mirror of
https://github.com/LadybirdBrowser/ladybird.git
synced 2024-11-26 01:20:25 +00:00
267eb3b329
This patch implements a basic hash index. It uses the extendible hashing algorith. Also includes a test file.
329 lines
6.3 KiB
C++
329 lines
6.3 KiB
C++
/*
|
|
* Copyright (c) 2021, Jan de Visser <jan@de-visser.net>
|
|
*
|
|
* SPDX-License-Identifier: BSD-2-Clause
|
|
*/
|
|
|
|
#include <LibSQL/HashIndex.h>
|
|
#include <LibSQL/Heap.h>
|
|
#include <LibSQL/Meta.h>
|
|
#include <LibSQL/Tuple.h>
|
|
#include <LibSQL/Value.h>
|
|
#include <LibTest/TestCase.h>
|
|
#include <unistd.h>
|
|
|
|
constexpr static int keys[] = {
|
|
39,
|
|
87,
|
|
77,
|
|
42,
|
|
98,
|
|
40,
|
|
53,
|
|
8,
|
|
37,
|
|
12,
|
|
90,
|
|
72,
|
|
73,
|
|
11,
|
|
88,
|
|
22,
|
|
10,
|
|
82,
|
|
25,
|
|
61,
|
|
97,
|
|
18,
|
|
60,
|
|
68,
|
|
21,
|
|
3,
|
|
58,
|
|
29,
|
|
13,
|
|
17,
|
|
89,
|
|
81,
|
|
16,
|
|
64,
|
|
5,
|
|
41,
|
|
36,
|
|
91,
|
|
38,
|
|
24,
|
|
32,
|
|
50,
|
|
34,
|
|
94,
|
|
49,
|
|
47,
|
|
1,
|
|
6,
|
|
44,
|
|
76,
|
|
};
|
|
constexpr static u32 pointers[] = {
|
|
92,
|
|
4,
|
|
50,
|
|
47,
|
|
68,
|
|
73,
|
|
24,
|
|
28,
|
|
50,
|
|
93,
|
|
60,
|
|
36,
|
|
92,
|
|
72,
|
|
53,
|
|
26,
|
|
91,
|
|
84,
|
|
25,
|
|
43,
|
|
88,
|
|
12,
|
|
62,
|
|
35,
|
|
96,
|
|
27,
|
|
96,
|
|
27,
|
|
99,
|
|
30,
|
|
21,
|
|
89,
|
|
54,
|
|
60,
|
|
37,
|
|
68,
|
|
35,
|
|
55,
|
|
80,
|
|
2,
|
|
33,
|
|
26,
|
|
93,
|
|
70,
|
|
45,
|
|
44,
|
|
3,
|
|
66,
|
|
75,
|
|
4,
|
|
};
|
|
|
|
NonnullRefPtr<SQL::HashIndex> setup_hash_index(SQL::Heap& heap);
|
|
void insert_and_get_to_and_from_hash_index(int num_keys);
|
|
void insert_into_and_scan_hash_index(int num_keys);
|
|
|
|
NonnullRefPtr<SQL::HashIndex> setup_hash_index(SQL::Heap& heap)
|
|
{
|
|
SQL::TupleDescriptor tuple_descriptor;
|
|
tuple_descriptor.append({ "key_value", SQL::SQLType::Integer, SQL::Order::Ascending });
|
|
tuple_descriptor.append({ "text_value", SQL::SQLType::Text, SQL::Order::Ascending });
|
|
|
|
auto directory_pointer = heap.user_value(0);
|
|
if (!directory_pointer) {
|
|
directory_pointer = heap.new_record_pointer();
|
|
heap.set_user_value(0, directory_pointer);
|
|
}
|
|
auto hash_index = SQL::HashIndex::construct(heap, tuple_descriptor, directory_pointer);
|
|
return hash_index;
|
|
}
|
|
|
|
void insert_and_get_to_and_from_hash_index(int num_keys)
|
|
{
|
|
ScopeGuard guard([]() { unlink("test.db"); });
|
|
{
|
|
auto heap = SQL::Heap::construct("test.db");
|
|
auto hash_index = setup_hash_index(heap);
|
|
|
|
for (auto ix = 0; ix < num_keys; ix++) {
|
|
SQL::Key k(hash_index->descriptor());
|
|
k[0] = keys[ix];
|
|
k[1] = String::formatted("The key value is {} and the pointer is {}", keys[ix], pointers[ix]);
|
|
k.set_pointer(pointers[ix]);
|
|
hash_index->insert(k);
|
|
}
|
|
#ifdef LIST_HASH_INDEX
|
|
hash_index->list_hash();
|
|
#endif
|
|
}
|
|
|
|
{
|
|
auto heap = SQL::Heap::construct("test.db");
|
|
auto hash_index = setup_hash_index(heap);
|
|
|
|
for (auto ix = 0; ix < num_keys; ix++) {
|
|
SQL::Key k(hash_index->descriptor());
|
|
k[0] = keys[ix];
|
|
k[1] = String::formatted("The key value is {} and the pointer is {}", keys[ix], pointers[ix]);
|
|
auto pointer_opt = hash_index->get(k);
|
|
EXPECT(pointer_opt.has_value());
|
|
EXPECT_EQ(pointer_opt.value(), pointers[ix]);
|
|
}
|
|
}
|
|
}
|
|
|
|
TEST_CASE(hash_index_one_key)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(1);
|
|
}
|
|
|
|
TEST_CASE(hash_index_four_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(4);
