12 unstable releases (3 breaking)

new 0.4.4 Sep 18, 2020
0.4.3 Sep 16, 2020
0.3.2 Sep 9, 2020
0.2.2 Sep 3, 2020
0.1.1 Sep 2, 2020

#79 in Database interfaces

Download history 60/week @ 2020-08-31 73/week @ 2020-09-07 94/week @ 2020-09-14

68 downloads per month

MIT license

97KB
939 lines

Aragog

pipeline status MIT licensed Crates.io aragog

aragog is a simple lightweight ODM library for ArangoDB using the arangors driver. The main concept is to provide behaviors allowing to synchronize documents and structs as simply an lightly as possible. In the future versions aragog will also be able to act as a ORM and OGM for ArangoDB

Features

By now the available features are:

  • Creating a database connection pool from a defined schema.json
  • Structures can implement different behaviors:
    • Record: The structure can be written into a ArangoDB collection as well as retrieved, from its _key or other query arguments.
    • New: The structure can be initialized from an other type (a form for example). It allows to maintain a privacy level in the model and to use different data formats.
    • Update: The structure can be updated from an other type (a form for example). It allows to maintain a privacy level in the model and to use different data formats.
    • Validate: The structure can perform simple validations before being created or saved into the database.
    • Authenticate: The structure can define a authentication behaviour from a secret (a password for example) (see password_hashing section)
    • AuthorizeAction: The structure can define authorization behavior on a target record with custom Action type.
  • Different operations can return a ServiceError error that can easily be transformed into a Http Error (can be used for the actix framework)

Cargo features

Actix Http Error

If you use this crate with the actix framework, you may want the aragog errors to be usable as http errors. To do so cou can add to your cargo.toml the following feature: actix_http_error.

aragog = { version = "^0.4", features = ["actix_http_error"] }
Password hashing

You may want aragog to provide a more complete Authenticate trait allowing to hash and verify passwords. To do so cou can add to your cargo.toml the following feature: password_hashing.

aragog = { version = "^0.4", features = ["password_hashing"] }

It will add two functions in the Authenticate trait:

fn hash_password(password: &str, secret_key: &str) -> Result<String, ServiceError>;
fn verify_password(password: &str, password_hash: &str, secret_key: &str) -> Result<(), ServiceError>;
  • hash_password will return a Argon2 encrypted password hash you can safely store to your database
  • verify_password will check if the provided password matches the Argon2 encrypted hash you stored.

The Argon2 encryption is based on the argonautica crate. That crate requires the clang lib, so if you deploy on docker you will need to install it or define a custom image.

Schema and collections

In order for everything yo work you need to specify a schema.json file. The path of the schema must be set in SCHEMA_PATH environment variable or by default the pool will look for it in src/config/db/schema.json.

There is an example schema.json file in /examples/simple_food_order_app

The json must look like this:

{
  "collections": [
    {
      "name": "collection1",
      "indexes": []
    },
    {
      "name": "collection2",
      "indexes": [
        {
          "name": "byUsernameAndEmail",
          "fields": ["username", "email"],
          "settings": {
            "type": "persistent",
            "unique": true,
            "sparse": false,
            "deduplicate": false
          }
        } 
      ]
    }
  ]
}

When initializing the DatabaseConnectionPool every collection name will be searched in the database and if not found the collection will be automatically created.

You don't need to create the collections yourself

Indexes

The array of Index in indexes must have that exact format:

  • name: the index name,
  • fields: an array of the fields concerned on that compound index,
  • settings: this json bloc must be the serialized version of an IndexSettings variant from arangors driver.

Database Record

The global architecture is simple, every Model you define that can be synced with the database must implement Record and derive from serde::Serialize, serde::Deserialize and Clone. If you want any of the other behaviors you can implement the associated trait

The final Model structure will be an Exact representation of the content of a ArangoDB document, so without its _key, _id and _rev. Your project should contain some models folder with every struct representation of your database documents.

The real representation of a complete document is DatabaseRecord<T> where T is your model structure.

