Parse a open.kattis.com problem listing page to fetch the individual problems for offline use.

parse-kattis-problem-page.sh

#!/bin/sh

# Expects a problems.txt to exist that is the source code of a problems page
if [ ! -f problems.txt ]; then
  echo "You should get the source of a page with a list of problems. Save it in problems.txt."
  exit 1
fi

problems=$(sed '/problems\//!d' problems.txt | sed '/\/submit/d' | sed '/\/statistics/d' | sed 's/^.*problems\/\(.*\)".*$/\1/')

mkdir -p problems

for problem in $problems; do
  if [ ! -f "problems/$problem.html" ]; then
    wget -nv -O "problems/$problem.html" "https://open.kattis.com/problems/$problem"
    sleep 1
  else
    echo "Skipping $problem. We already have that one."
  fi
done