Mercurial > hg > ucx / file revision

 /*

  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS HEADER.

  *

  * Copyright 2017 Mike Becker, Olaf Wintermann All rights reserved.

  *

  * Redistribution and use in source and binary forms, with or without

  * modification, are permitted provided that the following conditions are met:

  *

  *   1. Redistributions of source code must retain the above copyright

  *      notice, this list of conditions and the following disclaimer.

  *

  *   2. Redistributions in binary form must reproduce the above copyright

  *      notice, this list of conditions and the following disclaimer in the

  *      documentation and/or other materials provided with the distribution.

  *

  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"

  * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE

  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE

  * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE

  * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR

  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF

  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS

  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN

  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)

  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE

  * POSSIBILITY OF SUCH DAMAGE.

  */

 /**

  * Bounded string implementation.

  * 

  * The UCX strings (<code>sstr_t</code>) provide an alternative to C strings.

  * The main difference to C strings is, that <code>sstr_t</code> does <b>not

  * need to be <code>NULL</code>-terminated</b>. Instead the length is stored

  * within the structure.

  * 

  * When using <code>sstr_t</code>, developers must be full aware of what type

  * of string (<code>NULL</code>-terminated) or not) they are using, when 

  * accessing the <code>char* ptr</code> directly.

  * 

  * The UCX string module provides some common string functions, known from

  * standard libc, working with <code>sstr_t</code>.

  * 

  * @file   string.h

  * @author Mike Becker

  * @author Olaf Wintermann

  */

 #ifndef UCX_STRING_H

 #define	UCX_STRING_H

 #include "ucx.h"

 #include "allocator.h"

 #include <stddef.h>

 /** Shortcut for a <code>sstr_t struct</code> literal. */

 #define ST(s) { (char*)s, sizeof(s)-1 }

 /** Shortcut for the conversion of a C string to a <code>sstr_t</code>. */

 #define S(s) sstrn((char*)s, sizeof(s)-1)

 /** Expands a sstr_t to printf arguments. */

 #define SFMT(s) (int) (s).length, (s).ptr

 /** Format specifier for a sstr_t. */

 #define PRIsstr ".*s"

 #ifdef	__cplusplus

 extern "C" {

 #endif

 /**

  * The UCX string structure.

  */

 typedef struct {

    /** A reference to the string (<b>not necessarily  <code>NULL</code>

     * -terminated</b>) */

     char   *ptr;

     /** The length of the string */

     size_t length;

 } sstr_t;

 typedef struct {

     const char *ptr;

     size_t     length;

 } scstr_t;

 #ifdef	__cplusplus

 }

 #endif

 #ifdef __cplusplus

 inline scstr_t s2scstr(sstr_t s) {

     scstr_t c;

     c.ptr = s.ptr;

     c.length = s.ptr;

     return c;

 }

 inline scstr_t s2scstr(scstr_t c) {

     return c;

 }

 #define SCSTR s2scstr

 #else

 scstr_t ucx_sc2sc(scstr_t c);

 scstr_t ucx_ss2sc(sstr_t str);

 #if __STDC_VERSION__ >= 201112L

 #define SCSTR(str) _Generic(str, sstr_t: ucx_ss2sc, scstr_t: ucx_sc2sc)(str)

 #elif defined(__GNUC__) || defined(__clang__)

 #define SCSTR(str) __builtin_choose_expr( \

         __builtin_types_compatible_p(typeof(str), sstr_t), \

         ucx_ss2sc, \

         ucx_sc2sc)(str)

 #elif defined(__sun)

 #define SCSTR(str) ({typeof(str) ucx_tmp_var_str = str; \

 	scstr_t ucx_tmp_var_c; \

 	ucx_tmp_var_c.ptr = ucx_tmp_var_str.ptr;\

 	ucx_tmp_var_c.length = ucx_tmp_var_str.length;\

 	ucx_tmp_var_c; })

 #else

 scstr_t ucx_ss2c_s();

 #define SCSTR ucx_ss2c_s

 #endif /* C11 feature test */

 #endif /* C++ */

 #ifdef	__cplusplus

 extern "C" {

 #endif

 /**

  * Creates a new sstr_t based on a C string.

