第 5 部分 - 持久化到磁盘
"世界上没有任何东西可以取代坚持"。- 卡尔文-柯立芝
我们的数据库允许您插入记录并将其读出，但前提是您必须保持程序运行。如果杀死程序并重新启动，所有记录都会消失。下面是我们想要的行为规范：
it 'keeps data after closing connection' do
result1 = run_script([
"insert 1 user1 person1@example.com",
".exit",
])
expect(result1).to match_array([
"db > Executed.",
"db > ",
])
result2 = run_script([
"select",
".exit",
])
expect(result2).to match_array([
"db > (1, user1, person1@example.com)",
"Executed.",
"db > ",
])
end 与 sqlite 一样，我们要通过将整个数据库保存到文件来持久化记录。
通过将记录序列化为页面大小的内存块，我们已经为此做好了准备。为了增加持久性，我们可以简单地将这些内存块写入文件，并在下次程序启动时将它们读回内存。
为了简化操作，我们将创建一个名为 "页管理器 "的抽象概念。我们向页管理器请求页码 x，页管理器就会返回给我们一个内存块。它首先会查看缓存。在缓存未命中时，它会（通过读取数据库文件）将数据从磁盘复制到内存中。
我们的程序如何与 SQLite 架构相匹配
页管理器访问页面缓存和文件。表对象通过页管理器请求页面：
+typedef struct {
+ int file_descriptor;
+ uint32_t file_length;
+ void* pages[TABLE_MAX_PAGES];
+} Pager;
+
typedef struct {
- void* pages[TABLE_MAX_PAGES];
+ Pager* pager;
uint32_t num_rows;
} Table;我把 new_table() 改名为 db_open()，因为它现在具有打开数据库连接的作用。我所说的打开连接是指
- 打开数据库文件
- 初始化页面数据结构
- 初始化表格数据结构
-Table* new_table() {
+Table* db_open(const char* filename) {
+ Pager* pager = pager_open(filename);
+ uint32_t num_rows = pager->file_length / ROW_SIZE;
+
Table* table = malloc(sizeof(Table));
- table->num_rows = 0;
+ table->pager = pager;
+ table->num_rows = num_rows;
return table;
}db_open() 会调用 pager_open()，打开数据库文件并记录其大小。它还会将页面缓存初始化为全部 NULL。
+Pager* pager_open(const char* filename) {
+ int fd = open(filename,
+ O_RDWR | // Read/Write mode
+ O_CREAT, // Create file if it does not exist
+ S_IWUSR | // User write permission
+ S_IRUSR // User read permission
+ );
+
+ if (fd == -1) {
+ printf("Unable to open file\n");
+ exit(EXIT_FAILURE);
+ }
+
+ off_t file_length = lseek(fd, 0, SEEK_END);
+
+ Pager* pager = malloc(sizeof(Pager));
+ pager->file_descriptor = fd;
+ pager->file_length = file_length;
+
+ for (uint32_t i = 0; i < TABLE_MAX_PAGES; i++) {
+ pager->pages[i] = NULL;
+ }
+
+ return pager;
+}按照新的抽象方法，我们将获取页面的逻辑移到它自己的方法中：
void* row_slot(Table* table, uint32_t row_num) {
uint32_t page_num = row_num / ROWS_PER_PAGE;
- void* page = table->pages[page_num];
- if (page == NULL) {
- // Allocate memory only when we try to access page
- page = table->pages[page_num] = malloc(PAGE_SIZE);
- }
+ void* page = get_page(table->pager, page_num);
uint32_t row_offset = row_num % ROWS_PER_PAGE;
uint32_t byte_offset = row_offset * ROW_SIZE;
return page + byte_offset;
}get_page() 方法具有处理缓存缺失的逻辑。我们假设数据库文件中的页面是一个接一个保存的：第 0 页在偏移量 0 处，第 1 页在偏移量 4096 处，第 2 页在偏移量 8192 处，等等。如果请求的页面位于文件边界之外，我们知道它应该是空白的，因此我们只需分配一些内存并返回。稍后将缓存刷新到磁盘时，该页面将被添加到文件中。
+void* get_page(Pager* pager, uint32_t page_num) {
+ if (page_num > TABLE_MAX_PAGES) {
+ printf("Tried to fetch page number out of bounds. %d > %d\n", page_num,
+ TABLE_MAX_PAGES);
+ exit(EXIT_FAILURE);
+ }
+
+ if (pager->pages[page_num] == NULL) {
+ // Cache miss. Allocate memory and load from file.
