README.md

FAW-encoding

A flag aligned word-based encoding algorithm to compress sorted positive integers. FAW guarantees the upper space complexity O(N). Since FAW-encoded value never exceeds its original position in an array, FAW allows to use the same array to store encoded data. However, most of the existing integers encoding algorithms, such as, WAH, doesn't guarantee the upper space complexity O(N), encoded data cann't be stored in the same array.

Usages

#include "faw_encoding.h"

...

...

unsigned int list2_size = 500;

unsigned int list2[] = {42, 56, 71, 121, 138, 166, 177, 194, 246, ......, 29840, 29894, 29914, 29942, 29958};

Encode using FAW

list2 = encode(list2, &list2_size);

Show items from encoded data

show_items(list2, list2_size);

Search an item in FAW-encoded data

unsigned int needle = 1232;

short result = exists(list2, list2_size, needle); /** if result==1, needle exists **/

Intersect: Find common values in two FAW-encoded data

int result_size = 0; /** store the result size **/

unsigned int* results = intersect(list1, faw_encoded_list1_size, list2, faw_encoded_list2_size, &result_size);

Union: Merged values from two FAW-encoded data

int result_size = 0; /** store the result size **/

unsigned int* results = merge(list1, faw_encoded_list1_size, list2, faw_encoded_list2_size, &result_size);

Decode

list2 = decode(list2, &list2_size);