Crate garde

A Rust validation library

• Basic usage example
• Validation rules
• Custom validation
• Implementing rules
• Implementing Validate
• Feature flags
• Why garde?

Basic usage example

To get started, use the Validate derive macro and add some validation rules to your type. This generates an implementation of the Validate trait for you. To use it, call the validate method on an instance of the type.

Here’s what that looks like in full:

use garde::{Validate, Valid};

#[derive(Validate)]
struct User<'a> {
    #[garde(ascii, length(min=3, max=25))]
    username: &'a str,
    #[garde(length(min=15))]
    password: &'a str,
}

let user = User {
    username: "test",
    password: "not_a_very_good_password",
};

if let Err(e) = user.validate(&()) {
    println!("invalid user: {e}");
}

Garde can also validate enums:

use garde::{Validate, Valid};

#[derive(Validate)]
enum Data {
    Struct {
        #[garde(range(min=-10, max=10))]
        field: i32,
    },
    Tuple(
        #[garde(ascii)]
        String
    ),
}

let data = Data::Struct { field: 100 };
if let Err(e) = data.validate(&()) {
    println!("invalid data: {e}");
}

Available validation rules

Additional notes:

• For length and range, either min or max may be omitted, but not both.
• length and range use an inclusive upper bound (min..=max).
• length uses .chars().count() for UTF-8 strings instead of .len().
• For contains, prefix, and suffix, the pattern must be a string literal, because the Pattern API is currently unstable.
• Nested validation using dive may not be combined with any other rule.

Custom validation

Validation may be customized via the custom rule and the context attribute.

The context may be any type without generic parameters. By default, the context is ().

#[derive(garde::Validate)]
#[garde(context(PasswordContext))]
struct User {
    #[garde(custom(is_strong_password))]
    password: String,
}

struct PasswordContext {
    min_entropy: f32,
    entropy: cracken::password_entropy::EntropyEstimator,
}

fn is_strong_password(value: &str, context: &PasswordContext) -> garde::Result {
    let bits = context.entropy.estimate_password_entropy(value.as_bytes())
        .map(|e| e.mask_entropy)
        .unwrap_or(0.0);
    if bits < context.min_entropy {
        return Err(garde::Error::new("password is not strong enough"));
    }
    Ok(())
}

let ctx = PasswordContext { /* ... */ };
let user = User { /* ... */ };
user.validate(&ctx)?;

The validator function may accept the value as a reference to any type which it derefs to. In the above example, it is possible to use &str, because password is a String, and String derefs to &str.

Implementing rules

Say you want to implement length checking for a custom string-like type. To do this, you would implement the rules::length::HasLength trait for it.

#[repr(transparent)]
pub struct MyString(pub String);

impl garde::rules::length::HasLength for MyString {
    fn length(&self) -> usize {
        self.0.chars().count()
    }
}
#[derive(garde::Validate)]
struct Foo {
    // Now the `length` check may be used with `MyString`
    #[garde(length(min = 1, max = 1000))]
    field: MyString,
}

Each rule comes with its own trait that may be implemented by custom types in your code. They are all available under rules.

Implementing Validate

In case you have a container type for which you’d like to support nested validation (using the #[garde(dive)] rule), you may manually implement Validate for it:

#[repr(transparent)]
struct MyVec<T>(Vec<T>);

impl<T: garde::Validate> garde::Validate for MyVec<T> {
    type Context = T::Context;

    fn validate(&self, ctx: &Self::Context) -> Result<(), garde::Errors> {
        garde::Errors::list(|errors| {
            for item in self.0.iter() {
                errors.push(item.validate(ctx));
            }
        })
        .finish()
    }
}

#[derive(garde::Validate)]
struct Foo {
  #[garde(dive)]
  field: MyVec<Bar>,
}

#[derive(garde::Validate)]
struct Bar {
  #[garde(range(min = 1, max = 10))]
  value: u32,
}

To make implementing the trait easier, the error::Errors type supports a nesting builders.

• For list-like or tuple-like data structures, use Errors::list, and its .push method to attach nested Errors.
• For map-like data structures, use Errors::fields, and its .insert method to attach nested Errors.
• For a “flat” error list, use Errors::simple, and its .push method to attach individual errors.

The ListErrorBuilder::push and ListErrorBuilder::insert methods will ignore any errors which are empty (via Errors::is_empty).

Feature flags