1

/* $Id$ */

     2

     3

#include "stdafx.h"

     4

#include "openttd.h"

     5

#include "functions.h"

     6

#include "string.h"

     7

#include "macros.h"

     8

#include "table/control_codes.h"

     9

    10

#include <stdarg.h>

    11

#include <wctype.h> // required for towlower()

    12

#include <locale.h> // required for setlocale()

    13

    14

void ttd_strlcat(char *dst, const char *src, size_t size)

    15

{

    16

	assert(size > 0);

    17

	for (; size > 0 && *dst != '\0'; --size, ++dst) {}

    18

	assert(size > 0);

    19

	while (--size > 0 && *src != '\0') *dst++ = *src++;

    20

	*dst = '\0';

    21

}

    22

    23

    24

void ttd_strlcpy(char *dst, const char *src, size_t size)

    25

{

    26

	assert(size > 0);

    27

	while (--size > 0 && *src != '\0') *dst++ = *src++;

    28

	*dst = '\0';

    29

}

    30

    31

    32

char* strecat(char* dst, const char* src, const char* last)

    33

{

    34

	assert(dst <= last);

    35

	for (; *dst != '\0'; ++dst)

    36

		if (dst == last) return dst;

    37

	for (; *src != '\0' && dst != last; ++dst, ++src) *dst = *src;

    38

	*dst = '\0';

    39

	return strecpy(dst, src, last);

    40

}

    41

    42

    43

char* strecpy(char* dst, const char* src, const char* last)

    44

{

    45

	assert(dst <= last);

    46

	for (; *src != '\0' && dst != last; ++dst, ++src) *dst = *src;

    47

	*dst = '\0';

    48

#if 1

    49

	if (dst == last && *src != '\0') {

    50

		error("String too long for destination buffer");

    51

	}

    52

#endif

    53

	return dst;

    54

}

    55

    56

    57

char* CDECL str_fmt(const char* str, ...)

    58

{

    59

	char buf[4096];

    60

	va_list va;

    61

	int len;

    62

	char* p;

    63

    64

	va_start(va, str);

    65

	len = vsnprintf(buf, lengthof(buf), str, va);

    66

	va_end(va);

    67

	p = malloc(len + 1);

    68

	if (p != NULL) memcpy(p, buf, len + 1);

    69

	return p;

    70

}

    71

    72

    73

void str_validate(char *str)

    74

{

    75

	char *dst = str;

    76

	WChar c;

    77

	size_t len;

    78

    79

	for (len = Utf8Decode(&c, str); c != '\0'; len = Utf8Decode(&c, str)) {

    80

		if (IsPrintable(c) && (c < SCC_SPRITE_START || c > SCC_SPRITE_END ||

    81

			IsValidChar(c - SCC_SPRITE_START, CS_ALPHANUMERAL))) {

    82

			/* Copy the character back. Even if dst is current the same as str

    83

			 * (i.e. no characters have been changed) this is quicker than

    84

			 * moving the pointers ahead by len */

    85

			do {

    86

				*dst++ = *str++;

    87

			} while (--len != 0);

    88

		} else {

    89

			/* Replace the undesirable character with a question mark */

    90

			str += len;

    91

			*dst++ = '?';

    92

		}

    93

	}

    94

    95

	*dst = '\0';

    96

}

    97

    98

    99

void str_strip_colours(char *str)

   100

{

   101

	char *dst = str;

   102

	WChar c;

   103

	size_t len;

   104

   105

	strtolower(str);

   106

	for (len = Utf8Decode(&c, str); c != '\0'; len = Utf8Decode(&c, str)) {

   107

		if (c < SCC_BLUE || c > SCC_BLACK) {

   108

			/* Copy the character back. Even if dst is current the same as str

   109

			 * (i.e. no characters have been changed) this is quicker than

   110

			 * moving the pointers ahead by len */

   111

			do {

   112

				*dst++ = *str++;

   113

			} while (--len != 0);

   114

		} else {

   115

			/* Just skip (strip) the colour codes */

   116

			str += len;

   117

		}

   118

	}

   119

	*dst = '\0';

   120

}

   121

   122

   123

void strtolower(char *str)

   124

{

   125

	WChar c;

   126

	/* Convert according to native locale, needed for unicode characters

   127

	 * We backup the current locale, then set it to native "", the set back */

   128

	char *locale = strdup(setlocale(LC_CTYPE, NULL));

   129

   130

	setlocale(LC_CTYPE, "");

   131

	for (Utf8Decode(&c, str); c != '\0'; Utf8Decode(&c, str)) {

   132

		/* XXX - assume lowercase version does not use more bytes */

   133

		c = towlower(c);

   134

		str += Utf8Encode(str, c);

   135

	}

   136

	setlocale(LC_CTYPE, locale);

   137

	free(locale);

   138

}

   139

   140

   141

/**

   142

 * Only allow certain keys. You can define the filter to be used. This makes

   143

 *  sure no invalid keys can get into an editbox, like BELL.

