 37 */
7 , 37, /* & - 38 */
6 , 128, /* ' - 39 */
7 , 39, /* ( - 40 */
7 , 42, /* ) - 41 */
7 , 45, /* * - 42 */
8 , 3, /* + - 43 */
7 , 47, /* , - 44 */
6 , 130, /* - - 45 */
7 , 51, /* . - 46 */
7 , 53, /* / - 47 */
12 , 3, /* 0 - 48 */
12 , 33, /* 1 - 49 */
12 , 51, /* 2 - 50 */
12 , 70, /* 3 - 51 */
12 , 88, /* 4 - 52 */
12 , 106, /* 5 - 53 */
12 , 125, /* 6 - 54 */
12 , 144, /* 7 - 55 */
12 , 162, /* 8 - 56 */
12 , 180, /* 9 - 57 */
7 , 55, /* : - 58 */
7 , 58, /* ; - 59 */
8 , 14, /* < - 60 */
8 , 18, /* = - 61 */
8 , 20, /* > - 62 */
7 , 60, /* ? - 63 */
7 , 62, /* @ - 64 */
14 , 2, /* A - 65 */
14 , 9, /* B - 66 */
14 , 10, /* C - 67 */
14 , 26, /* D - 68 */
14 , 33, /* E - 69 */
14 , 35, /* F - 70 */
14 , 37, /* G - 71 */
14 , 44, /* H - 72 */
14 , 50, /* I - 73 */
14 , 53, /* J - 74 */
14 , 54, /* K - 75 */
14 , 72, /* L - 76 */
14 , 81, /* M - 77 */
14 , 112, /* N - 78 */
14 , 124, /* O - 79 */
14 , 126, /* P - 80 */
14 , 137, /* Q - 81 */
14 , 138, /* R - 82 */
14 , 145, /* S - 83 */
14 , 153, /* T - 84 */
14 , 159, /* U - 85 */
14 , 162, /* V - 86 */
14 , 164, /* W - 87 */
14 , 166, /* X - 88 */
14 , 167, /* Y - 89 */
14 , 169, /* Z - 90 */
7 , 63, /* [ - 91 */
7 , 65, /* \ - 92 */
7 , 66, /* ] - 93 */
7 , 67, /* ^ - 94 */
7 , 68, /* _ - 95 */
7 , 72, /* ` - 96 */
14 , 2, /* a - 97 */
14 , 9, /* b - 98 */
14 , 10, /* c - 99 */
14 , 26, /* d - 100 */
14 , 33, /* e - 101 */
14 , 35, /* f - 102 */
14 , 37, /* g - 103 */
14 , 44, /* h - 104 */
14 , 50, /* i - 105 */
14 , 53, /* j - 106 */
14 , 54, /* k - 107 */
14 , 72, /* l - 108 */
14 , 81, /* m - 109 */
14 , 112, /* n - 110 */
14 , 124, /* o - 111 */
14 , 126, /* p - 112 */
14 , 137, /* q - 113 */
14 , 138, /* r - 114 */
14 , 145, /* s - 115 */
14 , 153, /* t - 116 */
14 , 159, /* u - 117 */
14 , 162, /* v - 118 */
14 , 164, /* w - 119 */
14 , 166, /* x - 120 */
14 , 167, /* y - 121 */
14 , 169, /* z - 122 */
7 , 74, /* { - 123 */
7 , 76, /* | - 124 */
7 , 78, /* } - 125 */
7 , 80, /* ~ - 126 */
6 , 29, /*  - 127 */
6 , 30, /* € - 128 */
6 , 31, /*  - 129 */
7 , 123, /* ‚ - 130 */
14 , 35, /* ƒ - 131 */
7 , 127, /* „ - 132 */
10 , 21, /* … - 133 */
10 , 15, /* † - 134 */
10 , 16, /* ‡ - 135 */
7 , 67, /* ˆ - 136 */
10 , 22, /* ‰ - 137 */
14 , 145, /* Š - 138 */
7 , 136, /* ‹ - 139 */
14 + 16 , 124, /* Π- 140 */
6 , 43, /*  - 141 */
6 , 44, /* Ž - 142 */
6 , 45, /*  - 143 */
6 , 46, /*  - 144 */
7 , 121, /* ‘ - 145 */
7 , 122, /* ’ - 146 */
7 , 125, /* “ - 147 */
7 , 126, /* ” - 148 */
10 , 17, /* • - 149 */
6 , 137, /* – - 150 */
6 , 139, /* — - 151 */
7 , 93, /* ˜ - 152 */
14 , 156, /* ™ - 153 */
14 , 145, /* š - 154 */
7 , 137, /* › - 155 */
14 + 16 , 124, /* œ - 156 */
6 , 59, /*  - 157 */
6 , 60, /* ž - 158 */
14 , 167, /* Ÿ - 159 */
7 , 4, /*   - 160 */
7 , 81, /* ¡ - 161 */
10 , 2, /* ¢ - 162 */
10 , 3, /* £ - 163 */
10 , 4, /* ¤ - 164 */
10 , 5, /* ¥ - 165 */
7 , 82, /* ¦ - 166 */
10 , 6, /* § - 167 */
7 , 83, /* ¨ - 168 */
10 , 7, /* © - 169 */
14 , 2, /* ª - 170 */
8 , 24, /* « - 171 */
10 , 8, /* ¬ - 172 */
6 , 131, /* ­ - 173 */
10 , 9, /* ® - 174 */
7 , 84, /* ¯ - 175 */
