#! /bin/sh

#================================================================
# estxdwhtml
# Strip a file of DocuWorks and extract its text as HTML.
#================================================================


# set variables
LANG=C ; export LANG
LC_ALL=C ; export LC_ALL
progname="estxdwhtml"
tmpfile="/tmp/$progname-$$.xdw"
dummyfile="/tmp/$progname-$$.dummy"
nulldev="/dev/null"
infile="$1"
outfile="$2"


# show help message
if [ "$1" = "--help" ]
then
  printf 'Strip a file of DocuWorks and extract its text as HTML.\n'
  printf '\n'
  printf 'Usage:\n'
  printf '  %s [infile] [outfile]\n' "$progname"
  printf '  estindex register -xsuf .xdw \\\n'
  printf '      application/vnd.fujixerox.docuworks %s casket\n' "$progname"
  printf '\n'
  exit 0
fi


# function to remove the temporary file
tmpclean(){
  rm -rf "$tmpfile"
}


# function to create the temporary file for input
output(){
  if [ -z "$outfile" ]
  then
    cat
  else
    cat >> "$outfile"
  fi
}


# set the exit trap
trap tmpclean 1 2 3 13 15


# check the input file existence
if [ -n "$infile" ] && [ ! -f "$infile" ]
then
  printf '%s: %s: no such file\n' "$progname" "$infile"
  exit 1
fi


# create the temporary file
if [ -z "$infile" ]
then
  cat > "$tmpfile"
  infile="$tmpfile"
fi


# output the result
xdw2text -p "$infile" "$dummyfile" 2> "$nulldev" |
iconv -f Shift_JIS -t UTF-8 -c |
awk '
BEGIN {
  printf("<html>\n")
  printf("<head>\n")
  printf("<meta http-equiv=\"Content-Type\" content=\"text/html; charset=UTF-8\">\n")
  printf("</head>\n")
  printf("<body>\n")
  printf("<p>")
}
{
  gsub(/&/, "\\&amp;", $0)
  gsub(/</, "\\&lt;", $0)
  gsub(/>/, "\\&gt;", $0)
  printf("%s", $0)
}
END {
  printf("</p>\n")
  printf("</body>\n")
  printf("</html>\n")
}
' |
output


# clean up the temporary directory
tmpclean


# exit normally
exit 0



# END OF FILE
