fix: convert images off the event loop in LLMDocumentContentExtractor.run_async

[pcbeingused333]

Related Issues

• No issue; found while reading the component.

Proposed Changes:

LLMDocumentContentExtractor.run_async converts the documents to images with a direct call:

image_contents = self._document_to_image_content.run(documents=documents)["image_contents"]

DocumentToImageContent reads every file from disk, renders the requested page of each PDF and base64-encodes the result, and it implements no run_async. So the whole batch is converted on the event loop before the first LLM call is even scheduled, and anything else sharing that loop waits for it.

The module already imports _execute_component_async and uses it for the chat generator, which is exactly the helper for this: it prefers run_async when a component has one and otherwise runs run in a thread. This change routes the conversion through it. If DocumentToImageContent grows a run_async later, this call site picks it up for free.

run is untouched.

How did you test it?

• New unit test test_run_async_converts_images_off_the_event_loop: patches DocumentToImageContent.run to record the thread it executes on and asserts it is not the thread running the event loop. It fails on main, where the conversion runs on the main thread.
• hatch run test:unit — 6116 passed, 10 skipped.
• hatch run test:types and hatch run fmt clean.

Notes for the reviewer

I looked for the same pattern across the components with a run_async and this was the one remaining call that does blocking I/O; the other direct run calls in async methods are prompt builders, which are pure CPU and cheap.

I used an AI assistant while writing this change. I have reviewed it, reproduced the behaviour, and run the tests.

Checklist

• I have read the contributors guidelines and the code of conduct.
• I have added unit tests and updated the docstrings.
• I've used one of the conventional commit types for my PR title.
• I have documented my code.
• I have added a release note file.
• I have run pre-commit hooks and fixed any issue.

[vercel]

Someone is attempting to deploy a commit to the deepset Team on Vercel.

A member of the Team first needs to authorize it.

[CLAassistant]

All committers have signed the CLA.

[github-actions]

Hi @pcbeingused333, thanks for your interest in contributing to Haystack! 🙏

⚠️ You currently have 3 open pull requests in this repository (#12358, #12357 and this one). Our review capacity is limited, so please hold off opening more PRs until we've had a chance to review your first 2 open PRs. This helps us give each contribution the attention it deserves. Thank you!

This is an automated message to help us keep the review queue healthy.

[HaystackBot]

Hi @pcbeingused333, thanks a lot for your contribution! 🙏

We noticed that the Contributor License Agreement (CLA) check (license/cla) hasn't passed yet, so we've temporarily moved this PR to draft and paused the review assignment.

To get your PR reviewed, please sign the CLA via the link in the license/cla check below (or in the CLA bot comment). As soon as the check turns green, this PR will automatically be marked ready for review again and a reviewer will be re-assigned.

[pcbeingused333]

recheck

[pcbeingused333]

recheck

[HaystackBot]

Thanks for signing the CLA, @pcbeingused333! 🎉 This PR is now ready for review again and the reviewer has been re-assigned.