10 , 10, /* ° - 176 */
8 , 23, /* ± - 177 */
12 , 51, /* ² - 178 */
12 , 70, /* ³ - 179 */
7 , 85, /* ´ - 180 */
10 , 11, /* µ - 181 */
10 , 12, /* ¶ - 182 */
10 , 13, /* · - 183 */
7 , 86, /* ¸ - 184 */
12 , 33, /* ¹ - 185 */
14 , 124, /* º - 186 */
8 , 26, /* » - 187 */
12 , 21, /* ¼ - 188 */
12 , 25, /* ½ - 189 */
12 , 29, /* ¾ - 190 */
7 , 87, /* ¿ - 191 */
14 , 2, /* À - 192 */
14 , 2, /* Á - 193 */
14 , 2, /* Â - 194 */
14 , 2, /* Ã - 195 */
14 , 2, /* Ä - 196 */
14 , 2, /* Å - 197 */
14 + 16 , 2, /* Æ - 198 */
14 , 10, /* Ç - 199 */
14 , 33, /* È - 200 */
14 , 33, /* É - 201 */
14 , 33, /* Ê - 202 */
14 , 33, /* Ë - 203 */
14 , 50, /* Ì - 204 */
14 , 50, /* Í - 205 */
14 , 50, /* Î - 206 */
14 , 50, /* Ï - 207 */
14 , 26, /* Ð - 208 */
14 , 112, /* Ñ - 209 */
14 , 124, /* Ò - 210 */
14 , 124, /* Ó - 211 */
14 , 124, /* Ô - 212 */
14 , 124, /* Õ - 213 */
14 , 124, /* Ö - 214 */
8 , 28, /* × - 215 */
14 , 124, /* Ø - 216 */
14 , 159, /* Ù - 217 */
14 , 159, /* Ú - 218 */
14 , 159, /* Û - 219 */
14 , 159, /* Ü - 220 */
14 , 167, /* Ý - 221 */
14 + 32 , 153, /* Þ - 222 */
14 + 48 , 145, /* ß - 223 */
14 , 2, /* à - 224 */
14 , 2, /* á - 225 */
14 , 2, /* â - 226 */
14 , 2, /* ã - 227 */
14 , 2, /* ä - 228 */
14 , 2, /* å - 229 */
14 + 16 , 2, /* æ - 230 */
14 , 10, /* ç - 231 */
14 , 33, /* è - 232 */
14 , 33, /* é - 233 */
14 , 33, /* ê - 234 */
14 , 33, /* ë - 235 */
14 , 50, /* ì - 236 */
14 , 50, /* í - 237 */
14 , 50, /* î - 238 */
14 , 50, /* ï - 239 */
14 , 26, /* ð - 240 */
14 , 112, /* ñ - 241 */
14 , 124, /* ò - 242 */
14 , 124, /* ó - 243 */
14 , 124, /* ô - 244 */
14 , 124, /* õ - 245 */
14 , 124, /* ö - 246 */
8 , 29, /* ÷ - 247 */
14 , 124, /* ø - 248 */
14 , 159, /* ù - 249 */
14 , 159, /* ú - 250 */
14 , 159, /* û - 251 */
14 , 159, /* ü - 252 */
14 , 167, /* ý - 253 */
14 + 32 , 153, /* þ - 254 */
14 , 167 /* ÿ - 255 */ };
static const unsigned char LCM_Unicode_LUT_2[] = { 33, 44, 145 };
#define LCM_Diacritic_Start 131
static const unsigned char LCM_Diacritic_LUT[] = {
123, /* ƒ - 131 */
2, /* „ - 132 */
2, /* … - 133 */
2, /* † - 134 */
2, /* ‡ - 135 */
3, /* ˆ - 136 */
2, /* ‰ - 137 */
20, /* Š - 138 */
2, /* ‹ - 139 */
2, /* Π- 140 */
2, /*  - 141 */
2, /* Ž - 142 */
2, /*  - 143 */
2, /*  - 144 */
2, /* ‘ - 145 */
2, /* ’ - 146 */
2, /* “ - 147 */
2, /* ” - 148 */
2, /* • - 149 */
2, /* – - 150 */
2, /* — - 151 */
2, /* ˜ - 152 */
2, /* ™ - 153 */
20, /* š - 154 */
2, /* › - 155 */
2, /* œ - 156 */
2, /*  - 157 */
2, /* ž - 158 */
19, /* Ÿ - 159 */
2, /*   - 160 */
2, /* ¡ - 161 */
2, /* ¢ - 162 */
2, /* £ - 163 */
2, /* ¤ - 164 */
2, /* ¥ - 165 */
2, /* ¦ - 166 */
2, /* § - 167 */
2, /* ¨ - 168 */
2, /* © - 169 */
3, /* ª - 170 */
2, /* « - 171 */
2, /* ¬ - 172 */
2, /* ­ - 173 */
2, /* ® - 174 */
2, /* ¯ - 175 */
2, /* ° - 176 */
2, /* ± - 177 */
2, /* ² - 178 */
2, /* ³ - 179 */
2, /* ´ - 180 */
2, /* µ - 181 */
2, /* ¶ - 182 */
2, /* · - 183 */
2, /* ¸ - 184 */
2, /* ¹ - 185 */
3, /* º - 186 */
2, /* » - 187 */
2, /* ¼ - 188 */
2, /* ½ - 189 */
2, /* ¾ - 190 */
2, /* ¿ - 191 */
15, /* À - 192 */