  * 

  * The length is implicitly inferred by using a call to <code>strlen()</code>.

  *

  * <b>Note:</b> the sstr_t will hold a <i>reference</i> to the C string. If you

  * do want a copy, use sstrdup() on the return value of this function.

  * 

  * @param cstring the C string to wrap

  * @return a new sstr_t containing the C string

  * 

  * @see sstrn()

  */

 sstr_t sstr(char *cstring);

 /**

  * Creates a new sstr_t of the specified length based on a C string.

  *

  * <b>Note:</b> the sstr_t will hold a <i>reference</i> to the C string. If you

  * do want a copy, use sstrdup() on the return value of this function.

  * 

  * @param cstring  the C string to wrap

  * @param length   the length of the string

  * @return a new sstr_t containing the C string

  * 

  * @see sstr()

  * @see S()

  */

 sstr_t sstrn(char *cstring, size_t length);

 scstr_t scstr(const char *cstring);

 scstr_t scstrn(const char *cstring, size_t length);

 /**

  * Returns the cumulated length of all specified strings.

  *

  * At least one string must be specified.

  * 

  * <b>Attention:</b> if the count argument does not match the count of the

  * specified strings, the behavior is undefined.

  *

  * @param count    the total number of specified strings (so at least 1)

  * @param string   the first string

  * @param ...      all other strings

  * @return the cumulated length of all strings

  */

 size_t ucx_strnlen(size_t count, ...);

 #define sstrnlen(count, ...) ucx_strnlen(count, __VA_ARGS__)

 /**

  * Concatenates two or more strings.

  * 

  * The resulting string will be allocated by standard <code>malloc()</code>. 

  * So developers <b>MUST</b> pass the sstr_t.ptr to <code>free()</code>.

  * 

  * The sstr_t.ptr of the return value will <i>always</i> be <code>NULL</code>-

  * terminated.

  *

  * @param count   the total number of strings to concatenate

  * @param s1      first string

  * @param ...     all remaining strings

  * @return the concatenated string

  */

 sstr_t ucx_strcat(size_t count, scstr_t s1, ...);

 #define sstrcat(count, s1, ...) ucx_strcat(count, SCSTR(s1), __VA_ARGS__)

 /**

  * Concatenates two or more strings using a UcxAllocator.

  * 

  * See sstrcat() for details.

  *

  * @param a       the allocator to use

  * @param count   the total number of strings to concatenate

  * @param s1      first string

  * @param ...     all remaining strings

  * @return the concatenated string

  */

 sstr_t ucx_strcat_a(UcxAllocator *a, size_t count, scstr_t s1, ...);

 #define sstrcat_a(count, s1, ...) ucx_strcat_a(count, SCSTR(s1), __VA_ARGS__)

 /**

  * Returns a substring starting at the specified location.

  * 

  * <b>Attention:</b> the new string references the same memory area as the

  * input string and will <b>NOT</b> be <code>NULL</code>-terminated.

  * Use sstrdup() to get a copy.

  * 

  * @param string input string

  * @param start  start location of the substring

  * @return a substring of <code>string</code> starting at <code>start</code>

  * 

  * @see sstrsubsl()

  * @see sstrchr()

  */

 sstr_t sstrsubs(sstr_t string, size_t start);

 /**

  * Returns a substring with a maximum length starting at the specified location.

  * 

  * <b>Attention:</b> the new string references the same memory area as the

  * input string and will <b>NOT</b> be <code>NULL</code>-terminated.

  * Use sstrdup() to get a copy.

