Datahike Documentation

Welcome to the Datahike documentation! This index organizes all documentation by topic and user level.

🚀 Getting Started

New to Datahike? Start here:

• Main README - Project overview, installation, and quick start
• Why Datalog? - Query comparisons and when to use Datalog
• Configuration - Database configuration options
• Schema Flexibility - Schema-on-read vs schema-on-write

📚 Core Concepts

Essential concepts for working with Datahike:

• Time Variance - Time-travel queries (as-of, history, since), audit trails, and GDPR-compliant purging
• Storage Backends - Choosing the right backend for your needs (file, memory, PostgreSQL, etc.)
• Garbage Collection - Reclaim storage by removing old database snapshots
• Blobs and out-of-line values - :db.type/store-ref for objects you fetch, not query, kept alive by the collector (experimental)
• Entity Spec - Entity API and specifications
• Logging and Error Handling - Debugging and error management

🔧 Language Bindings (Beta)

Datahike supports multiple languages and platforms:

JVM Languages

• Java API - Fluent builder API, automatic collection conversion, comprehensive bindings
  • Examples: examples/java/
  • Status: Beta - Functional and tested, but may receive breaking changes

JavaScript/TypeScript

• JavaScript API - Promise-based API for Node.js and browsers
• Browser replicas - Datahike in the browser, writing through a server, with permissions
  • npm: npm install datahike@next
  • Status: Beta - Functional and tested, but may receive breaking changes

ClojureScript

• ClojureScript Support - Async operations, Node.js, IndexedDB, and browser backends
  • Backends: Memory, IndexedDB, TieredStore, File (Node.js)
  • Status: Beta - Functional, please try it out and provide feedback

Python

• Python Bindings - High-level Pythonic API with automatic EDN conversion
  • Status: Beta - Functional and tested, but may receive breaking changes

CLI & Shell

• CLI (dthk) - Native command-line tool compiled with GraalVM

  • Features: Instant startup, file backend, scriptable
  • Status: Beta - Functional and tested, but command structure may change

• Babashka Pod - Shell scripting with Datahike

  • Integration: Native pod for Babashka
  • Status: Beta - Functional but not yet used in production

Native Libraries

• libdatahike - C/C++ native bindings for embedding in non-JVM applications

  • Use cases: Embedding in C/C++/Rust/Go applications
  • Status: Beta - Functional and tested, but API may change

• EDN Conversion Rules - How data types map between languages

🏗️ Advanced Features

For experienced users building production systems:

• Compiled Query Engine - Fused scan execution, ORDER BY, query result cache, d/explain (beta; default engine)
• Secondary Indices - Full-text search, vector similarity, columnar aggregates (experimental)
• Distributed Architecture - Distributed Index Space and real-time sync with Kabel
• Reducing write amplification - Cut objects-per-commit on request-priced object stores: diff buffering, root fusion, commit-graph opt-out (experimental)
• Index warming - Cut a cold reader's misses × RTT on high-latency stores: budget-bounded breadth-first prefetch of index levels, fetched concurrently (experimental)
• Cross-database references - dh:// URIs and reified links across databases (living vs record references)
• Optimistic Overlay - Zero-latency UI updates over a remote writer via d/with (beta)
• Snapshot dependency tracking - Runtime tokens for caches of selected attributes (experimental)
• Versioning - Git-like branching and merging (beta)
• Migration - One record stream, three jobs: backup/restore via portable verifiable dumps, moving between storage backends, and reading from or writing to other systems via import-source / export-to-sink (beta)
• Migrating from Datomic - Datomic Pro in and out over that seam, with what survives a round trip and what cannot (experimental; may move out of datahike into its own library)
• Norms - Database migration system
• Unstructured Input Support - Schema inference from JSON/EDN (experimental)
• Graph Algorithms - Reachability, paths, centrality, community detection, flows, random walks over a GraphSpec (experimental)
• Anomaly Detection - Parameter-free ECOD outlier detection (experimental)

🔍 Reference

• Differences to Datomic - API compatibility and differences
• Benchmarking - Performance testing and optimization

🛠️ Development

Contributing to Datahike:

• Contributing Guide - How to contribute to the project
• Backend Development - Creating custom storage backends

---

Feature Maturity Levels

• Stable - Production-ready, API unlikely to change
• Beta - Tested and functional, but API may receive changes
• Experimental - Try it out, API likely to change significantly

Need Help?

• 💬 Discussions - Ask questions and share ideas
• 🐛 Issues - Report bugs or request features
• 📖 API Docs - Complete API reference on cljdoc