fix: Resolve asyncio event loop error and improve translate UI performance

[NotYuSheng]

User description

Summary

Fixed asyncio event loop error that caused the second file to fail on the translate page, and improved performance by removing the PDF preview feature.

---

Changes Made

• Fixed asyncio event loop binding issue where httpx.AsyncClient was shared across multiple asyncio.Runner instances
• Each document now creates its own client within the runner's event loop context
• Removed PDF preview iframe to eliminate lag caused by base64 encoding and rendering
• Kept download button functionality with proper authentication

---

Context / Rationale

Bug Fix: When processing multiple files on the translate page, the second file would fail with the error: "Event object is bound to a different event loop". This occurred because a single httpx.AsyncClient created at module level was being used across multiple asyncio.Runner() instances, each with their own event loop.

Performance Improvement: The translate page was experiencing significant lag due to automatically fetching and base64-encoding PDFs for preview in iframes. By removing the preview feature and keeping only the download button, the page now loads much faster while maintaining full functionality.

---

Related Docs or References

• asyncio event loop documentation
• Streamlit performance best practices

---

General Checklist

• I have tested these changes locally
• I have updated relevant documentation or added comments where needed
• I have linked relevant issues and tagged reviewers
• I have followed coding conventions and naming standards

---

PR Type

Bug fix, Enhancement

---

Description

• Fix asyncio event loop error for multi-file processing

• Remove PDF preview feature to improve performance

• Create separate HTTP client per document processing

• Keep download functionality with proper authentication

---

Diagram Walkthrough

flowchart LR
  A["Multiple Documents"] --> B["Individual AsyncIO Runner"]
  B --> C["Dedicated HTTP Client"]
  C --> D["Translation Processing"]
  D --> E["Download Button Only"]
  F["PDF Preview Removed"] --> G["Improved Performance"]

Loading

File Walkthrough

	Relevant files
Bug fix	6_translate_UI.py Fix asyncio client and remove PDF preview                                frontend/my_pages/6_translate_UI.py Remove global httpx.AsyncClient and create per-document clients Add client parameter to async functions for proper event loop binding Remove PDF preview iframe and base64 encoding functionality Keep download button with server-side PDF fetching +14/-28

---

[gemini-code-assist]

Summary of Changes

Hello @NotYuSheng, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses a critical bug affecting multi-file processing on the translation page and delivers a notable performance boost to the user interface. The changes ensure robust handling of asynchronous operations and streamline the user experience by removing a resource-intensive feature that was causing slowdowns.

Highlights

• Bug Fix: Asyncio Event Loop Error: Resolved an issue where processing multiple files on the translate page would fail with an 'Event object is bound to a different event loop' error. This was fixed by ensuring each document's processing creates its own httpx.AsyncClient within the asyncio.Runner's event loop context, preventing shared client conflicts.
• Performance Improvement: Translate UI: Significantly improved the performance of the translate UI by removing the PDF preview feature. The previous implementation, which involved base64 encoding and rendering PDFs in iframes, caused substantial lag. The download functionality for translated PDFs remains intact.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[codiumai-pr-agent-free]

PR Compliance Guide 🔍

Below is a summary of compliance checks for this PR:

Security Compliance
🟢	No security concerns identified No security vulnerabilities detected by AI analysis. Human verification advised for critical code.
Ticket Compliance
⚪	🎫 No ticket provided Create ticket/issue
Codebase Duplication Compliance
⚪	Codebase context is not defined Follow the guide to enable codebase context checks.
Custom Compliance
⚪	No custom compliance provided Follow the guide to enable custom compliance check.
Update

Compliance status legend

🟢 - Fully Compliant
🟡 - Partial Compliant
🔴 - Not Compliant
⚪ - Requires Further Human Verification
🏷️ - Compliance label

[gemini-code-assist]

Code Review

This pull request effectively resolves the asyncio event loop error by instantiating a new httpx.AsyncClient for each document, which is a good fix. The removal of the PDF preview for performance is also a sensible improvement. My review includes a critical suggestion to prevent resource leaks by properly closing the httpx.AsyncClient and a minor improvement for user-facing error messages.