14, /* Á - 193 */
18, /* Â - 194 */
25, /* Ã - 195 */
19, /* Ä - 196 */
26, /* Å - 197 */
2, /* Æ - 198 */
28, /* Ç - 199 */
15, /* È - 200 */
14, /* É - 201 */
18, /* Ê - 202 */
19, /* Ë - 203 */
15, /* Ì - 204 */
14, /* Í - 205 */
18, /* Î - 206 */
19, /* Ï - 207 */
104, /* Ð - 208 */
25, /* Ñ - 209 */
15, /* Ò - 210 */
14, /* Ó - 211 */
18, /* Ô - 212 */
25, /* Õ - 213 */
19, /* Ö - 214 */
2, /* × - 215 */
33, /* Ø - 216 */
15, /* Ù - 217 */
14, /* Ú - 218 */
18, /* Û - 219 */
19, /* Ü - 220 */
14, /* Ý - 221 */
2, /* Þ - 222 */
2, /* ß - 223 */
15, /* à - 224 */
14, /* á - 225 */
18, /* â - 226 */
25, /* ã - 227 */
19, /* ä - 228 */
26, /* å - 229 */
2, /* æ - 230 */
28, /* ç - 231 */
15, /* è - 232 */
14, /* é - 233 */
18, /* ê - 234 */
19, /* ë - 235 */
15, /* ì - 236 */
14, /* í - 237 */
18, /* î - 238 */
19, /* ï - 239 */
104, /* ð - 240 */
25, /* ñ - 241 */
15, /* ò - 242 */
14, /* ó - 243 */
18, /* ô - 244 */
25, /* õ - 245 */
19, /* ö - 246 */
2, /* ÷ - 247 */
33, /* ø - 248 */
15, /* ù - 249 */
14, /* ú - 250 */
18, /* û - 251 */
19, /* ü - 252 */
14, /* ý - 253 */
2, /* þ - 254 */
19, /* ÿ - 255 */
} ;
/******************************************************************************
* OLE2NLS_isPunctuation [INTERNAL]
*/
static int OLE2NLS_isPunctuation(unsigned char c)
{
/* "punctuation character" in this context is a character which is
considered "less important" during word sort comparison.
See LCMapString implementation for the precise definition
of "less important". */
return (LCM_Unicode_LUT[-2+2*c]==6);
}
/******************************************************************************
* OLE2NLS_isNonSpacing [INTERNAL]
*/
static int OLE2NLS_isNonSpacing(unsigned char c)
{
/* This function is used by LCMapStringA. Characters
for which it returns true are ignored when mapping a
string with NORM_IGNORENONSPACE */
return ((c==136) || (c==170) || (c==186));
}
/******************************************************************************
* OLE2NLS_isSymbol [INTERNAL]
* FIXME: handle current locale
*/
static int OLE2NLS_isSymbol(unsigned char c)
{
/* This function is used by LCMapStringA. Characters
for which it returns true are ignored when mapping a
string with NORM_IGNORESYMBOLS */
return ( (c!=0) && !(isalpha(c) || isdigit(c)) );
}
/******************************************************************************
* identity [Internal]
*/
static int identity(int c)
{
return c;
}
/*************************************************************************
* LCMapStringA [KERNEL32.@]
*
* Convert a string, or generate a sort key from it.
*
* RETURNS
* Success: The length of the string written to dststr.
* Failure: 0.
*
* NOTES
* -If mapflags includes LCMAP_SORTKEY, the function will generate a
* sort key for srcstr. Otherwise, srcstr is converted according to
* mapflags.
* -If scrlen is -1, the function will compute the length of strsrc
* (which must be NUL terminated) itself.
* -If dstlen is 0, The return value is the buffer length that is needed.
* -NORM_IGNOREWIDTH means to compare ASCII and wide characters
* as if they are equal.
