Parse your first document

In this tutorial you will parse a known PDF with the prebuilt document-to-markdown parser, inspect its Markdown, and find the same run in the Web UI.

Before you begin

Use a supported macOS or Linux installation. This tutorial runs from a ParseHawk repository checkout because it uses the bundled receipt fixture.

1. Start ParseHawk

parsehawk start

Wait for the services to become ready. If startup fails, run parsehawk doctor and follow the reported fix.

2. Parse the bundled receipt

parsehawk parse tests/fixtures/receipt/receipt.pdf \
  --wait \
  --output receipt.md

The command uploads the PDF, creates a parse job, waits for it with a 600-second client deadline, and writes the complete-document Markdown. It selects the read-only document-to-markdown parser because --parser is omitted.

The final line should confirm:

Wrote Markdown output: receipt.md

3. Inspect the Markdown

Confirm that the output is not empty, then print its first lines:

test -s receipt.md
sed -n '1,40p' receipt.md

The text and exact Markdown can vary with the model. You should see the receipt content in readable Markdown instead of a fixed JSON object.

List the job that produced it:

parsehawk parse-jobs list --parser document-to-markdown

The completed job contains both result.content for the complete document and result.pages for one-based source pages. ParseHawk joins multiple pages with an HTML page-break comment.

4. Inspect the same run in the Web UI

Open http://127.0.0.1:5173, select Parse, and open the newest parse job. You can switch between these result views:

• rendered complete-document Markdown
• rendered Markdown for one source page
• raw Markdown

The result panel also shows the provider, model, internal adapter, duration, and job state. You can copy or download the Markdown from the same page.

What you built

You exercised the complete parsing path:

document → uploaded file → parse job → page images → Markdown

Parsing keeps the document’s content and page structure. It does not select a predefined set of business fields. Compare the two output models in parsing and extraction.

Next, customize Markdown parsing or run parsing and extraction through the REST API.