#!/usr/bin/env bash
set -u
mkdir -p /tmp/spectre_extra_corpus
cd /tmp/spectre_extra_corpus

declare -a URLS=(
  "https://arxiv.org/pdf/2409.18839"
  "https://arxiv.org/pdf/2310.06825"
  "https://arxiv.org/pdf/2103.14030"
  "https://arxiv.org/pdf/2305.10601"
  "https://arxiv.org/pdf/2106.09685"
  "https://www.irs.gov/pub/irs-pdf/p17.pdf"
  "https://www.irs.gov/pub/irs-pdf/p334.pdf"
  "https://www.irs.gov/pub/irs-pdf/p15.pdf"
  "https://www.irs.gov/pub/irs-pdf/p525.pdf"
  "https://www.irs.gov/pub/irs-pdf/p463.pdf"
  "https://www.irs.gov/pub/irs-pdf/p550.pdf"
  "https://www.irs.gov/pub/irs-pdf/p590a.pdf"
)

for url in "${URLS[@]}"; do
  fname=$(basename "$url")
  case "$fname" in
    *.pdf) ;;
    *) fname="${fname}.pdf" ;;
  esac
  if [ ! -f "$fname" ]; then
    if curl -sSL --max-time 90 -o "$fname" "$url"; then
      sz=$(stat -c%s "$fname" 2>/dev/null || echo 0)
      echo "got $fname ($sz bytes)"
    else
      echo "FAIL $fname"
      rm -f "$fname"
    fi
  else
    sz=$(stat -c%s "$fname" 2>/dev/null || echo 0)
    echo "have $fname ($sz bytes)"
  fi
done
