Chinaunix首页 | 论坛 | 博客
  • 博客访问: 528949
  • 博文数量: 104
  • 博客积分: 2089
  • 博客等级: 大尉
  • 技术积分: 1691
  • 用 户 组: 普通用户
  • 注册时间: 2010-06-29 08:48
文章分类

全部博文(104)

文章存档

2015年(1)

2013年(13)

2012年(31)

2011年(59)

分类: SQLite/嵌入式数据库

2013-02-04 16:11:54

很早就知道sqlite,一直没有机会,因为我接触的项目要么不使用数据库(一两个文配置文件就可以搞定),要么就会有很多的数据,用到postgresql,操练sqlite的还没有。现在我有个自己的小测试例子,写个数据库对比的小项目例子,我就准备把数据存储在sqlite上,第一数据不是很大,百位级别;为桌面应用软件,嵌入式的。sqlite 很适合。

安装:

1> os:ubuntu。

    由于是使用代理上的网,估计没有配置好,apt-get intall sqlite3自动安装没有成功,就采用了源码安装。

2> package from : ,要选择这个autoconf的,不然就只有源码没有项目管理脚本(auto 工具集)

3> tar -xzvf  , cd,./configure, make , make install. 很顺利的就安装完毕

4> 运行 sqlite3. 有错误

# sqlite3
SQLite header and source version mismatch
2011-11-01 00:52:41 c7c6050ef060877ebe77b41d959e9df13f8c9b5e
2013-01-09 11:53:05 c0e09560d26f0a6456be9dd3447f5311eb4f238f


在参考了http://jianshusoft.blog.51cto.com/2380869/824575 这篇文章后,发现


    1. 在make install 的log中指出安装的路径都在/usr/local 下,在configure中也有对应的代码代码指出了这个路径
    2. 在/usr/lib/i386-linux-gnu中也确实有libsqlite3*的文件


遂果断处理:mv /usr/lib/i386-linux-gnu/*sqlite3* /tmp


在运行sqlite3, cmd的管理界面出现了。


快速入门:

shell中使用命令来了解sqlite的使用,文中大部分篇幅介绍了在sqlite3交互界面中的使用,还有些直接使用sh命令交互,非常好。


c语言使用

一较为详细的c语言使用sqlite3的例子,测试通过。包含了如下过程,打开数据库文件(没有则新建),建立表(没有则新建),插入数据,查询输出,关闭数据库文件。文件虽小,却包含整个过程,配合博主的解释,和评论者的积极参与,可窥探sqlite使用概貌。

源码:

#include
#include
#include

int main(int argc, char** args)
{
    // Create an int variable for storing the return code for each call
    int retval;

    // The number of queries to be handled,size of each query and pointer
    int q_cnt = 5,q_size = 150,ind = 0;
    char **queries = malloc(sizeof(char) * q_cnt * q_size);

    // A prepered statement for fetching tables
    sqlite3_stmt *stmt;

    // Create a handle for database connection, create a pointer to sqlite3
    sqlite3 *handle;

    // try to create the database. If it doesnt exist, it would be created
    // pass a pointer to the pointer to sqlite3, in short sqlite3**
    retval = sqlite3_open("sampledb.sqlite3",&handle);
    // If connection failed, handle returns NULL
    if(retval)
    {
        printf("Database connection failed\n");
        return -1;
    }
    printf("Connection successful\n");

    // Create the SQL query for creating a table
    char create_table[100] = "CREATE TABLE IF NOT EXISTS users (uname TEXT PRIMARY KEY,pass TEXT NOT NULL,activated INTEGER)";

    // Execute the query for creating the table
    retval = sqlite3_exec(handle,create_table,0,0,0);

    // Insert first row and second row
    queries[ind++] = "INSERT INTO users VALUES('manish','mani',1)";
    retval = sqlite3_exec(handle,queries[ind-1],0,0,0);
    queries[ind++] = "INSERT INTO users VALUES('mehul','pulsar',0)";
    retval = sqlite3_exec(handle,queries[ind-1],0,0,0);

    // select those rows from the table
    queries[ind++] = "SELECT * from users";
    retval = sqlite3_prepare_v2(handle,queries[ind-1],-1,&stmt,0);
    if(retval)
    {
        printf("Selecting data from DB Failed\n");
        return -1;
    }

    // Read the number of rows fetched
    int cols = sqlite3_column_count(stmt);

    while(1)
    {
        // fetch a row's status
        retval = sqlite3_step(stmt);

        if(retval == SQLITE_ROW)
        {
            // SQLITE_ROW means fetched a row

            // sqlite3_column_text returns a const void* , typecast it to const char*
            for(int col=0 ; col
	



编译: gcc basics.c -o basics -lsqlite3 -std=c99

运行: ./basics

Connection successful
uname = manish  pass = mani     activated = 1
uname = mehul   pass = pulsar   activated = 0
All rows fetched
文件列表如下



# ls
basics  basics.c  sampledb.sqlite3

例子原文:

不知道在不使用代理情况下是否能够看到,我仅贴原文如下(不包含评论):

------------------------------------------------------


‘C’ has always been my favourite language due to simple facts that it is beautiful and low level in nature. I don’t claim that am a ‘Geek’ in this language, its just my love that pulls me towards it. Let’s have a look at the other languages usually  liked by the public – VB, Java, Perl , Python. All of them may be good in their own ways but C kicks ass. VB?? urgh… Sorry! I vow not to code in it. It’s syntax is very unusual and every Tom,Dick and Harry claims to be a champ of that language.