  * 

  * @param string input string

  * @param start  start location of the substring

  * @param length the maximum length of the substring

  * @return a substring of <code>string</code> starting at <code>start</code>

  * with a maximum length of <code>length</code>

  * 

  * @see sstrsubs()

  * @see sstrchr()

  */

 sstr_t sstrsubsl(sstr_t string, size_t start, size_t length);

 scstr_t scstrsubs(scstr_t s, size_t start);

 scstr_t scstrsubsl(scstr_t string, size_t start, size_t length);

 int ucx_strchr(const char *string, size_t length, int chr, size_t *pos);

 int ucx_strrchr(const char *string, size_t length, int chr, size_t *pos);

 /**

  * Returns a substring starting at the location of the first occurrence of the

  * specified character.

  * 

  * If the string does not contain the character, an empty string is returned.

  * 

  * @param string the string where to locate the character

  * @param chr    the character to locate

  * @return       a substring starting at the first location of <code>chr</code>

  * 

  * @see sstrsubs()

  */

 sstr_t sstrchr(sstr_t string, int chr);

 /**

  * Returns a substring starting at the location of the last occurrence of the

  * specified character.

  * 

  * If the string does not contain the character, an empty string is returned.

  * 

  * @param string the string where to locate the character

  * @param chr    the character to locate

  * @return       a substring starting at the last location of <code>chr</code>

  * 

  * @see sstrsubs()

  */

 sstr_t sstrrchr(sstr_t string, int chr);

 scstr_t scstrchr(scstr_t string, int chr);

 scstr_t scstrrchr(scstr_t string, int chr);

 const char* ucx_strstr(

         const char *str,

         size_t length,

         const char *match,

         size_t matchlen,

         size_t *newlen);

 /**

  * Returns a substring starting at the location of the first occurrence of the

  * specified string.

  * 

  * If the string does not contain the other string, an empty string is returned.

  * 

  * If <code>match</code> is an empty string, the complete <code>string</code> is

  * returned.

  * 

  * @param string the string to be scanned

  * @param match  string containing the sequence of characters to match

  * @return       a substring starting at the first occurrence of

  *               <code>match</code>, or an empty string, if the sequence is not

  *               present in <code>string</code>

  */

 sstr_t ucx_sstrstr(sstr_t string, scstr_t match);

 #define sstrstr(string, match) ucx_sstrstr(string, SCSTR(match))

 scstr_t ucx_scstrstr(scstr_t string, scstr_t match);

 #define scstrstr(string, match) ucx_scstrstr(string, SCSTR(match))

 /**

  * Splits a string into parts by using a delimiter string.

  * 

  * This function will return <code>NULL</code>, if one of the following happens:

  * <ul>

  *   <li>the string length is zero</li>

  *   <li>the delimeter length is zero</li>

  *   <li>the string equals the delimeter</li>

  *   <li>memory allocation fails</li>

  * </ul>

  * 

  * The integer referenced by <code>count</code> is used as input and determines

  * the maximum size of the resulting array, i.e. the maximum count of splits to

  * perform + 1.

  * 

  * The integer referenced by <code>count</code> is also used as output and is

  * set to

  * <ul>

  *   <li>-2, on memory allocation errors</li>

  *   <li>-1, if either the string or the delimiter is an empty string</li>

  *   <li>0, if the string equals the delimiter</li>

  *   <li>1, if the string does not contain the delimiter</li>

  *   <li>the count of array items, otherwise</li>

  * </ul>

  * 

  * If the string starts with the delimiter, the first item of the resulting

  * array will be an empty string.

  * 

  * If the string ends with the delimiter and the maximum list size is not

  * exceeded, the last array item will be an empty string.

  * In case the list size would be exceeded, the last array item will be the

  * remaining string after the last split, <i>including</i> the terminating

  * delimiter.

  * 

  * <b>Attention:</b> The array pointer <b>AND</b> all sstr_t.ptr of the array

  * items must be manually passed to <code>free()</code>. Use sstrsplit_a() with

  * an allocator to managed memory, to avoid this.

