Downloads & PDFs

Web scraping is not always about extracting text. Frequently, you need to capture visual proofs (screenshots), generate formal reports (PDFs), or trigger file downloads (CSVs, MP4s, ZIPs) hosted on the site.

StepWright provides simple, one-line actions for handling Playwright's download and printing APIs.

---

📸 Capturing Screenshots

Full Page Storage

To grab the entire rendered DOM (even parts currently off-screen), use the screenshot action with data_type="full".

BaseStep(
    id="full_capture",
    action="screenshot",
    value="./storage/evidence.png",  # Relative or absolute Filepath destination
    data_type="full"                 # Required flag for full page
)

Element-Level Storage

You can perfectly crop an image to a specific DOM element, like a chart, a specific tweet, or an invoice table:

BaseStep(
    id="chart_capture",
    action="screenshot",
    object_type="id",
    object="quarterly-earnings-chart",
    value="./storage/chart.png"
)

---

🖨️ Generating PDF Documents

If you are scraping receipts, financial statements, or long articles, saving the result as a PDF is often cleaner than a rasterized image screenshot.

WARNING

PDF generation inside Python Playwright instances strictly requires the browser to be running in headless=True mode. You will receive an error if you attempt to use these actions while watching the browser visually.

savePDF (Standard Method)

Generates a straightforward PDF representation of the current page.

BaseStep(
    id="print_invoice",
    action="savePDF",
    value="./invoices/july_invoice.pdf"
)

printToPDF (Styling Method)

This action forces the browser to apply CSS @media print rules before rendering. Use this if the target website strips out ads and navigation bars specifically for printing.

BaseStep(
    id="print_clean",
    action="printToPDF",
    value="./invoices/july_invoice_clean.pdf"
)

---

📥 Intercepting Downloads

When you click an <a> tag or a button that forces the browser to download a file natively, traditional scrapers lose track of the binary blob. StepWright explicitly waits for the download event and reroutes the file to your disk.

eventBaseDownload

This requires two pieces: the selector to click, and the file destination.

BaseStep(
    id="download_report",
    action="eventBaseDownload",      # Tell the engine a download is expected
    object_type="id",
    object="export-csv-btn",         # The button that starts the download
    value="./downloads/export.csv"   # Where to save the intercepted payload
)

If you do not know the exact file name or extension beforehand (e.g., the server generates a random string), omit the explicit filename in the value param and just provide a directory path. Playwright will use the suggested filename from the server's Content-Disposition headers.

BaseStep(
    id="download_dynamic",
    action="eventBaseDownload",      
    object_type="class",
    object="dl-file",         
    value="./downloads/"             # Wait, intercept, and save with the server's name 
)