Skip to main content

Quick Start

Install the package for the ecosystem you use. Python is the shortest path for experiments and ingestion jobs, Node.js fits services and queues, Rust exposes the core API directly, and the CLI is useful for inspection.

pip install dongler
npm install @cristianexer/dongler
cargo install dongler

For Rust library usage, depend on dongler-core. The dongler crate is the CLI package.

Python​

Parse a PDF into all three output formats:

import dongler

doc = dongler.load("report.pdf")

print(doc.metadata["block_count"])
print(doc.to_markdown())
print(doc.to_latex())

to_dict() gives you the full document object for custom pipelines:

data = doc.to_dict()
for page in data["pages"]:
    print(page["number"], len(page["blocks"]))

TypeScript​

import { load, loadMany } from "@cristianexer/dongler";

const doc = load("report.pdf");

console.log(doc.metadata.block_count);
console.log(doc.toMarkdown());
console.log(doc.toLatex());

for (const result of loadMany(["report.pdf", "notes.txt"])) {
  if (!result.ok) {
    console.error(`${result.path}: ${result.error}`);
  }
}

Rust​

use dongler_core::load_path;

fn main() -> dongler_core::Result<()> {
    let doc = load_path("report.pdf")?;

    println!("blocks: {}", doc.metadata.block_count);
    println!("{}", doc.to_markdown()?);
    Ok(())
}

CLI​

dongler inspect report.pdf
dongler extract report.pdf --format markdown
dongler extract report.pdf --format latex
dongler extract report.pdf --format json

The CLI uses the same native extraction engine as the language bindings.

Next Steps​