tests/test_fetch_citation_live.py

"""Isolated test — just fetch_citation(). Feeds it the exact claim dict
extract_claims() produced in test_extract_claims_live.py, so we're chaining
real output, not synthetic input.

Requires S2_API_KEY set in your environment.

Run from the veriscite/ root:
    python -m tests.test_fetch_citation_live
"""

import asyncio
from app.graph.nodes import fetch_citation

# Exact output from test_extract_claims_live.py — first claim only
SAMPLE_CLAIM = {
    "claim": "Scientific claims spread faster than they can be verified, and "
             "LLMs are increasingly used to check facts against primary literature",
    "citation_marker": "[1]",
    "reference_string": "Wadden, D., Lin, S., Lo, K., Wang, L. L., van Zuylen, M., "
                         "Cohan, A., & Hajishirzi, H. (2020). Fact or fiction: "
                         "Verifying scientific claims. In Proceedings of EMNLP 2020.",
    "resolved_paper_id": None,
    "evidence_text": None,
}


async def main():
    state = {
        "source_text": "",
        "claims": [SAMPLE_CLAIM],
        "audits": [],
        "current_index": 0,
        "report": None,
    }
    result = await fetch_citation(state)
    cc = result["claims"][0]

    print("UPDATED CLAIM CITATION:")
    print("resolved_paper_id:", cc["resolved_paper_id"])
    print("evidence_text (first 200 chars):", (cc["evidence_text"] or "")[:200])
    print()
    print("resolution succeeded:", cc["resolved_paper_id"] is not None)


if __name__ == "__main__":
    asyncio.run(main())