[codiumai-pr-agent-free]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
Possible issue	Fix resource leaks and race conditions To prevent resource leaks and race conditions, refactor the main loop to use a single asyncio.Runner and manage the httpx.AsyncClient lifecycle with an async with block inside a dedicated async function. frontend/my_pages/6_translate_UI.py [143-180] # Main content if "processed_data" in st.session_state and st.session_state.processed_data: response_lst = list(st.session_state.processed_data.items()) if not response_lst: st.info("No documents have been processed yet. Upload a PDF to get started!") else: + runner = asyncio.Runner() for doc_id, data in response_lst: with st.expander(f"📄 {data['uploaded_filename']}", expanded=True): - runner = asyncio.Runner() - client = httpx.AsyncClient(cookies=st.session_state.httpx_cookies, timeout=300.0) - translation_data = runner.run(get_translation_status(doc_id, client)) + + async def process_document(): + async with httpx.AsyncClient(cookies=st.session_state.httpx_cookies, timeout=300.0) as client: + translation_data = await get_translation_status(doc_id, client) + renderer_data = None + if translation_data.get("status") == "completed": + renderer_data = await get_renderer_status(doc_id, client) + return translation_data, renderer_data + + translation_data, renderer_data = runner.run(process_document()) status = translation_data.get("status") if status == "not_found": st.warning("⚠️ No translation available for this document") st.info("Translation was not enabled when this document was uploaded, or it's still processing.") elif status == "error": st.error(f"❌ Error: {translation_data.get('error', 'Unknown error')}") elif status == "processing": st.info("⏳ Translation is currently processing...") elif status == "completed": display_translation_info(translation_data) # Check and display renderer section - renderer_data = runner.run(get_renderer_status(doc_id, client)) - display_renderer_section(doc_id, data['uploaded_filename'], renderer_data, runner, client) + # Create a new client for synchronous calls inside display_renderer_section + with httpx.Client(cookies=st.session_state.httpx_cookies, timeout=300.0) as sync_client: + display_renderer_section(doc_id, data['uploaded_filename'], renderer_data, runner, sync_client) elif status == "failed": st.error("❌ Translation failed") st.info("Please try uploading the document again.") else: st.warning(f"Unknown status: {status}") [To ensure code accuracy, apply this suggestion manually] Suggestion importance[1-10]: 9 __ Why: The suggestion correctly identifies a critical resource leak and unsafe asyncio usage where an httpx.AsyncClient is not closed and is used across multiple event loops, which could lead to instability.	High
General	Use synchronous client for synchronous operations Refactor display_rendered_pdf to be fully synchronous by using a synchronous httpx.Client instead of an httpx.AsyncClient with asyncio.Runner, simplifying the function and improving efficiency. frontend/my_pages/6_translate_UI.py [82-102] -def display_rendered_pdf(doc_id: str, filename: str, runner: asyncio.Runner, client: httpx.AsyncClient): +def display_rendered_pdf(doc_id: str, filename: str, client: httpx.Client): """Display download button for rendered PDF.""" # Server-side URL for fetching PDF (internal service communication) server_pdf_url = f"{PDF_PROCESSOR_URL}/renderer/{doc_id}/rendered.pdf" try: - pdf_response = runner.run(client.get(server_pdf_url)) - if pdf_response.status_code == 200: - st.download_button( - label="📥 Download Translated PDF", - data=pdf_response.content, - file_name=f"{filename.rsplit('.', 1)[0]}_translated.pdf", - mime="application/pdf", - key=f"download_{doc_id}", - use_container_width=True - ) - else: - st.error(f"Failed to load PDF (HTTP {pdf_response.status_code})") + pdf_response = client.get(server_pdf_url) + pdf_response.raise_for_status() # Raise an exception for bad status codes + + st.download_button( + label="📥 Download Translated PDF", + data=pdf_response.content, + file_name=f"{filename.rsplit('.', 1)[0]}_translated.pdf", + mime="application/pdf", + key=f"download_{doc_id}", + use_container_width=True + ) + except httpx.HTTPStatusError as e: + st.error(f"Failed to load PDF (HTTP {e.response.status_code})") + logger.error(f"Error fetching rendered PDF: {e}") except Exception as e: logger.error(f"Error fetching rendered PDF: {e}") st.error(f"Error loading PDF: {e}") Apply / Chat Suggestion importance[1-10]: 7 __ Why: The suggestion correctly points out that using runner.run for a single async call is inefficient and proposes a cleaner, fully synchronous implementation which improves code clarity and maintainability.	Medium
Update