* In the only code page implemented so far, there may not be
* wide characters in strings passed to this function,
* so there is nothing to be done for this flag.
*/
INT WINAPI LCMapStringA(
LCID lcid, /* [in] Locale Id */
DWORD mapflags, /* [in] Flags */
LPCSTR srcstr, /* [in] Source buffer */
INT srclen, /* [in] Length of srcstr */
LPSTR dststr, /* [out] Destination buffer */
INT dstlen) /* [in] Length of dststr */
{
int i;
TRACE("(0x%04lx,0x%08lx,%s,%d,%p,%d)\n",
lcid,mapflags,debugstr_an(srcstr,srclen),srclen,dststr,dstlen);
if ( ((dstlen!=0) && (dststr==NULL)) || (srcstr==NULL) )
{
ERR("(src=%s,dest=%s): Invalid NULL string\n",
debugstr_an(srcstr,srclen), dststr);
SetLastError(ERROR_INVALID_PARAMETER);
return 0;
}
if (srclen == -1)
srclen = strlen(srcstr) + 1 ; /* (include final '\0') */
#define LCMAPSTRINGA_SUPPORTED_FLAGS (LCMAP_UPPERCASE | \
LCMAP_LOWERCASE | \
LCMAP_SORTKEY | \
NORM_IGNORECASE | \
NORM_IGNORENONSPACE | \
SORT_STRINGSORT | \
NORM_IGNOREWIDTH | \
NORM_IGNOREKANATYPE)
/* FIXME: as long as we don't support Katakana nor Hiragana
* characters, we can support NORM_IGNOREKANATYPE
*/
if (mapflags & ~LCMAPSTRINGA_SUPPORTED_FLAGS)
{
FIXME("(0x%04lx,0x%08lx,%p,%d,%p,%d): "
"unimplemented flags: 0x%08lx\n",
lcid,
mapflags,
srcstr,
srclen,
dststr,
dstlen,
mapflags & ~LCMAPSTRINGA_SUPPORTED_FLAGS
);
}
if ( !(mapflags & LCMAP_SORTKEY) )
{
int i,j;
int (*f)(int) = identity;
int flag_ignorenonspace = mapflags & NORM_IGNORENONSPACE;
int flag_ignoresymbols = mapflags & NORM_IGNORESYMBOLS;
if (flag_ignorenonspace || flag_ignoresymbols)
{
/* For some values of mapflags, the length of the resulting
string is not known at this point. Windows does map the string
and does not SetLastError ERROR_INSUFFICIENT_BUFFER in
these cases. */
if (dstlen==0)
{
/* Compute required length */
for (i=j=0; i < srclen; i++)
{
if ( !(flag_ignorenonspace && OLE2NLS_isNonSpacing(srcstr[i]))
&& !(flag_ignoresymbols && OLE2NLS_isSymbol(srcstr[i])) )
j++;
}
return j;
}
}
else
{
if (dstlen==0)
return srclen;
if (dstlen<srclen)
{
SetLastError(ERROR_INSUFFICIENT_BUFFER);
return 0;
}
}
if (mapflags & LCMAP_UPPERCASE)
f = toupper;
else if (mapflags & LCMAP_LOWERCASE)
f = tolower;
/* FIXME: NORM_IGNORENONSPACE requires another conversion */
for (i=j=0; (i<srclen) && (j<dstlen) ; i++)
{
if ( !(flag_ignorenonspace && OLE2NLS_isNonSpacing(srcstr[i]))
&& !(flag_ignoresymbols && OLE2NLS_isSymbol(srcstr[i])) )
{
dststr[j] = (CHAR) f(srcstr[i]);
j++;
}
}
return j;
}
/* FIXME: This function completely ignores the "lcid" parameter. */
/* else ... (mapflags & LCMAP_SORTKEY) */
{
int unicode_len=0;
int case_len=0;
int diacritic_len=0;
int delayed_punctuation_len=0;
char *case_component;
char *diacritic_component;
char *delayed_punctuation_component;
int room,count;
int flag_stringsort = mapflags & SORT_STRINGSORT;
/* compute how much room we will need */
for (i=0;i<srclen;i++)
{
int ofs;
unsigned char source_char = srcstr[i];
if (source_char!='\0')
{
if (flag_stringsort || !OLE2NLS_isPunctuation(source_char))
{
unicode_len++;
if ( LCM_Unicode_LUT[-2+2*source_char] & ~15 )
unicode_len++; /* double letter */
}
else
{
delayed_punctuation_len++;
}
}
if (isupper(source_char))
case_len=unicode_len;
ofs = source_char - LCM_Diacritic_Start;
if ((ofs>=0) && (LCM_Diacritic_LUT[ofs]!=2))
diacritic_len=unicode_len;
}
if (mapflags & NORM_IGNORECASE)
case_len=0;
if (mapflags & NORM_IGNORENONSPACE)
diacritic_len=0;
room = 2 * unicode_len /* "unicode" component */
+ diacritic_len /* "diacritic" component */
+ case_len /* "case" component */
+ 4 * delayed_punctuation_len /* punctuation in word sort mode */
+ 4 /* four '\1' separators */
+ 1 ; /* terminal '\0' */
if (dstlen==0)
return room;
else if (dstlen<room)
{
SetLastError(ERROR_INSUFFICIENT_BUFFER);
return 0;
}
#if 0
/*FIXME the Pointercheck should not be nessesary */
if (IsBadWritePtr (dststr,room))
{ ERR("bad destination buffer (dststr) : %p,%d\n",dststr,dstlen);
SetLastError(ERROR_INSUFFICIENT_BUFFER);
return 0;
}
#endif
/* locate each component, write separators */
diacritic_component = dststr + 2*unicode_len ;
*diacritic_component++ = '\1';
case_component = diacritic_component + diacritic_len ;
*case_component++ = '\1';
delayed_punctuation_component = case_component + case_len ;
*delayed_punctuation_component++ = '\1';
*delayed_punctuation_component++ = '\1';