The biggest problem which I face in C is storing data or in short making data persistent. One way is to write the required to a file on the disk in a fixed format. This stored data can then be read and parsed as per requirement. This approach is good for small amount of data, but what about huge amount of data? You would spend a big share of your time just for structured file I/O. Finally you would land up writing a small module for this work. Why not use any such existing database software for the same? Here comes SQLite for rescue.

I have seen a lot of tutorials on the net, they are very good but none of them suited my needs. The requirement was to explain a sample code line by line. After lots of googling and tea, I managed to make it work! The code snippet which I made is able to create new database if it does not exist, create a table if it does not exist, enter two rows and then fetch those two rows and print them on the screen.  which I have committed the code to my personal google code repository.

Let me explain the code. Sorry for not aligning it. Please .

#include
#include
#include


int
 main(int argc, char** args)
{
// Create an int variable for storing the return code for each call
int retval;

Include stdio.h, sqlite3.h and stdlib.h , stdlib.h is for malloc and sqlite3.h contains the standard function declarations needed for the required functionality.

// The number of queries to be handled,size of each query and pointer
int q_cnt = 5,q_size = 150,ind = 0;
char **queries = malloc(sizeof(char) * q_cnt * q_size);

q_cnt stored the number of queries we may want to do, q_size stores the max size of a SQL query, ind is the index.

**queries is a double array or matrix which stores the multiple queries. The total amount of storage to be allocated is sizeof(char) * q_cnt * q_size

// A prepered statement for fetching tables
sqlite3_stmt *stmt;

// Create a handle for database connection, create a pointer to sqlite3
sqlite3 *handle;

// try to create the database. If it doesnt exist, it would be created
// pass a pointer to the pointer to sqlite3, in short sqlite3**
retval = sqlite3_open(“sampledb.sqlite3″,&handle);
// If connection failed, handle returns NULL
if(retval)
{
printf(“Database connection failed\n”);
return -1;
}
printf(“Connection successful\n”);

We need to create a pointer to sqlite3 and sqlite3_stmt structures. sqlite3 is the structure which is to hold the database connection handle. sqlite3_stmt is just like a cursor to a database.

sqlite3_open function needs the address of the sqlite3 database instance on the disk. The second parameter is the pointer to the pointer to sqlite3 structure. One mistake which I stumbled upon was to create a sqlite3 ** handle and then pass it to this function. The correct way is to create a sqlite3* handle and then pass the pointer to it using the & operator

// Create the SQL query for creating a table
char create_table[100] = “CREATE TABLE IF NOT EXISTS users (uname TEXT PRIMARY KEY,pass TEXT NOT NULL,activated INTEGER)”;

// Execute the query for creating the table
retval = sqlite3_exec(handle,create_table,0,0,0);

// Insert first row and second row
queries[ind++] = “INSERT INTO users VALUES(‘manish’,'manish’,1)”;
retval = sqlite3_exec(handle,queries[ind-1],0,0,0);
queries[ind++] = “INSERT INTO users VALUES(‘mehul’,'pulsar’,0)”;
retval = sqlite3_exec(handle,queries[ind-1],0,0,0);

Create a table if it does not exist and then insert two rows. Note that sqlite3 does not support inserting two rows in one single query. Maybe I need to confirm this fact again, but I never worked for me ever.

// select those rows from the table
queries[ind++] = “SELECT * from users”;
retval = sqlite3_prepare_v2(handle,queries[ind-1],-1,&stmt,0);
if(retval)
{
printf(“Selecting data from DB Failed\n”);
return -1;
}

// Read the number of rows fetched
int cols = sqlite3_column_count(stmt);

Create a prepared statement for fetching data from the database usingsqlite3_prepare_v2 function call. The first parameter is the database handle itself which is a sqlite3* pointer. The second parameter is the SQL statement which needs to be executed. The third parameter tells upto how long the second parameter to be read. Pass -1 to make it read till line terminator. Fourth statement is the pointer to pointer to prepared statement structure. Take care of the pointer concept as I told about sqlite3 structure. The fifth parameter is filled with the unused portion of the query. Have a look at the official documentation.

sqlite3_column_count function gets the number of columns for the result fetched.

while(1)
{
// fetch a row’s status
retval = sqlite3_step(stmt);

if(retval == SQLITE_ROW)
{
// SQLITE_ROW means fetched a row

// sqlite3_column_text returns a const void* , typecast it to const char*
for(int col=0 ; col {
const char *val = (const char*)sqlite3_column_text(stmt,col);
printf(“%s = %s\t”,sqlite3_column_name(stmt,col),val);
}
printf(“\n”);
}
else if(retval == SQLITE_DONE)
{
// All rows finished
printf(“All rows fetched\n”);
break;
}
else
{
// Some error encountered
printf(“Some error encountered\n”);
return -1;
}
}

We have put this code in infinite while loop as we are not sure how much rows it contains. Usually, the table returns n+1 rows, where 1 extra row is for telling that all rows have been fetched. sqlite3_step returns the status which is actually an enumeration. . Two most used are SQLITE_DONESQLITE_ROW. The former tells that all the rows have been fetched, now the user can come out of this loop and continue. SQLITE_ROW tells that a valid row has been fetched.

// Close the handle to free memory
sqlite3_close(handle);
return 0;
}

sqlite3_close simply closes the database connection.

Save the code in a file named, say dataman.c , compile it using the command

$ gcc dataman.c -o dataman -l sqlite –std=c99

You obviously need to have sqlite development headers installed for compiling the same. The name of the package on Ubuntu is libsqlite3-dev

Official SQLite Documentation



------------------------------------------------------




参考:





  1. 快速入门 
  2. “SQLite header and source version mismatch” http://jianshusoft.blog.51cto.com/2380869/824575
  3. c语言的较为详细的例子 




阅读(1493) | 评论(0) | 转发(1) |
给主人留下些什么吧!~~