TA-Lib

This is a Zig wrapper for TA-LIB.

It requires the TA-Lib C library, version 0.8.3, to be installed (for example, brew install ta-lib, or from the release tarball).

Installation

1. Add ta-lib-zig as a dependency in your build.zig.zon:

zig fetch --save "git+https://github.com/ta-lib/ta-lib-zig#main"

2. In your build.zig, add the ta-lib module as a dependency you your program:

const ta_lib = b.dependency("ta_lib", .{
    .target = target,
    .optimize = optimize,
});

// the executable from your call to b.addExecutable(...)
exe.root_module.addImport("ta_lib", ta_lib.module("ta_lib"));

Examples

A simple example using the Momentum indicator:

const ta = @import("ta_lib");

const prices = [_]f64{ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0 };

const result = try ta.MOM(allocator, &prices, 5);
defer allocator.free(result);

Streaming

Each indicator also has a streaming type in ta.stream, which warms up on some history and then evaluates one bar at a time, giving the same values as the batch function over the same series:

var rsi = try ta.stream.RSI.init(&history, 14);
defer rsi.deinit();

const latest = try rsi.update(new_close); // commit a closed bar
const preview = try rsi.peek(live_price); // what-if, commits nothing

initAndFill also returns the batch output over the history. Indicators with several outputs return an Output struct from update, peek and value.

Regenerating the wrappers

src/func.zig (batch functions), src/stream.zig (streaming types) and the function aliases in src/root.zig are generated from TA-Lib's ta_func.h; the rest of src/ is hand-written. After upgrading TA-Lib, regenerate them with:

python3 tools/generate_func.py /path/to/include/ta-lib/ta_func.h