Extracting hyperlinks from PDFs can be challenging, as links are often embedded within the document structure in non-obvious ways. Today we’re announcing support for hyperlink extraction in our parsing API.
How It Works
When you parse a PDF with Datalab, our API now automatically extracts and preserves hyperlinks. The output includes:
- Link text: The visible clickable text
- Destination URL: Where the link points to
- Position: Bounding box coordinates for the link
Using the API
To extract hyperlinks, simply add the extract_links parameter to your API request:
import datalab
client = datalab.Client()
result = client.parse(
file="document.pdf",
extract_links=True
)
for link in result.links:
print(f"Text: {link.text}")
print(f"URL: {link.url}")
print(f"Page: {link.page}")Supported Link Types
Our hyperlink extraction supports:
- Web URLs: Standard http/https links
- Email links: mailto: links
- Internal links: Links to other pages within the document
- Anchors: Named destinations within the document
Try It Now
Hyperlink extraction is available in our playground and through the API. Sign up for free credits at datalab.to to get started.