1

/* $Id$ */

     2

     3

/** @file string_func.h Functions related to low-level strings.

     4

 *

     5

 * @note Be aware of "dangerous" string functions; string functions that

     6

 * have behaviour that could easily cause buffer overruns and such:

     7

 * - strncpy: does not '\0' terminate when input string is longer than

     8

 *   the size of the output string. Use strecpy instead.

     9

 * - [v]snprintf: returns the length of the string as it would be written

    10

 *   when the output is large enough, so it can be more than the size of

    11

 *   the buffer and than can underflow size_t (uint-ish) which makes all

    12

 *   subsequent snprintf alikes write outside of the buffer. Use

    13

 *   [v]seprintf instead; it will return the number of bytes actually

    14

 *   added so no [v]seprintf will cause outside of bounds writes.

    15

 * - [v]sprintf: does not bounds checking: use [v]seprintf instead.

    16

 */

    17

    18

#ifndef STRING_FUNC_H

    19

#define STRING_FUNC_H

    20

    21

#include "core/bitmath_func.hpp"

    22

#include "string_type.h"

    23

    24

/**

    25

 * Appends characters from one string to another.

    26

 *

    27

 * Appends the source string to the destination string with respect of the

    28

 * terminating null-character and the maximum size of the destination

    29

 * buffer.

    30

 *

    31

 * @note usage ttd_strlcat(dst, src, lengthof(dst));

    32

 * @note lengthof() applies only to fixed size arrays

    33

 *

    34

 * @param dst The buffer containing the target string

    35

 * @param src The buffer containing the string to append

    36

 * @param size The maximum size of the destination buffer

    37

 */

    38

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

    39

    40

/**

    41

 * Copies characters from one buffer to another.

    42

 *

    43

 * Copies the source string to the destination buffer with respect of the

    44

 * terminating null-character and the maximum size of the destination

    45

 * buffer.

    46

 *

    47

 * @note usage ttd_strlcpy(dst, src, lengthof(dst));

    48

 * @note lengthof() applies only to fixed size arrays

    49

 *

    50

 * @param dst The destination buffer

    51

 * @param src The buffer containing the string to copy

    52

 * @param size The maximum size of the destination buffer

    53

 */

    54

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

    55

    56

/**

    57

 * Appends characters from one string to another.

    58

 *

    59

 * Appends the source string to the destination string with respect of the

    60

 * terminating null-character and and the last pointer to the last element

    61

 * in the destination buffer. If the last pointer is set to NULL no

    62

 * boundary check is performed.

    63

 *

    64

 * @note usage: strecat(dst, src, lastof(dst));

    65

 * @note lastof() applies only to fixed size arrays

    66

 *

    67

 * @param dst The buffer containing the target string

    68

 * @param src The buffer containing the string to append

    69

 * @param last The pointer to the last element of the destination buffer

    70

 * @return The pointer to the terminating null-character in the destination buffer

    71

 */

    72

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

    73

    74

/**

    75

 * Copies characters from one buffer to another.

    76

 *

    77

 * Copies the source string to the destination buffer with respect of the

    78

 * terminating null-character and the last pointer to the last element in

    79

 * the destination buffer. If the last pointer is set to NULL no boundary

    80

 * check is performed.

    81

 *

    82

 * @note usage: strecpy(dst, src, lastof(dst));

    83

 * @note lastof() applies only to fixed size arrays

    84

 *

    85

 * @param dst The destination buffer

    86

 * @param src The buffer containing the string to copy

    87

 * @param last The pointer to the last element of the destination buffer

    88

 * @return The pointer to the terminating null-character in the destination buffer

    89

 */

    90

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

    91

    92

int CDECL seprintf(char *str, const char *last, const char *format, ...);

    93

    94

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

    95

    96

/** Scans the string for valid characters and if it finds invalid ones,

    97

 * replaces them with a question mark '?' */

    98

void str_validate(char *str);

    99

   100

/** Scans the string for colour codes and strips them */

   101

void str_strip_colours(char *str);

   102

   103

/** Convert the given string to lowercase, only works with ASCII! */

   104

void strtolower(char *str);

   105

   106

/**

   107

 * Check if a string buffer is empty.