/* read source string char by char, write
corresponding weight in each component. */
for (i=0,count=0;i<srclen;i++)
{
unsigned char source_char=srcstr[i];
if (source_char!='\0')
{
int type,longcode;
type = LCM_Unicode_LUT[-2+2*source_char];
longcode = type >> 4;
type &= 15;
if (!flag_stringsort && OLE2NLS_isPunctuation(source_char))
{
WORD encrypted_location = (1<<15) + 7 + 4*count;
*delayed_punctuation_component++ = (unsigned char) (encrypted_location>>8);
*delayed_punctuation_component++ = (unsigned char) (encrypted_location&255);
/* big-endian is used here because it lets string comparison be
compatible with numerical comparison */
*delayed_punctuation_component++ = type;
*delayed_punctuation_component++ = LCM_Unicode_LUT[-1+2*source_char];
/* assumption : a punctuation character is never a
double or accented letter */
}
else
{
dststr[2*count] = type;
dststr[2*count+1] = LCM_Unicode_LUT[-1+2*source_char];
if (longcode)
{
if (count<case_len)
case_component[count] = ( isupper(source_char) ? 18 : 2 ) ;
if (count<diacritic_len)
diacritic_component[count] = 2; /* assumption: a double letter
is never accented */
count++;
dststr[2*count] = type;
dststr[2*count+1] = *(LCM_Unicode_LUT_2 - 1 + longcode);
/* 16 in the first column of LCM_Unicode_LUT --> longcode = 1
32 in the first column of LCM_Unicode_LUT --> longcode = 2
48 in the first column of LCM_Unicode_LUT --> longcode = 3 */
}
if (count<case_len)
case_component[count] = ( isupper(source_char) ? 18 : 2 ) ;
if (count<diacritic_len)
{
int ofs = source_char - LCM_Diacritic_Start;
diacritic_component[count] = (ofs>=0 ? LCM_Diacritic_LUT[ofs] : 2);
}
count++;
}
}
}
dststr[room-1] = '\0';
return room;
}
}
/*************************************************************************
* LCMapStringW [KERNEL32.@]
*
* See LCMapStringA.
*/
INT WINAPI LCMapStringW(
LCID lcid,DWORD mapflags,LPCWSTR srcstr,INT srclen,LPWSTR dststr,
INT dstlen)
{
int i;
TRACE("(0x%04lx,0x%08lx,%p,%d,%p,%d)\n",
lcid, mapflags, srcstr, srclen, dststr, dstlen);
if ( ((dstlen!=0) && (dststr==NULL)) || (srcstr==NULL) )
{
ERR("(src=%p,dst=%p): Invalid NULL string\n", srcstr, dststr);
SetLastError(ERROR_INVALID_PARAMETER);
return 0;
}
if (srclen==-1)
srclen = strlenW(srcstr)+1;
/* FIXME: Both this function and it's companion LCMapStringA()
* completely ignore the "lcid" parameter. In place of the "lcid"
* parameter the application must set the "LC_COLLATE" or "LC_ALL"
* environment variable prior to invoking this function. */
if (mapflags & LCMAP_SORTKEY)
{
/* Possible values of LC_COLLATE. */
char *lc_collate_default = 0; /* value prior to this function */
char *lc_collate_env = 0; /* value retrieved from the environment */
/* General purpose index into strings of any type. */
int str_idx = 0;
/* Lengths of various strings where the length is measured in
* wide characters for wide character strings and in bytes for
* native strings. The lengths include the NULL terminator. */
size_t returned_len = 0;
size_t src_native_len = 0;
size_t dst_native_len = 0;
size_t dststr_libc_len = 0;
/* Native (character set determined by locale) versions of the
* strings source and destination strings. */
LPSTR src_native = 0;
LPSTR dst_native = 0;
/* Version of the source and destination strings using the
* "wchar_t" Unicode data type needed by various libc functions. */
wchar_t *srcstr_libc = 0;
wchar_t *dststr_libc = 0;
if(!(srcstr_libc = (wchar_t *)HeapAlloc(GetProcessHeap(), 0,
srclen * sizeof(wchar_t))))
{
ERR("Unable to allocate %d bytes for srcstr_libc\n",
srclen * sizeof(wchar_t));
SetLastError(ERROR_NOT_ENOUGH_MEMORY);
return 0;
}
/* Convert source string to a libc Unicode string. */
for(str_idx = 0; str_idx < srclen; str_idx++)
{
srcstr_libc[str_idx] = srcstr[str_idx];
}
/* src_native should contain at most 3 bytes for each
* multibyte characters in the original srcstr string. */
src_native_len = 3 * srclen;
if(!(src_native = (LPSTR)HeapAlloc(GetProcessHeap(), 0,
src_native_len)))
{
ERR("Unable to allocate %d bytes for src_native\n", src_native_len);
SetLastError(ERROR_NOT_ENOUGH_MEMORY);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
return 0;
}
/* FIXME: Prior to to setting the LC_COLLATE locale category the
* current value is backed up so it can be restored after the
* last LC_COLLATE sensitive function returns.
