khorolets/diesel-derive-enum

Use enums with Diesel ORM

diesel-derive-enum

Use Rust enums directly with diesel ORM.

The latest release, 1.0.0, is tested against diesel 1.4 and rustc 1.39.0. It may work with older versions.

Note: The current master branch tracks diesel master, and will not work with diesel 1.x.

Example usage:

Cargo.toml

[dependencies]
diesel-derive-enum = { version = "1", features = ["..."] } # "postgres", "mysql" or "sqlite"

Rust

// define your enum
#[derive(DbEnum)]
pub enum MyEnum {
    Foo,  // All variants must be fieldless
    Bar,
    BazQuxx,
}

// define your table
table! {
    use diesel::types::Integer;
    use super::MyEnumMapping;
    my_table {
        id -> Integer,
        some_enum -> MyEnumMapping, // Generated Diesel type - see below for explanation
    }
}

// define a struct with which to populate/query the table
#[derive(Insertable, Queryable, Identifiable, Debug, PartialEq)]
#[table_name = "my_table"]
struct  MyRow {
    id: i32,
    some_enum: MyEnum,
}

SQL

Postgres:

-- by default the postgres ENUM values correspond to snake_cased Rust enum variant names
CREATE TYPE my_enum AS ENUM ('foo', 'bar', 'baz_quxx');

CREATE TABLE my_table (
  id SERIAL PRIMARY KEY,
  some_enum my_enum NOT NULL
);

MySQL:

CREATE TABLE my_table (
    id SERIAL PRIMARY KEY,
    my_enum enum('foo', 'bar', 'baz_quxx') NOT NULL  -- note: snake_case
);

sqlite:

CREATE TABLE my_table (
    id SERIAL PRIMARY KEY,
    my_enum TEXT CHECK(my_enum IN ('foo', 'bar', 'baz_quxx')) NOT NULL   -- note: snake_case
);

Now we can insert and retrieve MyEnum directly:

let data = vec![
    MyRow {
        id: 1,
        some_enum: MyEnum::Foo,
    },
    MyRow {
        id: 2,
        some_enum: MyEnum::BazQuxx,
    },
];
let connection = PgConnection::establish(/*...*/).unwrap();
let inserted = insert_into(my_table::table)
    .values(&data)
    .get_results(&connection)
    .unwrap();
assert_eq!(data, inserted);

Postgres arrays work too! See this example.

Enums Explained

Enums work slightly differently in each of the three databases.

• In Postgres, one declares an enum as a separate type within a schema, which may then be used in multiple tables. Internally, an enum value is encoded as an int (four bytes) and stored inline within a row - a much more efficient representation than a string.
• MySQL is similar except the enum is not declared as a separate type and is 'local' to it's parent table. It is encoded as either one or two bytes.
• sqlite does not have enums - in fact, it does not really have types; you can store any kind of data in any column. Instead we emulate static checking by adding the CHECK command, as per above. This does not give a more compact encoding but does ensure data integrity. Note that if you somehow retreive some other invalid text as an enum, diesel will error at the point of deserialization.

Type renaming

Diesel maintains a set of internal types which correspond one-to-one to the types available in various relational databases. Each internal type in turn maps to some kind of Rust native type. e.g. diesel::types::Integer maps to i32. So, when we create a new type in Postgres with CREATE TYPE ..., we must also create a corresponding type in Diesel, and then create a mapping to some native Rust type (our enum). Hence there are three types we need to be aware of.

By default, the Postgres and Diesel internal types are inferred from the name of the Rust enum. Specifically, we assume MyEnum corresponds to my_enum in Postgres and MyEnumMapping in Diesel. (The Diesel type is created by the plugin, the Postgres type must be created in SQL).

These defaults can be overridden with the attributes #[PgType = "..."] and #[DieselType = "..."]. (The PgType annotation has no effect on MySQL or sqlite).

Similarly, by default we assume that the possible ENUM variants are simply the Rust enum variants translated to snake_case. These can be renamed with the inline annotation #[db_rename = "..."].

See this test for an example of renaming.

You can override the snake_case assumption for the entire enum using the #[DbValueStyle = "..."] attribute. Individual variants can still be renamed using #[db_rename = "..."].

DbValueStyle	Variant	Value
camelCase	BazQuxx	"bazQuxx"
kebab-case	BazQuxx	"baz-quxx"
PascalCase	BazQuxx	"BazQuxx"
SCREAMING_SNAKE_CASE	BazQuxx	"BAZ_QUXX"
snake_case	BazQuxx	"baz_quxx"
verbatim	Baz__quxx	"Baz__quxx"

See this test for an example of changing the output style.

print-schema and infer-schema!

The print-schema command (from diesel_cli) attempts to connect to an existing DB and generate a correct mapping of Postgres columns to Diesel internal types. If a custom ENUM exists in the database, Diesel will simply assume that the internal mapping type is the ENUM name, Title-cased (e.g. my_enum -> My_enum). Therefore the derived mapping name must also be corrected with the DieselType attribute e.g. #[DieselType = "My_enum"].

If you are using a diesel.toml file to generate your schema.rs file, you should be sure to add an extra import for the enum mapping type, i.e. it should look something like the following:

# In diesel.toml

[print_schema]
file = "src/schema.rs"
import_types = ["diesel::sql_types::*", "crate::my_enum::*"]   # <- note the extra import

Unfortunately the infer_schema! is not compatible with this crate.

License

Licensed under either of these:

• Apache License, Version 2.0, (LICENSE-APACHE or https://www.apache.org/licenses/LICENSE-2.0)
• MIT license (LICENSE-MIT or https://opensource.org/licenses/MIT)