   108

 *

   109

 * @param s The pointer to the firste element of the buffer

   110

 * @return true if the buffer starts with the terminating null-character or

   111

 *         if the given pointer points to NULL else return false

   112

 */

   113

static inline bool StrEmpty(const char *s)

   114

{

   115

	return s == NULL || s[0] == '\0';

   116

}

   117

   118

/**

   119

 * Get the length of a string, within a limited buffer.

   120

 *

   121

 * @param str The pointer to the firste element of the buffer

   122

 * @param maxlen The maximum size of the buffer

   123

 * @return The length of the string

   124

 */

   125

static inline size_t ttd_strnlen(const char *str, size_t maxlen)

   126

{

   127

	const char *t;

   128

	for (t = str; (size_t)(t - str) < maxlen && *t != '\0'; t++) {}

   129

	return t - str;

   130

}

   131

   132

/** Convert the md5sum number to a 'hexadecimal' string, return next pos in buffer */

   133

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

   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

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

   145

size_t Utf8Encode(char *buf, WChar c);

   146

size_t Utf8TrimString(char *s, size_t maxlen);

   147

   148

   149

static inline WChar Utf8Consume(const char **s)

   150

{

   151

	WChar c;

   152

	*s += Utf8Decode(&c, *s);

   153

	return c;

   154

}

   155

   156

   157

/** Return the length of a UTF-8 encoded character.

   158

 * @param c Unicode character.

   159

 * @return Length of UTF-8 encoding for character.

   160

 */

   161

static inline int8 Utf8CharLen(WChar c)

   162

{

   163

	if (c < 0x80)       return 1;

   164

	if (c < 0x800)      return 2;

   165

	if (c < 0x10000)    return 3;

   166

	if (c < 0x110000)   return 4;

   167

   168

	/* Invalid valid, we encode as a '?' */

   169

	return 1;

   170

}

   171

   172

   173

/**

   174

 * Return the length of an UTF-8 encoded value based on a single char. This

   175

 * char should be the first byte of the UTF-8 encoding. If not, or encoding

   176

 * is invalid, return value is 0

   177

 * @param c char to query length of

   178

 * @return requested size

   179

 */

   180

static inline int8 Utf8EncodedCharLen(char c)

   181

{

   182

	if (GB(c, 3, 5) == 0x1E) return 4;

   183

	if (GB(c, 4, 4) == 0x0E) return 3;

   184

	if (GB(c, 5, 3) == 0x06) return 2;

   185

	if (GB(c, 7, 1) == 0x00) return 1;

   186

   187

	/* Invalid UTF8 start encoding */

   188

	return 0;

   189

}

   190

   191

   192

/* Check if the given character is part of a UTF8 sequence */

   193

static inline bool IsUtf8Part(char c)

   194

{

   195

	return GB(c, 6, 2) == 2;

   196

}

   197

   198

/**

   199

 * Retrieve the previous UNICODE character in an UTF-8 encoded string.

   200

 * @param s char pointer pointing to (the first char of) the next character

   201

 * @return a pointer in 's' to the previous UNICODE character's first byte

   202

 * @note The function should not be used to determine the length of the previous

   203

 * encoded char because it might be an invalid/corrupt start-sequence

   204

 */

   205

static inline char *Utf8PrevChar(const char *s)

   206

{

   207

	const char *ret = s;

   208

	while (IsUtf8Part(*--ret)) {}

   209

	return (char*)ret;

   210

}

   211

   212

   213

static inline bool IsPrintable(WChar c)

   214

{

   215

	if (c < 0x20)   return false;

   216

	if (c < 0xE000) return true;

   217

	if (c < 0xE200) return false;

   218

	return true;

   219

}

   220

   221

/**

   222

 * Check whether UNICODE character is whitespace or not, i.e. whether

   223

 * this is a potential line-break character.

   224

 * @param c UNICODE character to check

   225

 * @return a boolean value whether 'c' is a whitespace character or not

   226

 * @see http://www.fileformat.info/info/unicode/category/Zs/list.htm

   227

 */

   228

static inline bool IsWhitespace(WChar c)

   229

{

   230

	return

   231

	  c == 0x0020 /* SPACE */ ||

   232

	  c == 0x3000 /* IDEOGRAPHIC SPACE */

   233

	;

   234

}

   235

   236

#ifndef _GNU_SOURCE

   237

/* strndup is a GNU extension */

   238

char *strndup(const char *s, size_t len);

   239

#endif /* !_GNU_SOURCE */

   240

   241

#endif /* STRING_FUNC_H */