string-list.h

   1 #ifndef STRING_LIST_H
   2 #define STRING_LIST_H
   3
   4 /**
   5  * The string_list API offers a data structure and functions to handle
   6  * sorted and unsorted arrays of strings.  A "sorted" list is one whose
   7  * entries are sorted by string value in `strcmp()` order.
   8  *
   9  * The caller:
  10  *
  11  * . Allocates and clears a `struct string_list` variable.
  12  *
  13  * . Initializes the members. You might want to set the flag `strdup_strings`
  14  *   if the strings should be strdup()ed. For example, this is necessary
  15  *   when you add something like git_path("..."), since that function returns
  16  *   a static buffer that will change with the next call to git_path().
  17  *
  18  * If you need something advanced, you can manually malloc() the `items`
  19  * member (you need this if you add things later) and you should set the
  20  * `nr` and `alloc` members in that case, too.
  21  *
  22  * . Adds new items to the list, using `string_list_append`,
  23  *   `string_list_append_nodup`, `string_list_insert`,
  24  *   `string_list_split`, and/or `string_list_split_in_place`.
  25  *
  26  * . Can check if a string is in the list using `string_list_has_string` or
  27  *   `unsorted_string_list_has_string` and get it from the list using
  28  *   `string_list_lookup` for sorted lists.
  29  *
  30  * . Can sort an unsorted list using `string_list_sort`.
  31  *
  32  * . Can remove duplicate items from a sorted list using
  33  *   `string_list_remove_duplicates`.
  34  *
  35  * . Can remove individual items of an unsorted list using
  36  *   `unsorted_string_list_delete_item`.
  37  *
  38  * . Can remove items not matching a criterion from a sorted or unsorted
  39  *   list using `filter_string_list`, or remove empty strings using
  40  *   `string_list_remove_empty_items`.
  41  *
  42  * . Finally it should free the list using `string_list_clear`.
  43  *
  44  * Example:
  45  *
  46  *     struct string_list list = STRING_LIST_INIT_NODUP;
  47  *     int i;
  48  *
  49  *     string_list_append(&list, "foo");
  50  *     string_list_append(&list, "bar");
  51  *     for (i = 0; i < list.nr; i++)
  52  *             printf("%s\n", list.items[i].string)
  53  *
  54  * NOTE: It is more efficient to build an unsorted list and sort it
  55  * afterwards, instead of building a sorted list (`O(n log n)` instead of
  56  * `O(n^2)`).
  57  *
  58  * However, if you use the list to check if a certain string was added
  59  * already, you should not do that (using unsorted_string_list_has_string()),
  60  * because the complexity would be quadratic again (but with a worse factor).
  61  */
  62
  63 /**
  64  * Represents an item of the list. The `string` member is a pointer to the
  65  * string, and you may use the `util` member for any purpose, if you want.
  66  */
  67 struct string_list_item {
  68         char *string;
  69         void *util;
  70 };
  71
  72 typedef int (*compare_strings_fn)(const char *, const char *);
  73
  74 /**
  75  * Represents the list itself.
  76  *
  77  * . The array of items are available via the `items` member.
  78  * . The `nr` member contains the number of items stored in the list.
  79  * . The `alloc` member is used to avoid reallocating at every insertion.
  80  *   You should not tamper with it.
  81  * . Setting the `strdup_strings` member to 1 will strdup() the strings
  82  *   before adding them, see above.
  83  * . The `compare_strings_fn` member is used to specify a custom compare
  84  *   function, otherwise `strcmp()` is used as the default function.
  85  */
  86 struct string_list {
  87         struct string_list_item *items;
  88         unsigned int nr, alloc;
  89         unsigned int strdup_strings:1;
  90         compare_strings_fn cmp; /* NULL uses strcmp() */
  91 };
  92
  93 #define STRING_LIST_INIT_NODUP { NULL, 0, 0, 0, NULL }
  94 #define STRING_LIST_INIT_DUP   { NULL, 0, 0, 1, NULL }
  95
  96 /* General functions which work with both sorted and unsorted lists. */
  97
  98 /**
  99  * Initialize the members of the string_list, set `strdup_strings`
 100  * member according to the value of the second parameter.
 101  */
 102 void string_list_init(struct string_list *list, int strdup_strings);
 103
 104 /** Callback function type for for_each_string_list */
 105 typedef int (*string_list_each_func_t)(struct string_list_item *, void *);
 106
 107 /**
 108  * Apply `want` to each item in `list`, retaining only the ones for which
 109  * the function returns true.  If `free_util` is true, call free() on
 110  * the util members of any items that have to be deleted.  Preserve
 111  * the order of the items that are retained.
 112  */
 113 void filter_string_list(struct string_list *list, int free_util,
 114                         string_list_each_func_t want, void *cb_data);
 115
 116 /**
 117  * Free a string_list. The `string` pointer of the items will be freed
 118  * in case the `strdup_strings` member of the string_list is set. The
 119  * second parameter controls if the `util` pointer of the items should
 120  * be freed or not.
 121  */
 122 void string_list_clear(struct string_list *list, int free_util);
 123
 124 /**
 125  * Callback type for `string_list_clear_func`.  The string associated
 126  * with the util pointer is passed as the second argument
 127  */
 128 typedef void (*string_list_clear_func_t)(void *p, const char *str);
 129
 130 /** Call a custom clear function on each util pointer */
 131 void string_list_clear_func(struct string_list *list, string_list_clear_func_t clearfunc);
 132
 133 /**
 134  * Apply `func` to each item. If `func` returns nonzero, the
 135  * iteration aborts and the return value is propagated.
