It is an ORM, a small compiler, a dynamic SQL languages
Box<dyn Driver> separation#{arg}, ${arg}, ? placeholder (pg/mssql auto processing '?' to '$1' and '@P1')py_sql query lang and html_sql (Inspired Mybatis).log crate#![forbid(unsafe_code)] enabledpy_sql/ html_sql functions (see)Thanks to SQLX, deadpool, Tiberius, MyBatis, xorm and so on reference design or code implementation. Release of V4.0 is Inspired and supported by these frameworks.
rbatis.query_decode::<Vec<i32>>("", vec![]).await; on benches bench_raw()MockTable::insert(&mut rbatis.clone(),&t).await; on benches bench_insert()MockTable::select_all(&mut rbatis.clone()).await.unwrap(); on benches bench_select()``` ---- bench_raw stdout ---- Time: 84.6463ms ,each:846 ns/op QPS: 1181286 QPS/s
---- bench_insert stdout ----(macos,cpu-M1Max) Time: 378.186333ms ,each:3781 ns/op QPS: 264418 QPS/s
---- bench_select stdout ----(macos,cpu-M1Max) Time: 112.927916ms ,each:1129 ns/op QPS: 885486 QPS/s ```
| data structure | is supported |
|---------------------------------------------------------------------------|--------------|
| Option | √ |
| Vec | √ |
| HashMap | √ |
| i32,i64,f32,f64,bool,String...more rust type | √ |
| rbatis::rbdc::types::{Date,DateTime,Time,Timestamp,Decimal,Json} | √ |
| rbatis::plugin::page::{Page
| database(crates.io) | github_link | |-----------------------------------------------------|--------------------------------------------------------------------------------| | Mysql | rbatis/rbdc-mysql | | Postgres | rbatis/rbdc-pg | | Sqlite | rbatis/rbdc-sqlite | | Mssql | rbatis/rbdc-mssql | | MariaDB | rbatis/rbdc-mysql | | TiDB | rbatis/rbdc-mysql | | CockroachDB | rbatis/rbdc-pg | | Oracle | chenpengfan/rbdc-oracle | | TDengine | tdcare/rbdc-tdengine |
| platform | is supported |
|-------------------------|--------------|
| Linux(unbutu laster*) | √ |
| Apple/MacOS(laster) | √ |
| Windows(latest) | √ |
Rust v1.63+ later
Cargo.toml
```toml tokio = { version = "1", features = ["full"] } log = "0.4" fast_log = "1.5" serde = { version = "1", features = ["derive"] } rbs = { version = "4.1"} rbatis = { version = "4.1"} rbdc-sqlite = { version = "4.1" }
```
```rust //#[macro_use] define in 'root crate' or 'mod.rs' or 'main.rs'
extern crate rbatis; extern crate rbdc; use rbatis::rbdc::datetime::DateTime;
pub struct BizActivity {
pub id: Option
implselect!(BizActivity{selectallbyid(id:&str,name:&str) => "where id = #{id} and name = #{name}"});
implselect!(BizActivity{selectbyid(id:String) -> Option => "where id = #{id} limit 1"});
implupdate!(BizActivity{updatebyname(name:&str) => "where id = 1"});
impldelete!(BizActivity {deletebyname(name:&str) => "where name= '2'"});
implselectpage!(BizActivity{selectpage(name:&str) => "where name != #{name}"});
async fn main() {
/// enable log crate to show sql logs
fastlog::init(fastlog::Config::new().console()).expect("rbatis init fail");
/// initialize rbatis. also you can call rb.clone(). this is an Arc point
let rb = Rbatis::new();
/// connect to database
// sqlite
rb.init(SqliteDriver {}, "sqlite://target/sqlite.db").unwrap();
// mysql
// rb.init(MysqlDriver{},"mysql://root:123456@localhost:3306/test").unwrap();
// postgresql
// rb.init(PgDriver{},"postgres://postgres:123456@localhost:5432/postgres").unwrap();
// mssql/sqlserver
// rb.init(MssqlDriver{},"jdbc:sqlserver://localhost:1433;User=SA;Password={TestPass!123456};Database=test").unwrap();
let activity = BizActivity {
id: Some("2".into()),
name: Some("2".into()),
pc_link: Some("2".into()),
h5_link: Some("2".into()),
pc_banner_img: None,
h5_banner_img: None,
sort: None,
status: Some(2),
remark: Some("2".into()),
create_time: Some(DateTime::now()),
version: Some(1),
delete_flag: Some(1),
};
let data = BizActivity::insert(&mut rb, &activity).await;
println!("insert = {:?}", data);
let data = BizActivity::select_all_by_id(&mut rb, "1", "1").await;
println!("select_all_by_id = {:?}", data);
