Crate grumpydb 

GrumpyDB — A disk-based object storage engine.

GrumpyDB stores schema-less documents (JSON-like) on disk with B+Tree indexing, page-based storage, WAL for durability, and SWMR concurrency.

Example

use grumpydb::{Database, Value};
use uuid::Uuid;
use std::collections::BTreeMap;

let mut db = Database::open(std::path::Path::new("./my_database")).unwrap();
db.create_collection("docs").unwrap();

let key = Uuid::new_v4();
let value = Value::Object(BTreeMap::from([
    ("name".into(), Value::String("GrumpyDB".into())),
    ("version".into(), Value::Integer(1)),
]));

db.insert("docs", key, value).unwrap();

let doc = db.get("docs", &key).unwrap();
assert!(doc.is_some());
db.close().unwrap();

Re-exports

pub use concurrency::lock_manager::SharedDb;Deprecated

pub use concurrency::shared::SharedDatabase;

pub use concurrency::shared::SharedReadTx;

pub use concurrency::shared::SharedServer;

pub use database::Database;

pub use database::ReadTx;

pub use document::value::Value;

pub use engine::CompactResult;

pub use engine::GrumpyDb;Deprecated

pub use error::GrumpyError;

pub use error::Result;

pub use index::IndexDefinition;

pub use server::GrumpyServer;

pub use server::client::Client;

Modules

btree

Generic B+Tree index.

buffer

Buffer pool: LRU cache for pages in memory.

collection

Collection: a named unit of document storage.

concurrency

Concurrency: SWMR (Single-Writer, Multi-Reader) thread-safe access.

database

Database: manages multiple named collections with a shared WAL.

document

Document model: schema-less JSON-like values with binary codec.

engine

Storage engine: orchestrates all subsystems to provide CRUD operations.

error

index

Secondary index: maps field values to document UUIDs via a BTree<Vec<u8>>.

naming

Name validation for clients, databases, collections, and indexes.

page

Page management: constants, types, and page I/O.

server

Server: multi-tenant management of clients and databases.

wal

Write-Ahead Log for crash recovery and durability.