+ void* page = malloc(PAGE_SIZE);
+ uint32_t num_pages = pager->file_length / PAGE_SIZE;
+
+ // We might save a partial page at the end of the file
+ if (pager->file_length % PAGE_SIZE) {
+ num_pages += 1;
+ }
+
+ if (page_num <= num_pages) {
+ lseek(pager->file_descriptor, page_num * PAGE_SIZE, SEEK_SET);
+ ssize_t bytes_read = read(pager->file_descriptor, page, PAGE_SIZE);
+ if (bytes_read == -1) {
+ printf("Error reading file: %d\n", errno);
+ exit(EXIT_FAILURE);
+ }
+ }
+
+ pager->pages[page_num] = page;
+ }
+
+ return pager->pages[page_num];
+}目前，我们将等到用户关闭数据库连接后再将缓存刷新到磁盘。当用户退出时，我们将调用一个名为 db_close()的新方法。
- 将页面缓存刷新到磁盘
- 关闭数据库文件
- 释放 Pager 和 Table 数据结构的内存
+void db_close(Table* table) {
+ Pager* pager = table->pager;
+ uint32_t num_full_pages = table->num_rows / ROWS_PER_PAGE;
+
+ for (uint32_t i = 0; i < num_full_pages; i++) {
+ if (pager->pages[i] == NULL) {
+ continue;
+ }
+ pager_flush(pager, i, PAGE_SIZE);
+ free(pager->pages[i]);
+ pager->pages[i] = NULL;
+ }
+
+ // There may be a partial page to write to the end of the file
+ // This should not be needed after we switch to a B-tree
+ uint32_t num_additional_rows = table->num_rows % ROWS_PER_PAGE;
+ if (num_additional_rows > 0) {
+ uint32_t page_num = num_full_pages;
+ if (pager->pages[page_num] != NULL) {
+ pager_flush(pager, page_num, num_additional_rows * ROW_SIZE);
+ free(pager->pages[page_num]);
+ pager->pages[page_num] = NULL;
+ }
+ }
+
+ int result = close(pager->file_descriptor);
+ if (result == -1) {
+ printf("Error closing db file.\n");
+ exit(EXIT_FAILURE);
+ }
+ for (uint32_t i = 0; i < TABLE_MAX_PAGES; i++) {
+ void* page = pager->pages[i];
+ if (page) {
+ free(page);
+ pager->pages[i] = NULL;
+ }
+ }
+ free(pager);
+ free(table);
+}
+
-MetaCommandResult do_meta_command(InputBuffer* input_buffer) {
+MetaCommandResult do_meta_command(InputBuffer* input_buffer, Table* table) {
if (strcmp(input_buffer->buffer, ".exit") == 0) {
+ db_close(table);
exit(EXIT_SUCCESS);
} else {
return META_COMMAND_UNRECOGNIZED_COMMAND;在我们当前的设计中，文件长度编码了数据库中的行数，因此我们需要在文件末尾写入部分页面。这就是 pager_flush()同时需要页码和大小的原因。这不是最棒的设计，但当我们开始实现 B 树时，它很快就会消失。
+void pager_flush(Pager* pager, uint32_t page_num, uint32_t size) {
+ if (pager->pages[page_num] == NULL) {
+ printf("Tried to flush null page\n");
+ exit(EXIT_FAILURE);
+ }
+
+ off_t offset = lseek(pager->file_descriptor, page_num * PAGE_SIZE, SEEK_SET);
+
+ if (offset == -1) {
+ printf("Error seeking: %d\n", errno);
+ exit(EXIT_FAILURE);
+ }
+
+ ssize_t bytes_written =
+ write(pager->file_descriptor, pager->pages[page_num], size);
+
+ if (bytes_written == -1) {
+ printf("Error writing: %d\n", errno);
+ exit(EXIT_FAILURE);
+ }
+}最后，我们需要接受文件名作为命令行参数。不要忘记在 do_meta_command 中添加额外的参数：
int main(int argc, char* argv[]) {
- Table* table = new_table();
+ if (argc < 2) {
+ printf("Must supply a database filename.\n");
+ exit(EXIT_FAILURE);
+ }
+
+ char* filename = argv[1];
+ Table* table = db_open(filename);
+
InputBuffer* input_buffer = new_input_buffer();
while (true) {
print_prompt();
read_input(input_buffer);
if (input_buffer->buffer[0] == '.') {
- switch (do_meta_command(input_buffer)) {
+ switch (do_meta_command(input_buffer, table)) {有了这些更改，我们就可以关闭数据库，然后重新打开，我们的记录仍然存在！
~ ./db mydb.db
db > insert 1 cstack foo@bar.com
Executed.
