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

#include "helpers.hpp"

    10

    11

#include <stdarg.h>

    12

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

    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

    63

	va_start(va, str);

    64

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

    65

	va_end(va);

    66

	char* p = MallocT<char>(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

	char *p = buf;

   188

   189

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

   190

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

   191

		if (p >= last) break;

   192

	}

   193

   194

	return p;

   195

}

   196

   197

   198

/* UTF-8 handling routines */

   199

   200

   201

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

   202

 * @param c Buffer to place decoded character.

   203

 * @param s Character stream to retrieve character from.

   204

 * @return Number of characters in the sequence.

   205

 */

   206

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

   207

{

   208

	assert(c != NULL);

   209

   210

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

   211

		/* Single byte character: 0xxxxxxx */

   212

		*c = s[0];

   213

		return 1;

   214

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

   215

		if (IsUtf8Part(s[1])) {

   216

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

   217

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

   218

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

   219

		}

   220

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

   221

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

   222

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

   223

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

   224

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

   225

		}

   226

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

   227

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

   228

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

   229

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

   230

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

   231

		}

   232

	}

   233

   234

	//DEBUG(misc, 1, "[utf8] invalid UTF-8 sequence");

   235

	*c = '?';

   236

	return 1;

   237

}

   238

   239

   240

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

   241

 * @param buf Buffer to place character.

   242

 * @param c   Unicode character to encode.

   243

 * @return Number of characters in the encoded sequence.

   244

 */

   245

size_t Utf8Encode(char *buf, WChar c)

   246

{

   247

	if (c < 0x80) {

   248

		*buf = c;

   249

		return 1;

   250

	} else if (c < 0x800) {

   251

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

   252

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

   253

		return 2;

   254

	} else if (c < 0x10000) {

   255

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

   256

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

   257

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

   258

		return 3;

   259

	} else if (c < 0x110000) {

   260

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

   261

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

   262

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

   263

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

   264

		return 4;

   265

	}

   266

   267

	//DEBUG(misc, 1, "[utf8] can't UTF-8 encode value 0x%X", c);

   268

	*buf = '?';

   269

	return 1;

   270

}

   271

   272

/**

   273

 * Properly terminate an UTF8 string to some maximum length

   274

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

   275

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

   276

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

   277

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

   278

 */

   279

size_t Utf8TrimString(char *s, size_t maxlen)

   280

{

   281

	size_t length = 0;

   282

   283

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

   284

		size_t len = Utf8EncodedCharLen(*s);

   285

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

   286

		if (len == 0) len = 1;

   287

   288

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

   289

		 * the last UTF8 sequence is invalid */

   290

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

   291

		s += len;

   292

		length += len;

   293

	}

   294

   295

	*s = '\0';

   296

	return length;

   297

}