[BUG] Composite multi-document cache key incorrectly stored as document_id in ChatMessage table

[vipul674]

Description of the Bug

In backend/app/routes/chat.py, when a user queries across multiple documents, the code creates a composite cache key:

cache_doc_key = "multi:" + ",".join(sorted(payload.document_ids))

This composite string (e.g., "multi:doc1,doc2,doc3") is then passed directly to _save_message() as the document_id parameter:

msg = _save_message(
    db=db,
    user_id=user_id,
    document_id=cache_doc_key,  # ← composite key, not a real document ID
    ...
)

This value is stored in the ChatMessage.document_id column, which is supposed to be a foreign key referencing the Document table. By storing a fabricated composite string:

• The foreign key contract is broken
• Queries filtering by document_id won't find these messages
• Any UI that looks up document title/metadata using document_id will fail
• Chat history for multi-document queries is effectively orphaned from document records

Steps to Reproduce

1. Upload at least two documents
2. Send a multi-document query (selecting both documents)
3. Check the chat_messages table in the database
4. Observe that document_id contains "multi:doc1,doc2" instead of a valid document UUID

Expected Behavior

Multi-document queries should store document_id as None (since there is no single document). The cache key should be stored separately, perhaps in a new field or as part of a cache subsystem, not in the document_id column.

Affected File

backend/app/routes/chat.py (lines ~259-260 and ~374)

Suggested Fix

Pass document_id=None for multi-document queries and maintain the cache key separately. Add a cache_key field to the ChatMessage model for this purpose, or use a separate cache data structure.

GSSoC '26

• Yes, I am participating in GirlScript Summer of Code and would like to fix this.

[github-actions]

Hi @vipul674! 👋

Thanks for opening this issue! We are thrilled to have you contribute to PDF-Assistant-RAG for GSSoC 2026.

📌 Before you start:

• Make sure you have read our Contributing Guidelines.
• Wait for a project admin to assign this issue to you before working on it.
• Remember to branch off dev and not main.

Happy Coding! 🚀

[vipul674]

@param20h I would like to work on this issue under GSSoC 2026. Please assign it to me.