db > insert 2 voltorb volty@example.com
Executed.
db > .exit
~
~ ./db mydb.db
db > select
(1, cstack, foo@bar.com)
(2, voltorb, volty@example.com)
Executed.
db > .exit
~为了增加乐趣，让我们看看 mydb.db 是如何存储数据的。我将使用 vim 作为十六进制编辑器来查看文件的内存布局：
vim mydb.db
:%!xxd
当前文件格式
前四个字节是第一行的 id（4 个字节，因为我们存储的是 uint32_t）。它是按小字节顺序存储的，所以最小有效字节在前（01），然后是高阶字节（00 00 00）。我们使用 memcpy() 将 Row 结构中的字节复制到页面缓存中，这意味着该结构在内存中是按照 little-endian 字节顺序排列的。这是我编译程序的机器属性。如果我们想在我的机器上写一个数据库文件，然后在大前沿机器上读取它，我们就必须改变我们的 serialize_row() 和 deserialize_row() 方法，使其始终以相同的顺序存储和读取字节。
接下来的 33 个字节以空字符串的形式存储用户名。显然，ASCII 十六进制的 "cstack "是 63 73 74 61 63 6b，后面是一个空字符（00）。其余 33 个字节未使用。
接下来的 256 个字节以同样的方式存储电子邮件。在这里，我们可以看到终止空字符后的一些随机垃圾。这很可能是 Row 结构中未初始化的内存造成的。我们将整个 256 字节的电子邮件缓冲区复制到文件中，包括字符串结束后的所有字节。我们分配该结构时内存中的任何内容都还在。不过，由于我们使用了终止空字符，因此对行为没有影响。
void serialize_row(Row* source, void* destination) {
memcpy(destination + ID_OFFSET, &(source->id), ID_SIZE);
- memcpy(destination + USERNAME_OFFSET, &(source->username), USERNAME_SIZE);
- memcpy(destination + EMAIL_OFFSET, &(source->email), EMAIL_SIZE);
+ strncpy(destination + USERNAME_OFFSET, source->username, USERNAME_SIZE);
+ strncpy(destination + EMAIL_OFFSET, source->email, EMAIL_SIZE);
}结论
好吧我们有毅力它不是最棒的。例如，如果你不输入 .exit 就关闭程序，你就会丢失更改的内容。此外，我们会将所有页面写回磁盘，即使是从磁盘读取后没有更改的页面也不例外。这些问题我们可以稍后解决。
下一次我们将引入游标，这将使 B 树的实现更加容易。
在此之前
完整差异
+#include <errno.h>
+#include <fcntl.h>
#include <stdbool.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stdint.h>
+#include <unistd.h>
struct InputBuffer_t {
char* buffer;
@@ -62,9 +65,16 @@ const uint32_t PAGE_SIZE = 4096;
const uint32_t ROWS_PER_PAGE = PAGE_SIZE / ROW_SIZE;
const uint32_t TABLE_MAX_ROWS = ROWS_PER_PAGE * TABLE_MAX_PAGES;
+typedef struct {
+ int file_descriptor;
+ uint32_t file_length;
+ void* pages[TABLE_MAX_PAGES];
+} Pager;
+
typedef struct {
uint32_t num_rows;
- void* pages[TABLE_MAX_PAGES];
+ Pager* pager;
} Table;
@@ -84,32 +94,81 @@ void deserialize_row(void *source, Row* destination) {
memcpy(&(destination->email), source + EMAIL_OFFSET, EMAIL_SIZE);
}
+void* get_page(Pager* pager, uint32_t page_num) {
+ if (page_num > TABLE_MAX_PAGES) {
+ printf("Tried to fetch page number out of bounds. %d > %d\n", page_num,
+ TABLE_MAX_PAGES);
+ exit(EXIT_FAILURE);
+ }
+
+ if (pager->pages[page_num] == NULL) {
+ // Cache miss. Allocate memory and load from file.