 136  */
 137 int for_each_string_list(struct string_list *list,
 138                          string_list_each_func_t func, void *cb_data);
 139
 140 /** Iterate over each item, as a macro. */
 141 #define for_each_string_list_item(item,list)            \
 142         for (item = (list)->items;                      \
 143              item && item < (list)->items + (list)->nr; \
 144              ++item)
 145
 146 /**
 147  * Remove any empty strings from the list.  If free_util is true, call
 148  * free() on the util members of any items that have to be deleted.
 149  * Preserve the order of the items that are retained.
 150  */
 151 void string_list_remove_empty_items(struct string_list *list, int free_util);
 152
 153 /* Use these functions only on sorted lists: */
 154
 155 /** Determine if the string_list has a given string or not. */
 156 int string_list_has_string(const struct string_list *list, const char *string);
 157 int string_list_find_insert_index(const struct string_list *list, const char *string,
 158                                   int negative_existing_index);
 159
 160 /**
 161  * Insert a new element to the string_list. The returned pointer can
 162  * be handy if you want to write something to the `util` pointer of
 163  * the string_list_item containing the just added string. If the given
 164  * string already exists the insertion will be skipped and the pointer
 165  * to the existing item returned.
 166  *
 167  * Since this function uses xrealloc() (which die()s if it fails) if the
 168  * list needs to grow, it is safe not to check the pointer. I.e. you may
 169  * write `string_list_insert(...)->util = ...;`.
 170  */
 171 struct string_list_item *string_list_insert(struct string_list *list, const char *string);
 172
 173 /**
 174  * Remove the given string from the sorted list.  If the string
 175  * doesn't exist, the list is not altered.
 176  */
 177 void string_list_remove(struct string_list *list, const char *string,
 178                         int free_util);
 179
 180 /**
 181  * Check if the given string is part of a sorted list. If it is part of the list,
 182  * return the corresponding string_list_item, NULL otherwise.
 183  */
 184 struct string_list_item *string_list_lookup(struct string_list *list, const char *string);
 185
 186 /*
 187  * Remove all but the first of consecutive entries with the same
 188  * string value.  If free_util is true, call free() on the util
 189  * members of any items that have to be deleted.
 190  */
 191 void string_list_remove_duplicates(struct string_list *sorted_list, int free_util);
 192
 193
 194 /* Use these functions only on unsorted lists: */
 195
 196 /**
 197  * Add string to the end of list.  If list->strdup_string is set, then
 198  * string is copied; otherwise the new string_list_entry refers to the
 199  * input string.
 200  */
 201 struct string_list_item *string_list_append(struct string_list *list, const char *string);
 202
 203 /**
 204  * Like string_list_append(), except string is never copied.  When
 205  * list->strdup_strings is set, this function can be used to hand
 206  * ownership of a malloc()ed string to list without making an extra
 207  * copy.
 208  */
 209 struct string_list_item *string_list_append_nodup(struct string_list *list, char *string);
 210
 211 /**
 212  * Sort the list's entries by string value in `strcmp()` order.
 213  */
 214 void string_list_sort(struct string_list *list);
 215
 216 /**
 217  * Like `string_list_has_string()` but for unsorted lists. Linear in
 218  * size of the list.
 219  */
 220 int unsorted_string_list_has_string(struct string_list *list, const char *string);
 221
 222 /**
 223  * Like `string_list_lookup()` but for unsorted lists. Linear in size
 224  * of the list.
 225  */
 226 struct string_list_item *unsorted_string_list_lookup(struct string_list *list,
 227                                                      const char *string);
 228 /**
 229  * Remove an item from a string_list. The `string` pointer of the
 230  * items will be freed in case the `strdup_strings` member of the
 231  * string_list is set. The third parameter controls if the `util`
 232  * pointer of the items should be freed or not.
 233  */
 234 void unsorted_string_list_delete_item(struct string_list *list, int i, int free_util);
 235
 236 /**
 237  * Split string into substrings on character `delim` and append the
 238  * substrings to `list`.  The input string is not modified.
 239  * list->strdup_strings must be set, as new memory needs to be
 240  * allocated to hold the substrings.  If maxsplit is non-negative,
 241  * then split at most maxsplit times.  Return the number of substrings
 242  * appended to list.
 243  *
 244  * Examples:
 245  *   string_list_split(l, "foo:bar:baz", ':', -1) -> ["foo", "bar", "baz"]
 246  *   string_list_split(l, "foo:bar:baz", ':', 0) -> ["foo:bar:baz"]
 247  *   string_list_split(l, "foo:bar:baz", ':', 1) -> ["foo", "bar:baz"]
 248  *   string_list_split(l, "foo:bar:", ':', -1) -> ["foo", "bar", ""]
 249  *   string_list_split(l, "", ':', -1) -> [""]
 250  *   string_list_split(l, ":", ':', -1) -> ["", ""]
 251  */
 252 int string_list_split(struct string_list *list, const char *string,
 253                       int delim, int maxsplit);
 254
 255 /*
 256  * Like string_list_split(), except that string is split in-place: the
 257  * delimiter characters in string are overwritten with NULs, and the
 258  * new string_list_items point into string (which therefore must not
 259  * be modified or freed while the string_list is in use).
 260  * list->strdup_strings must *not* be set.
 261  */
 262 int string_list_split_in_place(struct string_list *list, char *string,
 263                                int delim, int maxsplit);
 264 #endif /* STRING_LIST_H */