Example:

use aragog::{Record, DatabaseConnectionPool, DatabaseRecord, Validate};
use serde::{Serialize, Deserialize};
use tokio;

#[derive(Serialize, Deserialize, Clone)]
pub struct User {
    pub username: String,
    pub first_name: String,
    pub last_name: String,
    pub age: usize
}

impl Record for User {
    fn collection_name() -> &'static str { "Users" }
}

impl Validate for User {
    fn validations(&self,errors: &mut Vec<String>) { }
}

#[tokio::main]
async fn main() {
    // Database connection Setup
    let database_pool = DatabaseConnectionPool::new("http://localhost:8529", "db", "user", "password").await;
    // Define a document
    let mut user = User {
        username: String::from("LeRevenant1234"),
        first_name: String::from("Robert"),
        last_name: String::from("Surcouf"),
        age: 18
    };
    // user_record is a DatabaseRecord<User>
    let mut user_record = DatabaseRecord::create(user, &database_pool).await.unwrap();
    // You can access and edit the document
    user_record.record.username = String::from("LeRevenant1524356");
    // And directly save it
    user_record.save(&database_pool).await;
}

Querying

You can retrieve a document from the database as simply as it gets, from the unique ArangoDB _key or from multiple conditions. The example below show different ways to retrieve records, look at each function documentation for more exhaustive explanations.

Example

let record = DatabaseRecord::create(user, &database_pool).await.unwrap();

// Find with the primary key
let user_record = User::find(&record.key, &database_pool).await.unwrap();

// Find a user with multiple conditions
let mut query = Query::new(QueryItem::field("last_name").equals_str("Surcouf")).and(QueryItem::field("age").greater_than(15));
let user_record = User::find_where(query, &database_pool).await.unwrap();

// Find all users with multiple conditions
let mut query = Query::new(QueryItem::field("last_name").like("%Surc%")).and(QueryItem::field("age").in_array(&[15,16,17,18]));
let user_records = User::get_where(query, &database_pool).await.unwrap();

The querying system hierarchy works this way:

Query::new(comparison_1).and(comparison_2).or(comparison_3)

Each comparison is a QueryItem built via QueryItemBuilder:

// for a simple field comparison
QueryItem::field("some_field").some_comparison("compared_value");
// for field veing arrays (see ArangoDB operators)
QueryItem::all("some_field_array").some_comparison("compared_value");
QueryItem::any("some_field_array").some_comparison("compared_value");
QueryItem::none("some_field_array").some_comparison("compared_value");

All the currently implemented comparison methods are listed under QueryItemBuilder documentation page.

TODO

  • Query system:
    • Simple and modular query system
    • Macros for lighter queries
    • Advanced query system supporting:
      • Arithmetic Operators
      • Array variant querying (ANY, NONE, ALL)
      • ArangoDB functions (LENGTH, ABS, etc.)
  • ORM and OGM
    • Pundit like authorizations (authorize actions on model)
    • Relations
      • Handle graph vertices and edges
      • Handle SQL-like relations (foreign keys)
    • Handle key-value pair system (redis like)
  • Middle and long term:
    • Handle revisions/concurrency correctly
    • Code Generation
      • Avoid string literals as collection names
      • Handle Migrations
    • Define possible async validations for database advance state check

Arango db setup

Installation (See official documentation [Here] arango_doc)

  • Download Link
  • Run it with /usr/local/sbin/arangod The default installation contains one database _system and a user named root
  • Create a user and database for the project with the arangosh shell
arangosh> db._createDatabase("DB_NAME");
arangosh> var users = require("@arangodb/users");
arangosh> users.save("DB_USER", "DB_PASSWORD");
arangosh> users.grantDatabase("DB_USER", "DB_NAME");

It is a good practice to create a test db and a development db.

  • you can connect to the new created db with
$> arangosh --server.username $DB_USER --server.database $DB_NAME

License

aragog is provided under the MIT license. See LICENSE. An simple lightweight ODM for ArangoDB based on arangors.

Special thanks to fMeow creator of arangors and inzanez

Dependencies

~5–10MB
~223K SLoC