*
* Even though the locale is adjusted for a minimum amount of
* time a race condition exists where other threads may be
* affected if they invoke LC_COLLATE sensitive functions. One
* possible solution is to wrap all LC_COLLATE sensitive Wine
* functions, like LCMapStringW(), in a mutex.
*
* Another enhancement to the following would be to set the
* LC_COLLATE locale category as a function of the "lcid"
* parameter instead of the "LC_COLLATE" environment variable. */
if(!(lc_collate_default = setlocale(LC_COLLATE, NULL)))
{
ERR("Unable to query the LC_COLLATE catagory\n");
SetLastError(ERROR_INVALID_PARAMETER);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
return 0;
}
if(!(lc_collate_env = setlocale(LC_COLLATE, "")))
{
ERR("Unable to inherit the LC_COLLATE locale category from the "
"environment. The \"LC_COLLATE\" environment variable is "
"\"%s\".\n", getenv("LC_COLLATE") ?
getenv("LC_COLLATE") : "<unset>");
SetLastError(ERROR_INVALID_PARAMETER);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
return 0;
}
TRACE("lc_collate_default = %s\n", lc_collate_default);
TRACE("lc_collate_env = %s\n", lc_collate_env);
/* Convert the libc Unicode string to a native multibyte character
* string. */
returned_len = wcstombs(src_native, srcstr_libc, src_native_len) + 1;
if(returned_len == 0)
{
ERR("wcstombs failed. The string specified (%s) may contain an invalid character.\n",
debugstr_w(srcstr));
SetLastError(ERROR_INVALID_PARAMETER);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
setlocale(LC_COLLATE, lc_collate_default);
return 0;
}
else if(returned_len > src_native_len)
{
src_native[src_native_len - 1] = 0;
ERR("wcstombs returned a string (%s) that was longer (%d bytes) "
"than expected (%d bytes).\n", src_native, returned_len,
dst_native_len);
/* Since this is an internal error I'm not sure what the correct
* error code is. */
SetLastError(ERROR_NOT_ENOUGH_MEMORY);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
setlocale(LC_COLLATE, lc_collate_default);
return 0;
}
src_native_len = returned_len;
TRACE("src_native = %s src_native_len = %d\n",
src_native, src_native_len);
/* dst_native seems to contain at most 4 bytes for each byte in
* the original src_native string. Change if need be since this
* isn't documented by the strxfrm() man page. */
dst_native_len = 4 * src_native_len;
if(!(dst_native = (LPSTR)HeapAlloc(GetProcessHeap(), 0, dst_native_len)))
{
ERR("Unable to allocate %d bytes for dst_native\n", dst_native_len);
SetLastError(ERROR_NOT_ENOUGH_MEMORY);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
setlocale(LC_COLLATE, lc_collate_default);
return 0;
}
/* The actual translation is done by the following call to
* strxfrm(). The surrounding code could have been simplified
* by calling wcsxfrm() instead except that wcsxfrm() is not
* available on older Linux systems (RedHat 4.1 with
* libc-5.3.12-17).