+ void* page = malloc(PAGE_SIZE);
+ uint32_t num_pages = pager->file_length / PAGE_SIZE;
+
+ // We might save a partial page at the end of the file
+ if (pager->file_length % PAGE_SIZE) {
+ num_pages += 1;
+ }
+
+ if (page_num <= num_pages) {
+ lseek(pager->file_descriptor, page_num * PAGE_SIZE, SEEK_SET);
+ ssize_t bytes_read = read(pager->file_descriptor, page, PAGE_SIZE);
+ if (bytes_read == -1) {
+ printf("Error reading file: %d\n", errno);
+ exit(EXIT_FAILURE);
+ }
+ }
+
+ pager->pages[page_num] = page;
+ }
+
+ return pager->pages[page_num];
+}
+
void* row_slot(Table* table, uint32_t row_num) {
uint32_t page_num = row_num / ROWS_PER_PAGE;
- void *page = table->pages[page_num];
- if (page == NULL) {
- // Allocate memory only when we try to access page
- page = table->pages[page_num] = malloc(PAGE_SIZE);
- }
+ void *page = get_page(table->pager, page_num);
uint32_t row_offset = row_num % ROWS_PER_PAGE;
uint32_t byte_offset = row_offset * ROW_SIZE;
return page + byte_offset;
}
-Table* new_table() {
- Table* table = malloc(sizeof(Table));
- table->num_rows = 0;
+Pager* pager_open(const char* filename) {
+ int fd = open(filename,
+ O_RDWR | // Read/Write mode
+ O_CREAT, // Create file if it does not exist
+ S_IWUSR | // User write permission
+ S_IRUSR // User read permission
+ );
+
+ if (fd == -1) {
+ printf("Unable to open file\n");
+ exit(EXIT_FAILURE);
+ }
+
+ off_t file_length = lseek(fd, 0, SEEK_END);
+
+ Pager* pager = malloc(sizeof(Pager));
+ pager->file_descriptor = fd;
+ pager->file_length = file_length;
+
for (uint32_t i = 0; i < TABLE_MAX_PAGES; i++) {
- table->pages[i] = NULL;
+ pager->pages[i] = NULL;
}
- return table;
+
+ return pager;
}
-void free_table(Table* table) {
- for (int i = 0; table->pages[i]; i++) {
- free(table->pages[i]);
- }
- free(table);
+Table* db_open(const char* filename) {
+ Pager* pager = pager_open(filename);
+ uint32_t num_rows = pager->file_length / ROW_SIZE;
+
+ Table* table = malloc(sizeof(Table));
+ table->pager = pager;
+ table->num_rows = num_rows;
+
+ return table;
}
InputBuffer* new_input_buffer() {
@@ -142,10 +201,76 @@ void close_input_buffer(InputBuffer* input_buffer) {
free(input_buffer);
}
+void pager_flush(Pager* pager, uint32_t page_num, uint32_t size) {
+ if (pager->pages[page_num] == NULL) {
+ printf("Tried to flush null page\n");
+ exit(EXIT_FAILURE);
+ }
+
+ off_t offset = lseek(pager->file_descriptor, page_num * PAGE_SIZE,
+ SEEK_SET);
+
+ if (offset == -1) {
+ printf("Error seeking: %d\n", errno);
+ exit(EXIT_FAILURE);
+ }
+
+ ssize_t bytes_written = write(
+ pager->file_descriptor, pager->pages[page_num], size
+ );
+
+ if (bytes_written == -1) {
+ printf("Error writing: %d\n", errno);
+ exit(EXIT_FAILURE);
+ }
+}
+
+void db_close(Table* table) {
+ Pager* pager = table->pager;
