๐Ÿ‘ฝRust Backend

Implementing New Handlers with Data Validation

When implementing a new handler in Rust, particularly for data processing or API request handling, it is crucial to ensure that the input data is valid and meets the expected criteria. This is where data validation comes into play.

The validator crate in Rust is an excellent tool for this purpose. It offers a comprehensive suite of validators that can be readily used to enforce various constraints on your data structures. From checking string lengths to validating numerical ranges and patterns, validator covers a wide array of common validation needs.

Utilizing the validator Crate

Hereโ€™s how you can integrate the validator crate into your Rust application:

  1. Integration: Include the validator crate in your Cargo.toml to get started. This crate provides the Validate trait that you can derive for your structs.

  2. Applying Validators: Use the provided validators to annotate your struct fields. For instance, to ensure a string field is not empty, you can use validators like length(min = "1").

  3. Custom Validators: For more specific or complex validation logic that isn't covered by the built-in validators, you can define custom validators. For example, required_not_empty_str is a custom validator defined in lib_infra::validator_fn.

Example: Validating Handler Input Data

In the context of your Rust application, consider the following struct representing data to be imported into AppFlowy:

use flowy_derive::ProtoBuf;
use validator::Validate;

#[derive(ProtoBuf, Validate, Default)]
pub struct ImportAppFlowyDataPB {
  #[pb(index = 1)]
  #[validate(custom = "lib_infra::validator_fn::required_not_empty_str")]
  pub path: String,

  #[pb(index = 2, one_of)]
  pub import_container_name: Option<String>,
}

In this example:

  • The ImportAppFlowyDataPB struct is defined with fields that need validation.

  • The path field is validated using a custom validator required_not_empty_str to ensure it's a non-empty string.

  • The import_container_name field is optional and does not require such strict validation.

Handling the Validation in Handlers

When writing a handler function, such as import_appflowy_data_folder_handler, you should include validation logic to ensure that incoming data meets the expected criteria before proceeding with the handlerโ€™s core functionality:

pub async fn import_appflowy_data_folder_handler(
    data: AFPluginData<ImportAppFlowyDataPB>,
) -> Result<(), FlowyError> {
    // Perform validation
    let data = data.try_into_inner()?;

    // Handler logic goes here
    Ok(())
}

In the handler:

  • The try_into_inner method is called on the data object. This method automatically performs all the validations defined in the struct.

  • If validation fails, an error is returned. This error can be converted or logged according to your application's error handling strategy.

Last updated