#!/usr/bin/env bash
# Run this after setting S2_API_KEY secret to confirm the worker + key work,
# then launch the full citation crawl.

set -e
WORKER="https://s2-proxy.researchstack.workers.dev"
SCRIPT_DIR="$(cd "$(dirname "$0")/../.." && pwd)/shim"

echo "=== health check ==="
curl -s "$WORKER/health" | python3 -m json.tool

echo ""
echo "=== test fetch (PageRank paper search) ==="
curl -sv "$WORKER/fetch?path=/graph/v1/paper/search&query=PageRank+Brin+Page+1998&fields=paperId,title,year&limit=2" 2>&1 | tail -20

echo ""
echo "=== launching citation crawl ==="
python3 "$SCRIPT_DIR/s2_citation_crawler.py" \
  --seed "PageRank Brin Page 1998 web search engine" \
  --workers "$WORKER" \
  --hops 2 \
  --max-refs 50 \
  --max-cites 30 \
  --max-papers 400 \
  --out /tmp/pagerank_citation_graph.json