|
|
}
|
|
|
|
TEST_CASE(hash_index_five_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(5);
|
|
}
|
|
|
|
TEST_CASE(hash_index_10_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(10);
|
|
}
|
|
|
|
TEST_CASE(hash_index_13_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(13);
|
|
}
|
|
|
|
TEST_CASE(hash_index_20_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(20);
|
|
}
|
|
|
|
TEST_CASE(hash_index_25_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(25);
|
|
}
|
|
|
|
TEST_CASE(hash_index_30_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(30);
|
|
}
|
|
|
|
TEST_CASE(hash_index_35_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(35);
|
|
}
|
|
|
|
TEST_CASE(hash_index_40_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(40);
|
|
}
|
|
|
|
TEST_CASE(hash_index_45_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(45);
|
|
}
|
|
|
|
TEST_CASE(hash_index_50_keys)
|
|
{
|
|
insert_and_get_to_and_from_hash_index(50);
|
|
}
|
|
|
|
void insert_into_and_scan_hash_index(int num_keys)
|
|
{
|
|
ScopeGuard guard([]() { unlink("test.db"); });
|
|
{
|
|
auto heap = SQL::Heap::construct("test.db");
|
|
auto hash_index = setup_hash_index(heap);
|
|
|
|
for (auto ix = 0; ix < num_keys; ix++) {
|
|
SQL::Key k(hash_index->descriptor());
|
|
k[0] = keys[ix];
|
|
k[1] = String::formatted("The key value is {} and the pointer is {}", keys[ix], pointers[ix]);
|
|
k.set_pointer(pointers[ix]);
|
|
hash_index->insert(k);
|
|
}
|
|
#ifdef LIST_HASH_INDEX
|
|
hash_index->list_hash();
|
|
#endif
|
|
}
|
|
|
|
{
|
|
auto heap = SQL::Heap::construct("test.db");
|
|
auto hash_index = setup_hash_index(heap);
|
|
Vector<bool> found;
|
|
for (auto ix = 0; ix < num_keys; ix++) {
|
|
found.append(false);
|
|
}
|
|
|
|
int count = 0;
|
|
for (auto iter = hash_index->begin(); !iter.is_end(); iter++, count++) {
|
|
auto key = (*iter);
|
|
auto key_value = (int)key[0];
|
|
for (auto ix = 0; ix < num_keys; ix++) {
|
|
if (keys[ix] == key_value) {
|
|
EXPECT_EQ(key.pointer(), pointers[ix]);
|
|
if (found[ix])
|
|
FAIL(String::formatted("Key {}, index {} already found previously", key_value, ix));
|
|
found[ix] = true;
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
|
|
#ifdef LIST_HASH_INDEX
|
|
hash_index->list_hash();
|
|
#endif
|
|
EXPECT_EQ(count, num_keys);
|
|
for (auto ix = 0; ix < num_keys; ix++) {
|
|
if (!found[ix])
|
|
FAIL(String::formatted("Key {}, index {} not found", keys[ix], ix));
|
|
}
|
|
}
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_one_key)
|
|
{
|
|
insert_into_and_scan_hash_index(1);
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_four_keys)
|
|
{
|
|
insert_into_and_scan_hash_index(4);
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_five_keys)
|
|
{
|
|
insert_into_and_scan_hash_index(5);
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_10_keys)
|
|
{
|
|
insert_into_and_scan_hash_index(10);
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_15_keys)
|
|
{
|
|
insert_into_and_scan_hash_index(15);
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_20_keys)
|
|
{
|
|
insert_into_and_scan_hash_index(20);
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_30_keys)
|
|
{
|
|
insert_into_and_scan_hash_index(30);
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_40_keys)
|
|
{
|
|
insert_into_and_scan_hash_index(40);
|
|
}
|
|
|
|
TEST_CASE(hash_index_scan_50_keys)
|
|
{
|
|
insert_into_and_scan_hash_index(50);
|
|
}
|