*
* Also, it is possible that the translation could be done by
* various tables as it is done in LCMapStringA(). However, I'm
* not sure what those tables are. */
returned_len = strxfrm(dst_native, src_native, dst_native_len) + 1;
if(returned_len > dst_native_len)
{
dst_native[dst_native_len - 1] = 0;
ERR("strxfrm returned a string (%s) that was longer (%d bytes) "
"than expected (%d bytes).\n", dst_native, returned_len,
dst_native_len);
/* Since this is an internal error I'm not sure what the correct
* error code is. */
SetLastError(ERROR_NOT_ENOUGH_MEMORY);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
if(dst_native) HeapFree(GetProcessHeap(), 0, dst_native);
setlocale(LC_COLLATE, lc_collate_default);
return 0;
}
dst_native_len = returned_len;
TRACE("dst_native = %s dst_native_len = %d\n",
dst_native, dst_native_len);
dststr_libc_len = dst_native_len;
if(!(dststr_libc = (wchar_t *)HeapAlloc(GetProcessHeap(), 0,
dststr_libc_len * sizeof(wchar_t))))
{
ERR("Unable to allocate %d bytes for dststr_libc\n",
dststr_libc_len * sizeof(wchar_t));
SetLastError(ERROR_NOT_ENOUGH_MEMORY);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
if(dst_native) HeapFree(GetProcessHeap(), 0, dst_native);
setlocale(LC_COLLATE, lc_collate_default);
return 0;
}
/* Convert the native multibyte string to a libc Unicode string. */
returned_len = mbstowcs(dststr_libc, dst_native, dst_native_len) + 1;
/* Restore LC_COLLATE now that the last LC_COLLATE sensitive
* function has returned. */
setlocale(LC_COLLATE, lc_collate_default);
if(returned_len == 0)
{
ERR("mbstowcs failed. The native version of the translated string "
"(%s) may contain an invalid character.\n", dst_native);
SetLastError(ERROR_INVALID_PARAMETER);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
if(dst_native) HeapFree(GetProcessHeap(), 0, dst_native);
if(dststr_libc) HeapFree(GetProcessHeap(), 0, dststr_libc);
return 0;
}
if(dstlen)
{
if(returned_len > dstlen)
{
ERR("mbstowcs returned a string that was longer (%d chars) "
"than the buffer provided (%d chars).\n", returned_len,
dstlen);
SetLastError(ERROR_INSUFFICIENT_BUFFER);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
if(dst_native) HeapFree(GetProcessHeap(), 0, dst_native);
if(dststr_libc) HeapFree(GetProcessHeap(), 0, dststr_libc);
return 0;
}
dstlen = returned_len;
/* Convert a libc Unicode string to the destination string. */
for(str_idx = 0; str_idx < dstlen; str_idx++)
{
dststr[str_idx] = dststr_libc[str_idx];
}
TRACE("1st 4 int sized chunks of dststr = %x %x %x %x\n",
*(((int *)dststr) + 0),
*(((int *)dststr) + 1),
*(((int *)dststr) + 2),
*(((int *)dststr) + 3));
}
else
{
dstlen = returned_len;
}
TRACE("dstlen (return) = %d\n", dstlen);
if(srcstr_libc) HeapFree(GetProcessHeap(), 0, srcstr_libc);
if(src_native) HeapFree(GetProcessHeap(), 0, src_native);
if(dst_native) HeapFree(GetProcessHeap(), 0, dst_native);
if(dststr_libc) HeapFree(GetProcessHeap(), 0, dststr_libc);
return dstlen;
}
else
{
int (*f)(int)=identity;
if (dstlen==0)
return srclen;
if (dstlen<srclen)
{
SetLastError(ERROR_INSUFFICIENT_BUFFER);
return 0;
}
if (mapflags & LCMAP_UPPERCASE)
f = toupper;
else if (mapflags & LCMAP_LOWERCASE)
f = tolower;
for (i=0; i < srclen; i++)
dststr[i] = (WCHAR) f(srcstr[i]);
return srclen;
}
}
/***********************************************************************
* OLE2NLS_EstimateMappingLength
*
* Estimates the number of characters required to hold the string
* computed by LCMapStringA.
*
* The size is always over-estimated, with a fixed limit on the
* amount of estimation error.
*
* Note that len == -1 is not permitted.
*/
static inline int OLE2NLS_EstimateMappingLength(LCID lcid, DWORD dwMapFlags,
LPCSTR str, DWORD len)
{
/* Estimate only for small strings to keep the estimation error from
* becoming too large. */
if (len < 128) return len * 8 + 5;
else return LCMapStringA(lcid, dwMapFlags, str, len, NULL, 0);
}
/******************************************************************************
* CompareStringA [KERNEL32.@]
* Compares two strings using locale
*
* RETURNS
*
* success: CSTR_LESS_THAN, CSTR_EQUAL, CSTR_GREATER_THAN
* failure: 0
*
* NOTES
*
* Defaults to a word sort, but uses a string sort if
* SORT_STRINGSORT is set.
* Calls SetLastError for ERROR_INVALID_FLAGS, ERROR_INVALID_PARAMETER.
*
* BUGS
*
* This implementation ignores the locale
*
* FIXME
*
* Quite inefficient.
