Fix/frontend parallel load

[jun-create]

User description

Summary

Fixes #166

---

Changes Made

• Modified to do parallel loading of files from sequential loading of files on images_UI, tables_UI
• Fixed embed results on images_UI
• Fixed timeout on chat_UI
• Chat now works

---

Context / Rationale

Feature refining

---

Related Docs or References

---

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

Enhancement, Bug fix

---

Description

• Implement parallel loading for images and tables UI

• Fix embedding results display and timeout handling

• Refactor document components for better reusability

• Update chat prompts and add timeout configuration

---

Diagram Walkthrough

flowchart LR
  A["Sequential Loading"] --> B["Parallel Loading"]
  B --> C["Images UI"]
  B --> D["Tables UI"]
  E["Chat Timeout"] --> F["300s Timeout"]
  G["Embedding Display"] --> H["Small Table Format"]
  I["Document Components"] --> J["Reusable Components"]

Loading

File Walkthrough

	Relevant files
Enhancement	rag_config.py Update RAG validation examples and chunk formatting            chat_service/models/rag_config.py Updated validation examples with more generic document-focused queries Removed similarity score from chunk headers for cleaner output +8/-17    2_images_UI.py Implement parallel loading and enhance image extraction UI frontend/my_pages/2_images_UI.py Implemented parallel loading using asyncio.gather() for multiple documents Added reusable document components import and usage Enhanced error handling with proper status messages and timeout handling Fixed embedding display to show results in small table format Added refresh button and improved UI layout +179/-168 3_tables_UI.py Implement parallel loading for table extraction                    frontend/my_pages/3_tables_UI.py Refactored to use parallel loading with asyncio.gather() Integrated reusable document components Improved error handling and status reporting Added refresh button and cleaner UI layout +63/-148
Bug fix	5_chat_UI.py Add timeout and update chat prompts                                            frontend/my_pages/5_chat_UI.py Added 300-second timeout to HTTP client configuration Updated preset query button prompts to be more document-specific +4/-4      7_metadata_UI.py Improve metadata display error handling                                    frontend/my_pages/7_metadata_UI.py Enhanced error handling for metadata display Added proper fallback messages for failed processing or no metadata +7/-1

---

[jun-create]

/gemini review

[gemini-code-assist]

Code Review

This pull request introduces parallel loading for images and tables in the frontend, which is a great performance improvement. The code has been refactored to use asyncio.gather effectively. Additionally, several fixes are included, such as increasing the timeout for chat requests and improving error handling and status reporting across different UI pages. My review focuses on a few areas for improvement: correcting a blocking I/O call in an async function, fixing incorrect or missing type hints to improve code quality and maintainability, and cleaning up some duplicated code. Overall, these are solid changes that enhance the application's performance and robustness.

[codiumai-pr-agent-free]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

🎫 Ticket compliance analysis ✅ 166 - Fully compliant Compliant requirements: Chat page User input Single file rag Multiple file rag Assistant reply Pending check chat backend

⏱️ Estimated effort to review: 3 🔵🔵🔵⚪⚪

🧪 No relevant tests

🔒 No security concerns identified

⚡ Recommended focus areas for review Unused Variable The image_bytes variable is initialized but may not be assigned a value before being referenced in the conditional check and download button logic. image_bytes = None # Initialize to handle scope issues with col1: # Display actual image from URL if "detail" in image_data: st.error("An error occurred while loading the image.") else: # Fetch image with authenticated client with httpx.Client( cookies=st.session_state.httpx_cookies ) as client: img_response = client.get(image_url) Potential Error The display_tables function attempts to convert a DataFrame to CSV even when df might be None, which could cause runtime errors. if df is not None: csv_string = df.to_csv(index=False) st.code(csv_string, language="csv") Commented Code There's a commented line of code that appears to be a duplicate of the line that was replaced, which could cause confusion during maintenance. # chunk_header = f"--- Document Section {i + 1} (Relevance: {chunk.get('similarity_score', 0):.2f}) ---" chunk_header = f"--- Document Section {i + 1} (Relevance: {chunk.get('similarity_score', 0):.2f}) ---" chunk_header = f"--- Document Section {i + 1} ---"

[codiumai-pr-agent-free]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
General	Improve embedding process efficiency Refactor the display_all function to initiate and poll for each document's embedding status in a single concurrent task, improving UI responsiveness by showing results as they become available. frontend/my_pages/2_images_UI.py [176-186] async def display_all(expanders: list[DocumentExpander]): displays = [display_images(expander) for expander in expanders] await asyncio.gather(*displays, return_exceptions=True) with st.spinner("Embedding document for RAG..."): - # Post - starts embedding, response from embed_pdf is 202 - embed_start = [embed_pdf(expander.doc_id, "semantic") for expander in expanders] - await asyncio.gather(*embed_start, return_exceptions=True) + # Start embedding and poll for each document individually + embedding_tasks = [] + for expander in expanders: + async def embed_and_display(expander): + await embed_pdf(expander.doc_id, "semantic") + await display_embedding(expander) + embedding_tasks.append(embed_and_display(expander)) + + await asyncio.gather(*embedding_tasks, return_exceptions=True) - # Get - checks if embedding is done - embed_response = [display_embedding(expander) for expander in expanders] - await asyncio.gather(*embed_response, return_exceptions=True) - Apply / Chat Suggestion importance[1-10]: 7 __ Why: The suggestion correctly identifies an inefficient async pattern and proposes a better one that improves UI responsiveness by processing and displaying results per document, rather than in separate bulk stages.	Medium
Update

[NotYuSheng]

Approved! Will verify status before merging