Mercurial > hg > ucx / file revision

 /*

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

  *

  * Copyright 2013 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)

 #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;

 /**

  * 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);

 /**

  * 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 sstrnlen(size_t count, sstr_t string, ...);

 /**

  * Concatenates strings.

  * 

  * At least one string must be specified and there must be enough memory

  * available referenced by the destination sstr_t.ptr for this function to

  * successfully concatenate all specified strings.

  * 

  * The sstr_t.length of the destination string specifies the capacity and

  * should match the total memory available referenced by the destination

  * sstr_t.ptr. This function <i>never</i> copies data beyond the capacity and

  * does not modify any of the source strings.

  * 

  * <b>Attention:</b>

  * <ul>

  *   <li>Any content in the destination string will be overwritten</li>

  *   <li>The destination sstr_t.ptr is <b>NOT</b>

  *       <code>NULL</code>-terminated</li>

  *   <li>The destination sstr_t.length is set to the total length of the

  *       concatenated strings</li>

  *   <li><i>Hint:</i> get a <code>NULL</code>-terminated string by performing

  *       <code>mystring.ptr[mystring.length]='\0'</code> after calling this

  *       function</li>

  * </ul>

  *

  * @param dest    new sstr_t with capacity information and allocated memory

  * @param count   the total number of strings to concatenate

  * @param src     the first string

  * @param ...     all other strings

  * @return the argument for <code>dest</code> is returned

  */

 sstr_t sstrncat(sstr_t dest, size_t count, sstr_t src, ...);

 /**

  * 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);

 /**

  * 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 least location of <code>chr</code>

  * 

  * @see sstrsubs()

  */

 sstr_t sstrchr(sstr_t string, int chr);

 /**

  * 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 list, 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 list items, otherwise</li>

  * </ul>

  * 

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

  * list will be an empty string.

  * 

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

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

  * 

  * <b>Attention:</b> All list items <b>AND</b> all sstr_t.ptr of the list

  * items must be manually passed to <code>free()</code>. Use sstrsplita() 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 list (0 for an

  *               unbounded list), OUT: the actual size of the list

  * @return a list of the split strings as sstr_t array or

  *         <code>NULL</code> on error

  * 

  * @see sstrsplita()

  */

 sstr_t* sstrsplit(sstr_t string, sstr_t delim, size_t *count);

 /**

  * Performing sstrsplit() using an UcxAllocator.

  * 

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

  * 

  * The memory for the sstr_t.ptr pointers of the list 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 string the string to split

  * @param delim  the delimiter string

  * @param count  IN: the maximum size of the resulting list (0 for an

  *               unbounded list), OUT: the actual size of the list

  * @param allocator the UcxAllocator used for allocating memory

  * @return a list of the split strings as sstr_t array or

  *         <code>NULL</code> on error

  * 

  * @see sstrsplit()

  */

 sstr_t* sstrsplita(sstr_t string, sstr_t delim, size_t *count,

         UcxAllocator *allocator);

 /**

  * 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 sstrcmp(sstr_t s1, sstr_t 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 sstrdupa()

  */

 sstr_t sstrdup(sstr_t string);

 /**

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

  * @param string the string to duplicate

  * @return a duplicate of the string

  * @see sstrdup()

  */

 sstr_t sstrdupa(UcxAllocator *allocator, sstr_t string);

 /**

  * 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);

 #ifdef	__cplusplus

 }

 #endif

 #endif	/* UCX_STRING_H */