#!/bin/sh
# This script downloads the "Spamwords" and "Not Evil" pages from the OHRRPGCE wiki
#
# Note that it depends on pcregrep, which is  not usually installed by default
# on a typical GNU/Linux box.

SPAMWORDSFILE="/var/www/wiki/blacklist.spamwords.txt"
SPAMWORDSURL="http://gilgamesh.hamsterrepublic.com/wiki/ohrrpgce/index.php/Spamwords"
SPAMWORDSSANE="this is a list of spammy keywords"

NOTEVILFILE="/var/www/wiki/not.evil.txt"
NOTEVILURL="http://gilgamesh.hamsterrepublic.com/wiki/ohrrpgce/index.php/Not_evil"
NOTEVILSANE="This is a list of wiki contributors"

wget -O - --quiet ${SPAMWORDSURL} \
 | pcregrep -M "(?s)<pre>.*</pre>" \
 | pcregrep -v "</?pre>" \
 > "${SPAMWORDSFILE}".new

SANITY=`grep "${SPAMWORDSSANE}" "${SPAMWORDSFILE}".new`

if [ "$SANITY" ] ; then
  mv "${SPAMWORDSFILE}".new "${SPAMWORDSFILE}"
else
  echo "ERROR: spamwords sanity check failed!"
  cat "${SPAMWORDSFILE}".new
  rm "${SPAMWORDSFILE}".new
fi

wget -O - --quiet ${NOTEVILURL} \
 | pcregrep -M "(?s)<pre>.*</pre>" \
 | pcregrep -v "</?pre>" \
 > "${NOTEVILFILE}".new

SANITY=`grep "${NOTEVILSANE}" "${NOTEVILFILE}".new`

if [ "$SANITY" ] ; then
  mv "${NOTEVILFILE}".new "${NOTEVILFILE}"
else
  echo "ERROR: not evil sanity check failed!"
  cat "${NOTEVILFILE}".new
  rm "${NOTEVILFILE}".new
fi