22 lines
774 B
Python
22 lines
774 B
Python
"""
|
|
pdf_parser.py — PDF-to-Question AI Parser
|
|
STATUS: DEFERRED — Phase 2 (AI Coach LLM integration)
|
|
|
|
This module will use PyMuPDF to extract text from uploaded PDFs and pass it
|
|
to an LLM to structure questions into the database JSON schema.
|
|
|
|
When ready to enable:
|
|
1. Add `PyMuPDF==1.23.26` and `openai==1.14.0` back to requirements.txt
|
|
2. Set LLM_BASE_URL, LLM_MODEL_NAME, OPENAI_API_KEY in backend/.env
|
|
3. Replace this stub with the full implementation.
|
|
"""
|
|
from typing import List, Dict, Any
|
|
|
|
|
|
async def parse_pdf_to_questions(pdf_bytes: bytes, exam_pattern: str) -> List[Dict[str, Any]]:
|
|
"""
|
|
Stub: returns empty list. Real implementation uses PyMuPDF + LLM JSON extraction.
|
|
Replace in Phase 2 once LLM infrastructure is provisioned.
|
|
"""
|
|
return []
|