update fnv-1a seed value (#18)

* update fnv-1a seed value
* update README

Author: barrust

CHANGELOG.md

@@ -1,4 +1,7 @@
 ## Current Version
+* ***BACKWARD INCOMPATIBLE CHANGES***
+ * Update to the FNV_1a hash function **NOTE:** Breaks backwards compatibility with
+ previously exported blooms using the default hash!
 
 ### Version 0.1.8
 * Added in-depth testing

README.md

@@ -88,3 +88,36 @@ cms_destroy(&cms);
 
 ## Required Compile Flags
 -lm
+
+
+## Backward Compatible Hash Function
+To use the older count-min sketch (v0.1.8 or lower) that utilized the default hashing
+algorithm, then change use the following code as the hash function:
+
+``` c
+/* NOTE: The caller will free the results */
+static uint64_t* original_default_hash(unsigned int num_hashes, const char* str) {
+    uint64_t *results = (uint64_t*)calloc(num_hashes, sizeof(uint64_t));
+    char key[17] = {0}; // largest value is 7FFF,FFFF,FFFF,FFFF
+    results[0] = __fnv_1a(str);
+    for (unsigned int i = 1; i < num_hashes; ++i) {
+        sprintf(key, "%" PRIx64 "", results[i-1]);
+        results[i] = old_fnv_1a(key);
+    }
+    return results;
+}
+
+static uint64_t old_fnv_1a(const char* key) {
+    // FNV-1a hash (http://www.isthe.com/chongo/tech/comp/fnv/)
+    int i, len = strlen(key);
+    uint64_t h = 14695981039346656073ULL; // FNV_OFFSET 64 bit
+    for (i = 0; i < len; ++i){
+            h = h ^ (unsigned char) key[i];
+            h = h * 1099511628211ULL; // FNV_PRIME 64 bit
+    }
+    return h;
+}
+```
+
+If using only older count-min sketch, then you can update the // FNV_OFFSET 64 bit
+to use `14695981039346656073ULL`

src/count_min_sketch.c

@@ -402,7 +402,7 @@ static uint64_t* __default_hash(unsigned int num_hashes, const char* str) {
 static uint64_t __fnv_1a(const char* key) {
     // FNV-1a hash (http://www.isthe.com/chongo/tech/comp/fnv/)
     int i, len = strlen(key);
-    uint64_t h = 14695981039346656073ULL; // FNV_OFFSET 64 bit
+    uint64_t h = 14695981039346656037ULL; // FNV_OFFSET 64 bit
     for (i = 0; i < len; ++i){
             h = h ^ (unsigned char) key[i];
             h = h * 1099511628211ULL; // FNV_PRIME 64 bit

tests/test_cms.c

@@ -94,7 +94,7 @@ MU_TEST(test_insertions_different) {
 }
 
 MU_TEST(test_insertions_max) {
-    uint32_t too_large = (uint32_t)INT32_MAX + 5;
+    uint64_t too_large = (uint32_t)INT32_MAX + 5;
     mu_assert_int_eq(INT32_MAX, cms_add_inc(&cms, "this is a test", too_large));
     mu_assert_int_eq(too_large, cms.elements_added);
     mu_assert_int_eq(INT32_MAX, cms_add_inc(&cms, "this is a test", 2));
@@ -270,7 +270,7 @@ MU_TEST(test_cms_export) {
     cms_export(&cms, "./tests/test.cms");
     char digest[33] = {0};
     calculate_md5sum("./tests/test.cms", digest);
-    mu_assert_string_eq("61d2ea9d0cb09b7bb284e1cf1a860449", digest);
+    mu_assert_string_eq("a53b06b40aae73ae2b8fc6c9dd113781", digest);
     remove("./tests/test.cms");
 }