1

/* $Id$ */

     2

     3

/** @file string.cpp */

     4

     5

#include "stdafx.h"

     6

#include "openttd.h"

     7

#include "functions.h"

     8

#include "string.h"

     9

#include "macros.h"

    10

#include "table/control_codes.h"

    11

#include "helpers.hpp"

    12

    13

#include <stdarg.h>

    14

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

    15

    16

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

    17

{

    18

	assert(size > 0);

    19

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

    20

	assert(size > 0);

    21

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

    22

	*dst = '\0';

    23

}

    24

    25

    26

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

    27

{

    28

	assert(size > 0);

    29

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

    30

	*dst = '\0';

    31

}

    32

    33

    34

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

    35

{

    36

	assert(dst <= last);

    37

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

    38

		if (dst == last) return dst;

    39

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

    40

	*dst = '\0';

    41

	return strecpy(dst, src, last);

    42

}

    43

    44

    45

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

    46

{

    47

	assert(dst <= last);

    48

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

    49

	*dst = '\0';

    50

#if 1

    51

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

    52

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

    53

	}

    54

#endif

    55

	return dst;

    56

}

    57

    58

    59

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

    60

{

    61

	char buf[4096];

    62

	va_list va;

    63

	int len;

    64

    65

	va_start(va, str);

    66

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

    67

	va_end(va);

    68

	char* p = MallocT<char>(len + 1);

    69

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

    70

	return p;

    71

}

    72

    73

    74

void str_validate(char *str)

    75

{

    76

	char *dst = str;

    77

	WChar c;

    78

	size_t len;

    79

    80

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

    81

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

    82

			IsValidChar(c - SCC_SPRITE_START, CS_ALPHANUMERAL))) {

    83

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

    84

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

    85

			 * moving the pointers ahead by len */

    86

			do {

    87

				*dst++ = *str++;

    88

			} while (--len != 0);

    89

		} else {

    90

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

    91

			str += len;

    92

			*dst++ = '?';

    93

		}

    94

	}

    95

    96

	*dst = '\0';

    97

}

    98

    99

   100

void str_strip_colours(char *str)

   101

{

   102

	char *dst = str;

   103

	WChar c;

   104

	size_t len;

   105

   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

/** Convert a given ASCII string to lowercase.

   123

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

   124

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

   125

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

   126

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

   127

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

   128

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

   129

 * @param str string to convert */

   130

void strtolower(char *str)

   131

{

   132

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

   133

}

   134

   135

/**

   136

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

   137

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

   138

 * @param key character to be checked

   139

 * @param afilter the filter to use

   140

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

   141

 */

   142

bool IsValidChar(WChar key, CharSetFilter afilter)

   143

{

   144

	switch (afilter) {

   145

		case CS_ALPHANUMERAL: return IsPrintable(key);

   146

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

   147

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

   148

	}

   149

   150

	return false;

   151

}

   152

   153

#ifdef WIN32

   154

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

   155

{

   156

	va_list ap;

   157

	int ret;

   158

   159

	va_start(ap, format);

   160

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

   161

	va_end(ap);

   162

	return ret;

   163

}

   164

   165

#ifdef _MSC_VER

   166

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

   167

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

   168

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

   169

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

   170

 */

   171

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

   172

{

   173

	int ret;

   174

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

   175

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

   176

	return ret;

   177

}

   178

#endif /* _MSC_VER */

   179

   180

#endif /* WIN32 */

   181

   182

   183

/** Convert the md5sum to a hexadecimal string representation

   184

 * @param buf buffer to put the md5sum into

   185

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

   186

 * @param md5sum the md5sum itself

   187

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

   188

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

   189

{

   190

	char *p = buf;

   191

   192

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

   193

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

   194

		if (p >= last) break;

   195

	}

   196

   197

	return p;

   198

}

   199

   200

   201

/* UTF-8 handling routines */

   202

   203

   204

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

   205

 * @param c Buffer to place decoded character.

   206

 * @param s Character stream to retrieve character from.

   207

 * @return Number of characters in the sequence.

   208

 */

   209

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

   210

{

   211

	assert(c != NULL);

   212

   213

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

   214

		/* Single byte character: 0xxxxxxx */

   215

		*c = s[0];

   216

		return 1;

   217

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

   218

		if (IsUtf8Part(s[1])) {

   219

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

   220

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

   221

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

   222

		}

   223

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

   224

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

   225

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

   226

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

   227

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

   228

		}

   229

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

   230

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

   231

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

   232

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

   233

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

   234

		}

   235

	}

   236

   237

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

   238

	*c = '?';

   239

	return 1;

   240

}

   241

   242

   243

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

   244

 * @param buf Buffer to place character.

   245

 * @param c   Unicode character to encode.

   246

 * @return Number of characters in the encoded sequence.

   247

 */

   248

size_t Utf8Encode(char *buf, WChar c)

   249

{

   250

	if (c < 0x80) {

   251

		*buf = c;

   252

		return 1;

   253

	} else if (c < 0x800) {

   254

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

   255

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

   256

		return 2;

   257

	} else if (c < 0x10000) {

   258

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

   259

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

   260

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

   261

		return 3;

   262

	} else if (c < 0x110000) {

   263

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

   264

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

   265

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

   266

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

   267

		return 4;

   268

	}

   269

   270

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

   271

	*buf = '?';

   272

	return 1;

   273

}

   274

   275

/**

   276

 * Properly terminate an UTF8 string to some maximum length

   277

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

   278

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

   279

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

   280

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

   281

 */

   282

size_t Utf8TrimString(char *s, size_t maxlen)

   283

{

   284

	size_t length = 0;

   285

   286

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

   287

		size_t len = Utf8EncodedCharLen(*s);

   288

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

   289

		if (len == 0) len = 1;

   290

   291

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

   292

		 * the last UTF8 sequence is invalid */

   293

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

   294

		s += len;

   295

		length += len;

   296

	}

   297

   298

	*s = '\0';

   299

	return length;

   300

}