*/
int WINAPI CompareStringA(
LCID lcid, /* [in] locale ID */
DWORD fdwStyle, /* [in] comparison-style options */
LPCSTR s1, /* [in] first string */
int l1, /* [in] length of first string */
LPCSTR s2, /* [in] second string */
int l2) /* [in] length of second string */
{
int mapstring_flags;
int len1,len2;
int result;
LPSTR sk1,sk2;
TRACE("%s and %s\n",
debugstr_an (s1,l1), debugstr_an (s2,l2));
if ( (s1==NULL) || (s2==NULL) )
{
ERR("(s1=%s,s2=%s): Invalid NULL string\n",
debugstr_an(s1,l1), debugstr_an(s2,l2));
SetLastError(ERROR_INVALID_PARAMETER);
return 0;
}
if(fdwStyle & NORM_IGNORESYMBOLS)
FIXME("IGNORESYMBOLS not supported\n");
if (l1 == -1) l1 = strlen(s1);
if (l2 == -1) l2 = strlen(s2);
mapstring_flags = LCMAP_SORTKEY | fdwStyle ;
len1 = OLE2NLS_EstimateMappingLength(lcid, mapstring_flags, s1, l1);
len2 = OLE2NLS_EstimateMappingLength(lcid, mapstring_flags, s2, l2);
if ((len1==0)||(len2==0))
return 0; /* something wrong happened */
sk1 = (LPSTR)HeapAlloc(GetProcessHeap(), 0, len1 + len2);
sk2 = sk1 + len1;
if ( (!LCMapStringA(lcid,mapstring_flags,s1,l1,sk1,len1))
|| (!LCMapStringA(lcid,mapstring_flags,s2,l2,sk2,len2)) )
{
ERR("Bug in LCmapStringA.\n");
result = 0;
}
else
{
/* strcmp doesn't necessarily return -1, 0, or 1 */
result = strcmp(sk1,sk2);
}
HeapFree(GetProcessHeap(),0,sk1);
if (result < 0)
return 1;
if (result == 0)
return 2;
/* must be greater, if we reach this point */
return 3;
}
/******************************************************************************
* CompareStringW [KERNEL32.@]
* This implementation ignores the locale
* FIXME : Does only string sort. Should
* be reimplemented the same way as CompareStringA.
*/
int WINAPI CompareStringW(LCID lcid, DWORD fdwStyle,
LPCWSTR s1, int l1, LPCWSTR s2, int l2)
{
int len,ret;
if(fdwStyle & NORM_IGNORENONSPACE)
FIXME("IGNORENONSPACE not supported\n");
if(fdwStyle & NORM_IGNORESYMBOLS)
FIXME("IGNORESYMBOLS not supported\n");
if(s1==NULL || s2==NULL)
{
SetLastError(ERROR_INVALID_PARAMETER);
return 0;
}
/* Is strcmp defaulting to string sort or to word sort?? */
/* FIXME: Handle NORM_STRINGSORT */
l1 = (l1==-1)?strlenW(s1):l1;
l2 = (l2==-1)?strlenW(s2):l2;
len = l1<l2 ? l1:l2;
ret = (fdwStyle & NORM_IGNORECASE) ? strncmpiW(s1,s2,len) : strncmpW(s1,s2,len);
/* not equal, return 1 or 3 */
if(ret!=0) {
/* need to translate result */
return ((int)ret < 0) ? 1 : 3;
}
/* same len, return 2 */
if(l1==l2) return 2;
/* the longer one is lexically greater */
return (l1<l2)? 1 : 3;
}
/***********************************************************************
* lstrcmp (KERNEL32.@)
* lstrcmpA (KERNEL32.@)
*/
INT WINAPI lstrcmpA( LPCSTR str1, LPCSTR str2 )
{
return CompareStringA(LOCALE_SYSTEM_DEFAULT,0,str1,-1,str2,-1) - 2 ;
}
/***********************************************************************
* lstrcmpW (KERNEL32.@)
* FIXME : should call CompareStringW, when it is implemented.
* This implementation is not "word sort", as it should.
*/
INT WINAPI lstrcmpW( LPCWSTR str1, LPCWSTR str2 )
{
TRACE("%s and %s\n",
debugstr_w (str1), debugstr_w (str2));
if (!str1 || !str2) {
SetLastError(ERROR_INVALID_PARAMETER);
return 0;
}
while (*str1 && (*str1 == *str2)) { str1++; str2++; }
return (INT)(*str1 - *str2);
}
/***********************************************************************
* lstrcmpi (KERNEL32.@)
* lstrcmpiA (KERNEL32.@)
*/
INT WINAPI lstrcmpiA( LPCSTR str1, LPCSTR str2 )
{ TRACE("strcmpi %s and %s\n",
debugstr_a (str1), debugstr_a (str2));
return CompareStringA(LOCALE_SYSTEM_DEFAULT,NORM_IGNORECASE,str1,-1,str2,-1)-2;
}
/***********************************************************************
* lstrcmpiW (KERNEL32.@)
*/
INT WINAPI lstrcmpiW( LPCWSTR str1, LPCWSTR str2 )
{
if (!str1 || !str2) {
SetLastError(ERROR_INVALID_PARAMETER);
return 0;
}
return strcmpiW( str1, str2 );
}