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 <ctype.h> // required for tolower()

    12

    13

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

    14

{

    15

	assert(size > 0);

    16

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

    17

	assert(size > 0);

    18

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

    19

	*dst = '\0';

    20

}

    21

    22

    23

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

    24

{

    25

	assert(size > 0);

    26

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

    27

	*dst = '\0';

    28

}

    29

    30

    31

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

    32

{

    33

	assert(dst <= last);

    34

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

    35

		if (dst == last) return dst;

    36

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

    37

	*dst = '\0';

    38

	return strecpy(dst, src, last);

    39

}

    40

    41

    42

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

    43

{

    44

	assert(dst <= last);

    45

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

    46

	*dst = '\0';

    47

#if 1

    48

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

    49

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

    50

	}

    51

#endif

    52

	return dst;

    53

}

    54

    55

    56

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

    57

{

    58

	char buf[4096];

    59

	va_list va;

    60

	int len;

    61

	char* p;

    62

    63

	va_start(va, str);

    64

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

    65

	va_end(va);

    66

	p = malloc(len + 1);

    67

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

    68

	return p;

    69

}

    70

    71

    72

void str_validate(char *str)

    73

{

    74

	char *dst = str;

    75

	WChar c;

    76

	size_t len;

    77

    78

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

    79

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

    80

			IsValidChar(c - SCC_SPRITE_START, CS_ALPHANUMERAL))) {

    81

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

    82

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

    83

			 * moving the pointers ahead by len */

    84

			do {

    85

				*dst++ = *str++;

    86

			} while (--len != 0);

    87

		} else {

    88

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

    89

			str += len;

    90

			*dst++ = '?';

    91

		}

    92

	}

    93

    94

	*dst = '\0';

    95

}

    96

    97

    98

void str_strip_colours(char *str)

    99

{

   100

	char *dst = str;

   101

	WChar c;

   102

	size_t len;

   103

   104

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

   105

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

   106

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

   107

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

   108

			 * moving the pointers ahead by len */

   109

			do {

   110

				*dst++ = *str++;

   111

			} while (--len != 0);

   112

		} else {

   113

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

   114

			str += len;

   115

		}

   116

	}

   117

	*dst = '\0';

   118

}

   119

   120

/** Convert a given ASCII string to lowercase.

   121

 * NOTE: only support ASCII characters, no UTF8 fancy. As currently

   122

 * the function is only used to lowercase data-filenames if they are

   123

 * not found, this is sufficient. If more, or general functionality is

   124

 * needed, look to r7271 where it was removed because it was broken when

   125

 * using certain locales: eg in Turkish the uppercase 'I' was converted to

   126

 * '?', so just revert to the old functionality */

   127

void strtolower(char *str)

   128

{

   129

	for (; *str != '\0'; str++) *str = tolower(*str);

   130

}

   131

   132

/**

   133

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

   134

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

   135

 * @param key character to be checked

   136

 * @param afilter the filter to use

   137

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

   138

 */

   139

bool IsValidChar(WChar key, CharSetFilter afilter)

   140

{

   141

	switch (afilter) {

   142

		case CS_ALPHANUMERAL: return IsPrintable(key);

   143

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

   144

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

   145

	}

   146

   147

	return false;

   148

}

   149

   150

#ifdef WIN32

   151

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

   152

{

   153

	va_list ap;

   154

	int ret;

   155

   156

	va_start(ap, format);

   157

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

   158

	va_end(ap);

   159

	return ret;

   160

}

   161

   162

#ifdef _MSC_VER

   163

/* *nprintf broken, not POSIX compliant, MSDN description

   164

 * - If len < count, then len characters are stored in buffer, a null-terminator is appended, and len is returned.

   165

 * - If len = count, then len characters are stored in buffer, no null-terminator is appended, and len is returned.

   166

 * - If len > count, then count characters are stored in buffer, no null-terminator is appended, and a negative value is returned

   167

 */

   168

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

   169

{

   170

	int ret;

   171

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

   172

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

   173

	return ret;

   174

}

   175

#endif /* _MSC_VER */

   176

   177

#endif /* WIN32 */

   178

   179

   180

/** Convert the md5sum to a hexadecimal string representation

   181

 * @param buf buffer to put the md5sum into

   182

 * @param last last character of buffer (usually lastof(buf))

   183

 * @param md5sum the md5sum itself

   184

 * @return a pointer to the next character after the md5sum */

   185

char *md5sumToString(char *buf, const char *last, const uint8 md5sum[16])

   186

{

   187

	uint i;

   188

	char *p = buf;

   189

   190

	for (i = 0; i < 16; i++) {

   191

		p += snprintf(p, last + 1 - p, "%02X", md5sum[i]);

   192

		if (p >= last) break;

   193

	}

   194

   195

	return p;

   196

}

   197

   198

   199

/* UTF-8 handling routines */

   200

   201

   202

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

   203

 * @param c Buffer to place decoded character.

   204

 * @param s Character stream to retrieve character from.

   205

 * @return Number of characters in the sequence.

   206

 */

   207

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

   208

{

   209

	assert(c != NULL);

   210

   211

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

   212

		/* Single byte character: 0xxxxxxx */

   213

		*c = s[0];

   214

		return 1;

   215

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

   216

		if (IsUtf8Part(s[1])) {

   217

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

   218

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

   219

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

   220

		}

   221

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

   222

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

   223

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

   224

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

   225

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

   226

		}

   227

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

   228

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

   229

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

   230

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

   231

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

   232

		}

   233

	}

   234

   235

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

   236

	*c = '?';

   237

	return 1;

   238

}

   239

   240

   241

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

   242

 * @param buf Buffer to place character.

   243

 * @param c   Unicode character to encode.

   244

 * @return Number of characters in the encoded sequence.

   245

 */

   246

size_t Utf8Encode(char *buf, WChar c)

   247

{

   248

	if (c < 0x80) {

   249

		*buf = c;

   250

		return 1;

   251

	} else if (c < 0x800) {

   252

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

   253

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

   254

		return 2;

   255

	} else if (c < 0x10000) {

   256

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

   257

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

   258

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

   259

		return 3;

   260

	} else if (c < 0x110000) {

   261

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

   262

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

   263

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

   264

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

   265

		return 4;

   266

	}

   267

   268

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

   269

	*buf = '?';

   270

	return 1;

   271

}

   272

   273

/**

   274

 * Properly terminate an UTF8 string to some maximum length

   275

 * @param s string to check if it needs additional trimming

   276

 * @param maxlen the maximum length the buffer can have.

   277

 * @return the new length in bytes of the string (eg. strlen(new_string))

   278

 * @NOTE maxlen is the string length _INCLUDING_ the terminating '\0'

   279

 */

   280

size_t Utf8TrimString(char *s, size_t maxlen)

   281

{

   282

	size_t length = 0;

   283

	const char *ptr = strchr(s, '\0');

   284

	while (*s != '\0') {

   285

		size_t len = Utf8EncodedCharLen(*s);

   286

		/* Silently ignore invalid UTF8 sequences, our only concern trimming */

   287

		if (len == 0) len = 1;

   288

   289

		/* Take care when a hard cutoff was made for the string and

   290

		 * the last UTF8 sequence is invalid */

   291

		if (length + len >= maxlen || (s + len > ptr)) break;

   292

		s += len;

   293

		length += len;

   294

	}

   295

   296

	*s = '\0';

   297

	return length;

   298

}