let data = BizActivity::select_by_id(&mut rb, "1".to_string()).await;
println!("select_by_id = {:?}", data);
let data = BizActivity::update_by_column(&mut rb, &activity, "id").await;
println!("update_by_column = {:?}", data);
let data = BizActivity::update_by_name(&mut rb, &activity, "test").await;
println!("update_by_name = {:?}", data);
let data = BizActivity::delete_by_column(&mut rb, "id", &"2".into()).await;
println!("delete_by_column = {:?}", data);
let data = BizActivity::delete_by_name(&mut rb, "2").await;
println!("delete_by_column = {:?}", data);
let data = BizActivity::select_page(&mut rb, &PageRequest::new(1, 10), "2").await;
println!("select_page = {:?}", data);
} ///...more usage,see crud.rs ```
pub async fn main() {
use rbatis::Rbatis;
use rbdcsqlite::driver::SqliteDriver;
#[derive(Clone, Debug, serde::Serialize, serde::Deserialize)]
pub struct BizActivity {
pub id: Option
Because of the compile time, the annotations need to declare the database type to be used.
rust
#[py_sql("select * from biz_activity where delete_flag = 0
if name != '':
`and name=#{name}`")]
async fn py_sql_tx(rb: &Rbatis, tx_id: &String, name: &str) -> Vec<BizActivity> { impled!() }
Because of the compile time, the annotations need to declare the database type to be used
html
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
"https://raw.githubusercontent.com/rbatis/rbatis/master/rbatis-codegen/mybatis-3-mapper.dtd">
<mapper>
<select id="select_by_condition">
`select * from biz_activity where `
<if test="name != ''">
name like #{name}
</if>
</select>
</mapper>
```rust ///select page must have '?:&PageRequest' arg and return 'Page'
async fn selectbycondition(rb: &mut dyn Executor, page_req: &PageRequest, name: &str) -> Page
```rust use once_cell::sync::Lazy;
pub static RB: Lazy
/// Macro generates execution logic based on method definition, similar to @select dynamic SQL of Java/Mybatis
/// RB is the name referenced locally by Rbatis, for example DAO ::RB, com:: XXX ::RB... Can be
/// The second parameter is the standard driver SQL. Note that the corresponding database parameter mysql is? , pg is $1...
/// macro auto edit method to 'pub async fn select(name: &str) -> rbatis::core::Result
pub async fn select(rb: &Rbatis, name: &str) -> BizActivity {}
//or： pub async fn select(name: &str) -> rbatis::core::Result
pub async fn testmacro() { fastlog::init(fast_log::Config::new().console()).expect("rbatis init fail"); RB.link("mysql://root:123456@localhost:3306/test").await.unwrap(); let a = select(&RB, "1").await.unwrap(); println!("{:?}", a); } ```
Rely on rbatis-codegen to create the source code of the corresponding structure from the html file at compile time (with debugmode(Cargo.toml- rbatis = { features = ["debug_mode"]}) enabled, you can observe the code-generated function), and call the generated method directly at run time.
We know that compilation is generally divided into three steps, lexes, syntactic analysis, semantic analysis, and intermediate code generation. In rbatis,
Lexical analysis is handled by the dependent func.rs in rbatis-codegen, which relies on syn and quote.
Parsing is done by parserhtml and parserpysql in rbatis-codegen
The generated syntax tree is a structure defined in the syntaxtree package in rbatis-codegen
Intermediate code generation has func.rs generation function, all supported functions are defined in rbatis-codegen
What is described above occurs during the cargo build phase, which is the compilation phase of the rust procedural macro, where the code generated by rbatis-codegen is handed back to the rust compiler for LLVM compilation to produce pure machine code
So I think rbatis is Truly zero overhead dynamic SQL compile-time ORM.
You are welcome to submit the merge, and make sure that any functionality you add has the appropriate mock unit test function added under the test package.
捐赠
联系方式(添加好友请备注'rbatis') 微信群：先加微信，然后拉进群
