#!/bin/bash

ASPELL=aspell
HUNSPELL=hunspell

set -e

export LANG=C
export LC_ALL=C
export LC_CTYPE=C
export LC_COLLATE=C

prep() {
  echo prep 

  cat misc/{offensive.1,offensive.2,profane.1} | sort -u > nosug

  cp en.aff eng_affix.dat

  cat << EOF > eng.dat
name eng
charset iso8859-1
special ' -*-
affix eng
EOF
}

doit() {
  echo creating en_$1.dic

  cat en-common.wl en_$1-wo_accents-only.wl | sort -u > en_$1.0

  comm en_$1.0 nosug -12 > en_$1-nosug.1

  comm en_$1.0 nosug -23 > en_$1.1

  $ASPELL -l ./eng munch-list < en_$1-nosug.1 | grep -v '^\(XXX\|>>>\)' | ./add-no-suggest > en_$1.2

  $ASPELL -l ./eng munch-list < en_$1.1 | grep -v '^\(XXX\|>>>\)' >> en_$1.2

  cat en.dic.supp >> en_$1.2

  wc -l en_$1.2 | cut -d' ' -f1 > en_$1.dic
  cat en_$1.2 | sort >> en_$1.dic

  cp en.aff en_$1.aff
  
  cat README_en.txt.in ../Copyright > README_en_$1.txt
  echo >> README_en_$1.txt
  echo "Build Date: `date`" >> README_en_$1.txt

  zip -9 en_$1.zip README_en_$1.txt en_$1.dic en_$1.aff

  cp -p en_$1.zip hunspell/

  #echo check

  cat en_$1-nosug.1 en_$1.1 | sort -u > en_$1.tocheck

  #hunspell -l -d ./en_$1 < en_$1.dic.tocheck > misspelled
}


prep

doit US
doit CA

rm eng*.dat nosug en_US*.? en_CA*.?