The Collection Trait

The easiest and most type-safe way to work with Deeb is through the Collection trait. This allows you to define your data structures as Rust structs while giving you full access to Deeb’s database operations—without sacrificing compile-time safety.

If you prefer a looser, dynamic approach, check out the Programmatic Usage Guide.

Example

#[derive(Collection, Serialize, Deserialize)]
#[deeb(
    name = "user",
    primary_key = "id",
    associate = ("comment", "id", "user_id", "user_comment")
)]
struct User {
    id: i32,
    name: String,
    age: i32,
}

This defines a User collection backed by a JSON file and adds a relationship to a Comment collection. With this setup, you can immediately start performing operations like:

User::find_one(db, query, transaction)
User::find_many(db, query, find_options, transaction)
User::insert(db, document, transaction) // Use `insert_many` in v0.0.12+
User::insert_many(db, vec![documents], transaction)
User::update_one(db, query, update_doc, transaction)
User::update_many(db, query, update_doc, transaction)
User::delete_one(db, query, transaction)
User::delete_many(db, query, transaction)

⚠️ insert has been replaced with insert_many in version v0.0.12.

ℹ️ find_many accepts an optional FindManyOptions struct, allowing you to skip documents, limit results, or sort by fields:

#[derive(Serialize, Deserialize, Clone, Debug, PartialEq)]
pub struct FindManyOptions {
    pub skip: Option<i32>,
    pub limit: Option<i32>,
    pub order: Option<Vec<FindManyOrder>>,
}

Querying Your Data

Deeb includes a flexible query system that allows you to find documents by any field, using nested and compound logic. Learn more about how to construct powerful queries in the Query Guide.

Transactions

All operations can be wrapped in transactions, allowing you to commit or roll back changes as a unit. For more info, visit the Transactions Guide.

📚 For full API reference and deeper technical details, visit the Deeb Docs on docs.rs .