  *

  * @param string the string to split

  * @param delim  the delimiter string

  * @param count  IN: the maximum size of the resulting array (0 = no limit),

  *               OUT: the actual size of the array

  * @return a sstr_t array containing the split strings or

  *         <code>NULL</code> on error

  * 

  * @see sstrsplit_a()

  */

 sstr_t* ucx_strsplit(scstr_t string, scstr_t delim, ssize_t *count);

 #define sstrsplit(s, delim, count) ucx_strsplit(SCSTR(s), SCSTR(delim), count)

 /**

  * Performing sstrsplit() using a UcxAllocator.

  * 

  * <i>Read the description of sstrsplit() for details.</i>

  * 

  * The memory for the sstr_t.ptr pointers of the array items and the memory for

  * the sstr_t array itself are allocated by using the UcxAllocator.malloc()

  * function.

  * 

  * <b>Note:</b> the allocator is not used for memory that is freed within the

  * same call of this function (locally scoped variables).

  * 

  * @param allocator the UcxAllocator used for allocating memory

  * @param string the string to split

  * @param delim  the delimiter string

  * @param count  IN: the maximum size of the resulting array (0 = no limit),

  *               OUT: the actual size of the array

  * @return a sstr_t array containing the split strings or

  *         <code>NULL</code> on error

  * 

  * @see sstrsplit()

  */

 sstr_t* ucx_strsplit_a(UcxAllocator *allocator, scstr_t string, scstr_t delim,

         ssize_t *count);

 #define sstrsplit_a(a, s, d, c) ucx_strsplit_a(a, SCSTR(s), SCSTR(d, c))

 /**

  * Compares two UCX strings with standard <code>memcmp()</code>.

  * 

  * At first it compares the sstr_t.length attribute of the two strings. The

  * <code>memcmp()</code> function is called, if and only if the lengths match.

  * 

  * @param s1 the first string

  * @param s2 the second string

  * @return -1, if the length of s1 is less than the length of s2 or 1, if the 

  * length of s1 is greater than the length of s2 or the result of

  * <code>memcmp()</code> otherwise (i.e. 0 if the strings match)

  */

 int ucx_str_cmp(scstr_t s1, scstr_t s2);

 #define sstrcmp(s1, s2) ucx_str_cmp(SCSTR(s1), SCSTR(s2))

 /**

  * Compares two UCX strings ignoring the case.

  * 

  * At first it compares the sstr_t.length attribute of the two strings. If and

  * only if the lengths match, both strings are compared char by char ignoring

  * the case.

  * 

  * @param s1 the first string

  * @param s2 the second string

  * @return -1, if the length of s1 is less than the length of s2 or 1, if the 

  * length of s1 is greater than the length of s2 or the difference between the

  * first two differing characters otherwise (i.e. 0 if the strings match and

  * no characters differ)

  */

 int ucx_str_casecmp(scstr_t s1, scstr_t s2);

 #define sstrcasecmp(s1, s2) ucx_str_casecmp(SCSTR(s1), SCSTR(s2))

 /**

  * Creates a duplicate of the specified string.

  * 

  * The new sstr_t will contain a copy allocated by standard

  * <code>malloc()</code>. So developers <b>MUST</b> pass the sstr_t.ptr to

  * <code>free()</code>.

  * 

  * The sstr_t.ptr of the return value will <i>always</i> be <code>NULL</code>-

  * terminated.

  * 

  * @param string the string to duplicate

  * @return a duplicate of the string

  * @see sstrdup_a()

  */

 sstr_t scstrdup(scstr_t string);

 #define sstrdup(s) scstrdup(SCSTR(s))

 /**

  * Creates a duplicate of the specified string using a UcxAllocator.

  * 

  * The new sstr_t will contain a copy allocated by the allocators

  * ucx_allocator_malloc function. So it is implementation depended, whether the

  * returned sstr_t.ptr pointer must be passed to the allocators

  * ucx_allocator_free function manually.

  * 

  * The sstr_t.ptr of the return value will <i>always</i> be <code>NULL</code>-

  * terminated.

