Quick Start

We're going to build a simple text processing pipeline. By the end, you'll understand how pipelines work and be able to create your own.

What We're Building

A pipeline that takes text, reverses it, and converts to uppercase. Simple, but it shows how commands chain together.

Create the Project

mkdir my-pipeline
cd my-pipeline
divvun-runtime init

You now have pipeline.ts and a .divvun-rt/ directory with TypeScript types.

What Just Happened

Open pipeline.ts and you'll see:

import { Command, StringEntry } from './.divvun-rt/mod.ts';
import * as example from './.divvun-rt/example.ts';

export default function examplePipeline(entry: StringEntry): Command {
    let x = example.reverse(entry);
    x = example.upper(x);
    return x;
}

Breaking this down: - StringEntry is your input type (a string) - entry is the actual input text - example.reverse(entry) reverses the string - example.upper(x) uppercases it - The function returns the result

The example module is just for learning. Real pipelines use hfst, cg3, divvun, and speech.

Run It

divvun-runtime run ./pipeline.ts "Hello World"

Output:

DLROW OLLEH

The text flows through: input → reverse → uppercase → output.

Modify It

Change the pipeline to just uppercase:

export default function myPipeline(entry: StringEntry): Command {
    return example.upper(entry);
}

Run it again - now it just uppercases without reversing.

Use Real Modules

The example module is for learning. Real pipelines look like this:

import { StringEntry, Command } from './.divvun-rt/mod.ts';
import * as hfst from './.divvun-rt/hfst.ts';
import * as cg3 from './.divvun-rt/cg3.ts';
import * as divvun from './.divvun-rt/divvun.ts';

export default function grammarCheck(entry: StringEntry): Command {
    let x = hfst.tokenize(entry, { model_path: "tokeniser.pmhfst" });
    x = cg3.vislcg3(x, { model_path: "grammar.bin" });
    return divvun.suggest(x, { model_path: "generator.hfstol" });
}

This tokenizes text, applies grammar rules, and generates error suggestions. You'd need to add the model files to assets/ for this to work.

What You Learned

• Pipelines are TypeScript functions
• Commands chain together: one's output becomes the next's input
• divvun-runtime run executes your pipeline
• .divvun-rt/ contains generated types for all available commands

See Pipelines for more details, or jump to Grammar Checking or Text-to-Speech to build something real.