# orm_mysql

**Repository Path**: guoyucode/orm_mysql

## Basic Information

- **Project Name**: orm_mysql
- **Description**: orm 库 rust 实现
- **Primary Language**: Rust
- **License**: BSD-3-Clause
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2022-12-16
- **Last Updated**: 2025-11-30

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# orm_mysql

Committed to building a high-performance, lightweight orm library; Try to achieve zero copy;
>致力于构建高性能、轻量级的orm库; 尽量做到零拷贝;

The function is being developed rapidly. Those who are interested can participate in the development together; 
>功能正在加紧开发中, 有兴趣的可以一起参与开发; 

Join us: WeChat: wx_essence, Telegram: tg_essence;
>加入我们: 微信: wx_essence, Telegram: tg_essence;

[examples mysql_all](https://github.com/guoyucode/orm_mysql/blob/main/examples/mysql_all.rs)<br>
[examples mysql_insert](https://github.com/guoyucode/orm_mysql/blob/main/examples/mysql_insert.rs)<br>
[examples mysql_query](https://github.com/guoyucode/orm_mysql/blob/main/examples/mysql_query.rs)<br>

examples:

    add Cargo.toml dependencies
    mysql_async = "0.31"
    orm_mysql = "*"

    
    use orm_mysql::mysql_async::prelude::*;
    use orm_mysql::mysql::OrmMySqlTrait;
    use orm_mysql::OrmMySql;

    #[tokio::main]
    async fn main() -> common_uu::IResult {
        let ref pool: mysql_async::Pool = mysql_async::Pool::new("mysql://username:pwd@ip:port/db_name");
        let mut conn: mysql_async::Conn = pool.get_conn().await?;

        r"DROP TABLE if exists users_temp".ignore(&mut conn).await?;

        r"CREATE TABLE users_temp (
            user_id int   not null,
            username      varchar(128) null,
            username2      varchar(128) null,
            username3      varchar(128) null,
            username4      varchar(128) null,
            username5      varchar(128) null,
            username6      varchar(128) null,
            username7      varchar(128) null,
            username8      varchar(128) null,
            username9      varchar(128) null,
            username10      varchar(128) null,
            username11      varchar(128) null,
            username12      varchar(128) null,
            username13      varchar(128) null,
            username14      varchar(128) null,
            username15      varchar(128) null,
            float_v      float(8,2) null
        )"
        .ignore(&mut conn)
        .await?;

        // use conn insert 
        // let r: Option<i64>=conn.exec_first("insert into users_temp (user_id, username, username2)values(?,?,?)", (1, "11".to_string(), Some("111".to_string()))).await?;
        let mut user = UserData::default();
        user.user_id = 1;
        user.username = "11".to_string();
        user.username2 = Some("111".to_string());
        user.insert(&mut conn).await?;

        // use transaction
        let mut tx = pool.start_transaction(mysql_async::TxOpts::new()).await?;
        let mut user = UserData::default();
        user.user_id = 2;
        user.username = "22".to_string();
        user.username2 = Some("222".to_string());
        user.insert(&mut tx).await?;
        tx.commit().await?;

        // let r: Option<UserData> = conn.query_first("select * from users_temp").await?;
        let r: Option<UserData> = UserData::query_first(&mut conn, "where 1=1").await?;
        println!(r##"UserData::query_first {:?}"##, r);

        // let r: Vec<UserData> = conn.query("select * from users_temp").await?;
        let r: Vec<UserData> = UserData::query(&mut conn, "where 1=1", Some(99)).await?;
        println!("UserData::query: {:?}", r);

        // Auto Increment Example (Pseudo-code)
        // If your struct has #[orm_mysql(id, auto_increment)]
        // let mut user = UserAutoIncrement::default();
        // user.username = "test".to_string();
        // let id = user.insert(&mut conn).await?; // Returns the generated ID

        Ok(())
    }

    // #[derive(Default, Debug)]
    #[derive(OrmMySql, Default, Debug)]
    #[orm_mysql(table_name=users_temp)] // is not config: table_name => user_data
    struct UserData {
        #[orm_mysql(id)]
        user_id: i64,
        username: String,
        username2: Option<String>,
        username3: Option<String>,
        username4: Option<String>,
        username5: Option<String>,
        username6: Option<String>,
        username7: Option<String>,
        username8: Option<String>,
        username9: Option<String>,
        username10: Option<String>,
        username11: Option<String>,
        username12: Option<String>,
        username13: String,
        username14: String,
        username15: String,
        float_v: f64,
        
        datetime_v: chrono::NaiveDateTime,
        datetime_opt: Option<chrono::NaiveDateTime>,

        date: chrono::NaiveDate,
        date_opt: Option<chrono::NaiveDate>,

        // not support; 暂时不支持
        // datetime_utc_v: Option<chrono::DateTime<chrono::Utc>>,
        // datetime_local_v: chrono::DateTime<chrono::Local>,
        // naive_date_v: time::PrimitiveDateTime,
        // naive_date_opt_v: Option<time::Date>,

    }

---

## TODO LIST (Essential APIs)

### 🚀 Must-Have Features (核心缺失功能)

#### 1. ID-Based Operations (基于ID的操作)
- [x] `find_by_id(id)` - Retrieve a single record directly by ID.
  ```rust
  // Find by ID (requires #[orm_mysql(id)] on the id field or a field named `id`)
  let user: Option<UserData> = UserData::find_by_id(&mut conn, 1).await?;
  
  // For String id type:
  let user: Option<UserData> = UserData::find_by_id(&mut conn, "uuid-xxx".to_string()).await?;
  ```
- [x] `delete_by_id(id)` - Delete a record directly by ID without loading it.
  ```rust
  let affected = UserData::delete_by_id(&mut conn, 1).await?;
  ```

#### 2. Partial Updates (部分更新)
Currently, `update()` overwrites ALL fields, which is inefficient and risky for concurrency.
- [ ] `update_columns(id, updates)` - Update specific columns only.
  - *Requirement:* Allow updating just `status` or `last_login` without touching other fields.

#### 3. Aggregation & Checks (聚合与检查)
Currently, users must fetch all records to count them, which is a performance killer.
- [ ] `count(where_sql)` - Efficiently count records matching a condition.
  - *Current workaround:* `UserData::query(...).len()` (Very bad performance)
- [ ] `exists(where_sql)` - Check if a record exists without loading data.

#### 4. Safe Query Parameters (安全查询参数)
Currently, `where_sql` is a raw string, encouraging dangerous string concatenation.
- [ ] `query_with_params(where_sql, params)` - Support parameterized queries to prevent SQL injection.
  - *Goal:* `UserData::query_with_params("where name = ?", vec!["Alice"])`

---