简单数据库实现——Part9 - 二分搜索和重复键
上一部分我们的数据仍然没有排序,这个部分我们会修复这个问题,并检测和拒绝重复的键。
现在,我们的execute_insert()
函数总是在表的最后插入。而实际上我们应该在表中找到正确的位置插入,并且如果键已经存在,应该返回一个错误。
ExecuteResult execute_insert(Statement* statement, Table* table) {
void* node = get_page(table->pager, table->root_page_num);
- if ((*leaf_node_num_cells(node) >= LEAF_NODE_MAX_CELLS)) {
+ uint32_t num_cells = (*leaf_node_num_cells(node));
+ if (num_cells >= LEAF_NODE_MAX_CELLS) {
return EXECUTE_TABLE_FULL;
}
Row* row_to_insert = &(statement->row_to_insert);
- Cursor* cursor = table_end(table);
+ uint32_t key_to_insert = row_to_insert->id;
+ Cursor* cursor = table_find(table, key_to_insert);
+
+ if (cursor->cell_num < num_cells) {
+ uint32_t key_at_index = *leaf_node_key(node, cursor->cell_num);
+ if (key_at_index == key_to_insert) {
+ return EXECUTE_DUPLICATE_KEY;
+ }
+ }
leaf_node_insert(cursor, row_to_insert->id, row_to_insert);
我们不再需要table_end()
这个函数。
-Cursor* table_end(Table* table) {
- Cursor* cursor = malloc(sizeof(Cursor));
- cursor->table = table;
- cursor->page_num = table->root_page_num;
-
- void* root_node = get_page(table->pager, table->root_page_num);
- uint32_t num_cells = *leaf_node_num_cells(root_node);
- cursor->cell_num = num_cells;
- cursor->end_of_table = true;
-
- return cursor;
-}
我们将会用一个新的方法来替换它,这个新方法以给定的键在树中搜索位置。
+/*
+Return the position of the given key.
+If the key is not present, return the position
+where it should be inserted
+*/
+Cursor* table_find(Table* table, uint32_t key) {
+ uint32_t root_page_num = table->root_page_num;
+ void* root_node = get_page(table->pager, root_page_num);
+
+ if (get_node_type(root_node) == NODE_LEAF) {
+ return leaf_node_find(table, root_page_num, key);
+ } else {
+ printf("Need to implement searching an internal node\n");
+ exit(EXIT_FAILURE);
+ }
+}
因为我们还没有实现内部节点,所以我们可以直接用二分查找来查找叶节点。
+Cursor* leaf_node_find(Table* table, uint32_t page_num, uint32_t key) {
+ void* node = get_page(table->pager, page_num);
+ uint32_t num_cells = *leaf_node_num_cells(node);
+
+ Cursor* cursor = malloc(sizeof(Cursor));
+ cursor->table = table;
+ cursor->page_num = page_num;
+
+ // Binary search
+ uint32_t min_index = 0;
+ uint32_t one_past_max_index = num_cells;
+ while (one_past_max_index != min_index) {
+ uint32_t index = (min_index + one_past_max_index) / 2;
+ uint32_t key_at_index = *leaf_node_key(node, index);
+ if (key == key_at_index) {
+ cursor->cell_num = index;
+ return cursor;
+ }
+ if (key < key_at_index) {
+ one_past_max_index = index;
+ } else {
+ min_index = index + 1;
+ }
+ }
+
+ cursor->cell_num = min_index;
+ return cursor;
+}
这个函数会返回:
- 键的位置
- 如果要插入新键,需要移动的另一个键的位置
- 超过最后一个键的位置
因为我们现在正在检查节点类型,所以我们需要函数来获取和设置节点中的值。
+NodeType get_node_type(void* node) {
+ uint8_t value = *((uint8_t*)(node + NODE_TYPE_OFFSET));
+ return (NodeType)value;
+}
+
+void set_node_type(void* node, NodeType type) {
+ uint8_t value = type;
+ *((uint8_t*)(node + NODE_TYPE_OFFSET)) = value;
+}
我们首先要转换为uint8_t
来确保序列化(serialized)成一个字节(byte)。我们还需要初始化节点类型。
-void initialize_leaf_node(void* node) { *leaf_node_num_cells(node) = 0; }
+void initialize_leaf_node(void* node) {
+ set_node_type(node, NODE_LEAF);
+ *leaf_node_num_cells(node) = 0;
+}
最后,我们需要添加一个错误码。
-enum ExecuteResult_t { EXECUTE_SUCCESS, EXECUTE_TABLE_FULL };
+enum ExecuteResult_t {
+ EXECUTE_SUCCESS,
+ EXECUTE_DUPLICATE_KEY,
+ EXECUTE_TABLE_FULL
+};
case (EXECUTE_SUCCESS):
printf("Executed.\n");
break;
+ case (EXECUTE_DUPLICATE_KEY):
+ printf("Error: Duplicate key.\n");
+ break;
case (EXECUTE_TABLE_FULL):
printf("Error: Table full.\n");
break;
下一部分,我们将会分裂(split)节点并创建内部节点。