string_utils: revert changes and merge with upstream

src/address_dictionary.c

@@ -284,7 +284,7 @@ phrase_t search_address_dictionaries_suffix(char *str, size_t len, char *lang) {
 bool address_dictionary_init(void) {
     if (address_dict != NULL) return false;
 
-    address_dict = malloc(sizeof(address_dictionary_t));
+    address_dict = calloc(1, sizeof(address_dictionary_t));
     if (address_dict == NULL) return false;
 
     address_dict->canonical = cstring_array_new();

src/address_parser.c

@@ -17,7 +17,7 @@ static address_parser_t *parser = NULL;
 
 
 address_parser_t *address_parser_new(void) {
-    address_parser_t *parser = malloc(sizeof(address_parser_t));
+    address_parser_t *parser = calloc(1, sizeof(address_parser_t));
     return parser;
 }
 

src/averaged_perceptron.c

@@ -93,7 +93,7 @@ averaged_perceptron_t *averaged_perceptron_read(FILE *f) {
         return NULL;
     }
 
-    averaged_perceptron_t *perceptron = malloc(sizeof(averaged_perceptron_t));
+    averaged_perceptron_t *perceptron = calloc(1, sizeof(averaged_perceptron_t));
 
     if (!file_read_uint32(f, &perceptron->num_features) ||
         !file_read_uint32(f, &perceptron->num_classes) ||
@@ -216,4 +216,4 @@ void averaged_perceptron_destroy(averaged_perceptron_t *self) {
     }
 
     free(self);
-}
+}

src/averaged_perceptron_trainer.c

@@ -389,7 +389,7 @@ bool averaged_perceptron_trainer_train_example(averaged_perceptron_trainer_t *se
 }
 
 averaged_perceptron_trainer_t *averaged_perceptron_trainer_new(void) {
-    averaged_perceptron_trainer_t *self = malloc(sizeof(averaged_perceptron_trainer_t));
+    averaged_perceptron_trainer_t *self = calloc(1, sizeof(averaged_perceptron_trainer_t));
 
     if (self == NULL) return NULL;
 

src/bloom.c

@@ -56,7 +56,7 @@ int bloom_filter_add(bloom_filter_t *self, const char *key, size_t len) {
 }
 
 bloom_filter_t *bloom_filter_new(uint64_t capacity, double error) {
-    bloom_filter_t *bloom = malloc(sizeof(bloom_filter_t));
+    bloom_filter_t *bloom = calloc(1, sizeof(bloom_filter_t));
 
     if (bloom == NULL) {
         return NULL;
@@ -220,4 +220,4 @@ void bloom_filter_destroy(bloom_filter_t *self) {
     }
 
     free(self);
-}
+}

src/geodb.c

@@ -47,7 +47,7 @@ void geodb_destroy(geodb_t *self) {
 geodb_t *geodb_init(char *dir) {
     if (dir == NULL) return NULL;
 
-    geodb_t *gdb = malloc(sizeof(geodb_t));
+    geodb_t *gdb = calloc(1, sizeof(geodb_t));
 
     if (gdb == NULL) return NULL;
 

src/geodb_builder.c

@@ -338,7 +338,7 @@ void geodb_builder_destroy(geodb_builder_t *self) {
 }
 
 geodb_builder_t *geodb_builder_new(char *log_filename) {
-    geodb_builder_t *builder = malloc(sizeof(geodb_builder_t));
+    geodb_builder_t *builder = calloc(1, sizeof(geodb_builder_t));
 
     if (builder == NULL) return NULL;
 

src/graph.c

@@ -1,7 +1,7 @@
 #include "graph.h"
 
 graph_t *graph_new_dims(graph_type_t type, uint32_t m, uint32_t n, size_t nnz, bool fixed_rows) {
-    graph_t *graph = malloc(sizeof(graph_t));
+    graph_t *graph = calloc(1, sizeof(graph_t));
     graph->m = m;
     graph->fixed_rows = fixed_rows;
     graph->n = n;

src/language_classifier.c

@@ -35,7 +35,7 @@ void language_classifier_destroy(language_classifier_t *self) {
 }
 
 language_classifier_t *language_classifier_new(void) {
-    language_classifier_t *language_classifier = malloc(sizeof(language_classifier_t));
+    language_classifier_t *language_classifier = calloc(1, sizeof(language_classifier_t));
     return language_classifier;
 }
 

src/numex.c

@@ -51,7 +51,7 @@ numex_table_t *numex_table_init(void) {
     numex_table_t *numex_table = get_numex_table();
 
     if (numex_table == NULL) {
-        numex_table = malloc(sizeof(numex_table_t));
+        numex_table = calloc(1, sizeof(numex_table_t));
 
         if (numex_table == NULL) return NULL;
 

src/sparse_matrix.c

@@ -2,7 +2,7 @@
 #include "klib/ksort.h"
 
 sparse_matrix_t *sparse_matrix_new_shape(size_t m, size_t n) {
-    sparse_matrix_t *matrix = malloc(sizeof(sparse_matrix_t));
+    sparse_matrix_t *matrix = calloc(1, sizeof(sparse_matrix_t));
     if (matrix == NULL) return NULL;
     matrix->m = m;
     matrix->n = n;

src/string_utils.c

@@ -711,16 +711,17 @@ cstring_array *cstring_array_new_size(size_t size) {
 
 cstring_array *cstring_array_from_char_array(char_array *str) {
     cstring_array *array = malloc(sizeof(cstring_array));
-    if (array == NULL || str==NULL || str->n==0) return NULL;
+    if (array == NULL || str == NULL) return NULL;
 
     array->str = str;
     array->indices = uint32_array_new_size(1);
     uint32_array_push(array->indices, 0);
     char *ptr = str->a;
-    uint32_t i = 0;
-    for (i = 0; i < str->n - 1; i++, ptr++) {
-        if (*ptr == '\0') {
-            uint32_array_push(array->indices, i + 1);
+    if (str->n > 0) {
+        for (uint32_t i = 0; i < str->n - 1; i++, ptr++) {
+            if (*ptr == '\0') {
+                uint32_array_push(array->indices, i + 1);
+            }
         }
     }
     return array;

src/transliterate.c

@@ -1087,7 +1087,7 @@ transliteration_table_t *transliteration_table_init(void) {
     transliteration_table_t *trans_table = get_transliteration_table();
 
     if (trans_table == NULL) {
-        trans_table = malloc(sizeof(transliteration_table_t));
+        trans_table = calloc(1, sizeof(transliteration_table_t));
 
         trans_table->trie = trie_new();
         if (trans_table->trie == NULL) {

src/trie.c

@@ -32,7 +32,7 @@ Constructors
 */
 
 static trie_t *trie_new_empty(uint8_t *alphabet, uint32_t alphabet_size) {
-    trie_t *self = malloc(sizeof(trie_t));
+    trie_t *self = calloc(1, sizeof(trie_t));
     if (!self)
         goto exit_no_malloc;
 

src/vector.h

@@ -34,7 +34,7 @@ static inline void _aligned_free(void *p)
         name *array = malloc(sizeof(name));                                         \
         if (array == NULL) return NULL;                                             \
         array->n = array->m = 0;                                                    \
-        array->a = malloc(size * sizeof(type));                                     \
+        array->a = malloc((size > 0 ? size : 1) * sizeof(type));                    \
         if (array->a == NULL) return NULL;                                          \
         array->m = size;                                                            \
         return array;                                                               \