olive

KO | EN

gitlite — search

#python #ai #typescript #machine-learning #rust #deep-learning #linux #macos #nodejs #windows #kubernetes #security #openai #open-source

(★ 14)

The Dream Programming Language

olive Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

## Overview

**A general-purpose systems language that's easy to read, fast to run, and keeps your memory safe.**

Olive was built for when you want the speed of a low-level language without the headache of complex syntax. It uses a clean, indentation-based structure and a smart ownership model to provide consistent performance without a garbage collector.

## Why Olive?

- **Clean Syntax**: No braces, no semicolons. Indentation defines the structure, keeping your code readable and consistent.
- **Memory Safety, No Annotations**: The compiler infers ownership and frees values deterministically; references are borrow-checked, and a generation-checked allocator turns anything the analysis can't prove into a precise runtime fault instead of corruption. No garbage collector, no lifetime syntax.
- **Blazing Fast**: Optimized to native code via the Cranelift backend. It's designed to run close to the metal with zero-cost abstractions.
- **Modern Concurrency**: True async/await that's easy to use and extremely efficient.
- **C / Rust Interop**: Interface with C or Rust libraries through a C-compatible ABI with built-in FFI support.
- **Python Interop**: Import any Python module directly. Types are inferred automatically from `.pyi` stubs, so `glm.vec3` stays `glm.vec3` through the type checker. Native collections pass with zero-copy bidirectional proxies.
- **Friendly Errors**: When things go wrong, the compiler tells you exactly where and why, with suggestions on how to fix it.

## A Taste of Olive

```rust
// A generic function to calculate average
fn average[T: Numeric](numbers: [T]) -> float:
    let mut total = 0.0
    for n in numbers:
        total += float(n)
    return total / float(len(numbers))

async fn process_data(data: [int]):
    print(f"Processing {len(data)} items...")
    let avg = average(data)
    print(f"Result: {avg:.2f}")

fn main():
    let data = [10, 20, 30, 40, 50]
    // Spawning an async task
    async:
        await process_data(data)
```

## Getting Started

**Linux and macOS:**

```bash
curl -sSL https://raw.githubusercontent.com/ecnivslabs/olive/master/install.sh | sh
```

**Windows:** download from the [releases page](https://github.com/ecnivslabs/olive/releases/latest).

Then:

```bash
pit new my_app
cd my_app
pit run
```

## Documentation

- [Introduction](docs/introduction.md): Philosophy and goals.
- [Basics](docs/basics.md): Variables, types, and control flow.
- [Ownership](docs/ownership.md): How memory safety works.
- [Generics](docs/generics.md): Writing reusable code.
- [Traits](docs/traits.md): Defining shared behavior between types.
- [C / Rust Interop (FFI)](docs/ffi.md): Calling C or Rust code and using `unsafe`.
- [Python Interop](docs/python.md): Typed Python integration with automatic `.pyi` stub introspection.
- [Standard Library](docs/modules.md): What's in the box.
- [Full Index](docs/index.md): Everything in one place.

## Contributing

Contributions are welcome! Fork the repo, make a branch, and open a PR. Keep it simple, keep it clean.

olive

olive Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

olive

olive Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README