   144

 * @param key character to be checked

   145

 * @param afilter the filter to use

   146

 * @return true or false depending if the character is printable/valid or not

   147

 */

   148

bool IsValidChar(WChar key, CharSetFilter afilter)

   149

{

   150

	switch (afilter) {

   151

		case CS_ALPHANUMERAL: return IsPrintable(key);

   152

		case CS_NUMERAL:      return (key >= '0' && key <= '9');

   153

		case CS_ALPHA:        return IsPrintable(key) && !(key >= '0' && key <= '9');

   154

	}

   155

   156

	return false;

   157

}

   158

   159

#ifdef WIN32

   160

int CDECL snprintf(char *str, size_t size, const char *format, ...)

   161

{

   162

	va_list ap;

   163

	int ret;

   164

   165

	va_start(ap, format);

   166

	ret = vsnprintf(str, size, format, ap);

   167

	va_end(ap);

   168

	return ret;

   169

}

   170

   171

#ifdef _MSC_VER

   172

int CDECL vsnprintf(char *str, size_t size, const char *format, va_list ap)

   173

{

   174

	int ret;

   175

	ret = _vsnprintf(str, size, format, ap);

   176

	if (ret < 0) str[size - 1] = '\0';

   177

	return ret;

   178

}

   179

#endif /* _MSC_VER */

   180

   181

#endif /* WIN32 */

   182

   183

   184

/* UTF-8 handling routines */

   185

   186

   187

/* Decode and consume the next UTF-8 encoded character

   188

 * @param c Buffer to place decoded character.

   189

 * @param s Character stream to retrieve character from.

   190

 * @return Number of characters in the sequence.

   191

 */

   192

size_t Utf8Decode(WChar *c, const char *s)

   193

{

   194

	assert(c != NULL);

   195

   196

	if (!HASBIT(s[0], 7)) {

   197

		/* Single byte character: 0xxxxxxx */

   198

		*c = s[0];

   199

		return 1;

   200

	} else if (GB(s[0], 5, 3) == 6) {

   201

		if (IsUtf8Part(s[1])) {

   202

			/* Double byte character: 110xxxxx 10xxxxxx */

   203

			*c = GB(s[0], 0, 5) << 6 | GB(s[1], 0, 6);

   204

			if (*c >= 0x80) return 2;

   205

		}

   206

	} else if (GB(s[0], 4, 4) == 14) {

   207

		if (IsUtf8Part(s[1]) && IsUtf8Part(s[2])) {

   208

			/* Triple byte character: 1110xxxx 10xxxxxx 10xxxxxx */

   209

			*c = GB(s[0], 0, 4) << 12 | GB(s[1], 0, 6) << 6 | GB(s[2], 0, 6);

   210

			if (*c >= 0x800) return 3;

   211

		}

   212

	} else if (GB(s[0], 3, 5) == 30) {

   213

		if (IsUtf8Part(s[1]) && IsUtf8Part(s[2]) && IsUtf8Part(s[3])) {

   214

			/* 4 byte character: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */

   215

			*c = GB(s[0], 0, 3) << 18 | GB(s[1], 0, 6) << 12 | GB(s[2], 0, 6) << 6 | GB(s[3], 0, 6);

   216

			if (*c >= 0x10000 && *c <= 0x10FFFF) return 4;

   217

		}

   218

	}

   219

   220

	//DEBUG(misc, 1) ("Invalid UTF-8 sequence");

   221

	*c = '?';

   222

	return 1;

   223

}

   224

   225

   226

/* Encode a unicode character and place it in the buffer

   227

 * @param buf Buffer to place character.

   228

 * @param c   Unicode character to encode.

   229

 * @return Number of characters in the encoded sequence.

   230

 */

   231

size_t Utf8Encode(char *buf, WChar c)

   232

{

   233

	if (c < 0x80) {

   234

		*buf = c;

   235

		return 1;

   236

	} else if (c < 0x800) {

   237

		*buf++ = 0xC0 + GB(c,  6, 5);

   238

		*buf   = 0x80 + GB(c,  0, 6);

   239

		return 2;

   240

	} else if (c < 0x10000) {

   241

		*buf++ = 0xE0 + GB(c, 12, 4);

   242

		*buf++ = 0x80 + GB(c,  6, 6);

   243

		*buf   = 0x80 + GB(c,  0, 6);

   244

		return 3;

   245

	} else if (c < 0x110000) {

   246

		*buf++ = 0xF0 + GB(c, 18, 3);

   247

		*buf++ = 0x80 + GB(c, 12, 6);

   248

		*buf++ = 0x80 + GB(c,  6, 6);

   249

		*buf   = 0x80 + GB(c,  0, 6);

   250

		return 4;

   251

	}

   252

   253

	//DEBUG(misc, 1) ("Can't UTF-8 encode value 0x%X", c);

   254

	*buf = '?';

   255

	return 1;

   256

}