+ uint32_t num_full_pages = table->num_rows / ROWS_PER_PAGE;
+
+ for (uint32_t i = 0; i < num_full_pages; i++) {
+ if (pager->pages[i] == NULL) {
+ continue;
+ }
+ pager_flush(pager, i, PAGE_SIZE);
+ free(pager->pages[i]);
+ pager->pages[i] = NULL;
+ }
+
+ // There may be a partial page to write to the end of the file
+ // This should not be needed after we switch to a B-tree
+ uint32_t num_additional_rows = table->num_rows % ROWS_PER_PAGE;
+ if (num_additional_rows > 0) {
+ uint32_t page_num = num_full_pages;
+ if (pager->pages[page_num] != NULL) {
+ pager_flush(pager, page_num, num_additional_rows * ROW_SIZE);
+ free(pager->pages[page_num]);
+ pager->pages[page_num] = NULL;
+ }
+ }
+
+ int result = close(pager->file_descriptor);
+ if (result == -1) {
+ printf("Error closing db file.\n");
+ exit(EXIT_FAILURE);
+ }
+ for (uint32_t i = 0; i < TABLE_MAX_PAGES; i++) {
+ void* page = pager->pages[i];
+ if (page) {
+ free(page);
+ pager->pages[i] = NULL;
+ }
+ }
+
+ free(pager);
+ free(table);
+}
+
MetaCommandResult do_meta_command(InputBuffer* input_buffer, Table *table) {
if (strcmp(input_buffer->buffer, ".exit") == 0) {
close_input_buffer(input_buffer);
- free_table(table);
+ db_close(table);
exit(EXIT_SUCCESS);
} else {
return META_COMMAND_UNRECOGNIZED_COMMAND;
@@ -182,6 +308,7 @@ PrepareResult prepare_insert(InputBuffer* input_buffer, Statement* statement) {
return PREPARE_SUCCESS;
}
+
PrepareResult prepare_statement(InputBuffer* input_buffer,
Statement* statement) {
if (strncmp(input_buffer->buffer, "insert", 6) == 0) {
@@ -227,7 +354,14 @@ ExecuteResult execute_statement(Statement* statement, Table *table) {
}
int main(int argc, char* argv[]) {
- Table* table = new_table();
+ if (argc < 2) {
+ printf("Must supply a database filename.\n");
+ exit(EXIT_FAILURE);
+ }
+
+ char* filename = argv[1];
+ Table* table = db_open(filename);
+
InputBuffer* input_buffer = new_input_buffer();
while (true) {
print_prompt();以及我们测试的差异：
describe 'database' do
+ before do
+ `rm -rf test.db`
+ end
+
def run_script(commands)
raw_output = nil
- IO.popen("./db", "r+") do |pipe|
+ IO.popen("./db test.db", "r+") do |pipe|
commands.each do |command|
pipe.puts command
end
@@ -28,6 +32,27 @@ describe 'database' do
])
end
+ it 'keeps data after closing connection' do
+ result1 = run_script([
+ "insert 1 user1 person1@example.com",
+ ".exit",
+ ])
+ expect(result1).to match_array([
+ "db > Executed.",
+ "db > ",
+ ])
+
+ result2 = run_script([
+ "select",
+ ".exit",
+ ])
+ expect(result2).to match_array([
+ "db > (1, user1, person1@example.com)",
+ "Executed.",
+ "db > ",
+ ])
+ end
+
it 'prints error message when table is full' do
script = (1..1401).map do |i|
"insert #{i} user#{i} person#{i}@example.com"
862
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