  * 

  * @param allocator a valid instance of a UcxAllocator

  * @param string the string to duplicate

  * @return a duplicate of the string

  * @see sstrdup()

  */

 sstr_t scstrdup_a(UcxAllocator *allocator, scstr_t string);

 #define sstrdup_a(allocator, s) scstrdup_a(allocator, SCSTR(s))

 size_t ucx_strtrim(const char *str, size_t length, size_t *newlen);

 /**

  * Omits leading and trailing spaces.

  * 

  * This function returns a new sstr_t containing a trimmed version of the

  * specified string.

  * 

  * <b>Note:</b> the new sstr_t references the same memory, thus you

  * <b>MUST NOT</b> pass the sstr_t.ptr of the return value to

  * <code>free()</code>. It is also highly recommended to avoid assignments like

  * <code>mystr = sstrtrim(mystr);</code> as you lose the reference to the

  * source string. Assignments of this type are only permitted, if the

  * sstr_t.ptr of the source string does not need to be freed or if another

  * reference to the source string exists.

  * 

  * @param string the string that shall be trimmed

  * @return a new sstr_t containing the trimmed string

  */

 sstr_t sstrtrim(sstr_t string);

 scstr_t scstrtrim(scstr_t string);

 /**

  * Checks, if a string has a specific prefix.

  * @param string the string to check

  * @param prefix the prefix the string should have

  * @return 1, if and only if the string has the specified prefix, 0 otherwise

  */

 int ucx_strprefix(scstr_t string, scstr_t prefix);

 #define sstrprefix(string, prefix) ucx_strprefix(SCSTR(string), SCSTR(prefix))

 /**

  * Checks, if a string has a specific suffix.

  * @param string the string to check

  * @param suffix the suffix the string should have

  * @return 1, if and only if the string has the specified suffix, 0 otherwise

  */

 int ucx_strsuffix(scstr_t string, scstr_t suffix);

 #define sstrsuffix(string, prefix) ucx_strsuffix(SCSTR(string), SCSTR(prefix))

 /**

  * Returns a lower case version of a string.

  * 

  * This function creates a duplicate of the input string, first. See the

  * documentation of sstrdup() for the implications.

  * 

  * @param string the input string

  * @return the resulting lower case string

  * @see sstrdup()

  */

 sstr_t ucx_strlower(scstr_t string);

 #define sstrlower(string) ucx_strlower(SCSTR(string))

 /**

  * Returns a lower case version of a string.

  * 

  * This function creates a duplicate of the input string, first. See the

  * documentation of sstrdup_a() for the implications.

  * 

  * @param allocator the allocator used for duplicating the string

  * @param string the input string

  * @return the resulting lower case string

  * @see sstrdup_a()

  */

 sstr_t ucx_strlower_a(UcxAllocator *allocator, scstr_t string);

 #define sstrlower_a(allocator, string) ucx_strlower_a(allocator, SCSTR(string))

 /**

  * Returns a upper case version of a string.

  * 

  * This function creates a duplicate of the input string, first. See the

  * documentation of sstrdup() for the implications.

  * 

  * @param string the input string

  * @return the resulting upper case string

  * @see sstrdup()

  */

 sstr_t ucx_strupper(scstr_t string);

 #define sstrupper(string) ucx_strupper(SCSTR(string))

 /**

  * Returns a upper case version of a string.

  * 

  * This function creates a duplicate of the input string, first. See the

  * documentation of sstrdup_a() for the implications.

  * 

  * @param allocator the allocator used for duplicating the string

  * @param string the input string

  * @return the resulting upper case string

  * @see sstrdup_a()

  */

 sstr_t ucx_strupper_a(UcxAllocator *allocator, scstr_t string);

 #define sstrupper_a(allocator, string) ucx_strupper_a(allocator, string)

 #ifdef	__cplusplus

 }

 #endif

 #endif	/* UCX_STRING_H */