#!/usr/bin/perl
# pluckbook: generates a Plucker compatible pdb file given a html
# file. Handles compressed html files.
# Version 2.0
# Author: elf@ee.ryerson.ca, Jun 6, 2002
# $Id: pb,v 1.1 2002/09/21 15:32:34 elf Exp $

my $PLUCKDIR="$ENV{HOME}/palm/plucker";	# this is where plucked files go

#location of the plucker program
my $PLUCKER="/usr/local/plucker-1.1.12/bin/plucker-build ";

my $archtype=0;			# 0=uncompressed, 1=gzip, 2=zip
my $DOC_TITLE;

if ($#ARGV <2) {
  print("Usage: pluckbook htmlfile depth pdbfile\n\"htmlfile\" may be compressed (.gz, .Z, .zip).\n");
  exit(0);
}

#if the file is compressed, uncompress it before plucking
if ($ARGV[0] =~ /\.(gz|Z)/) {
  print("Uncompressing $ARGV[0] with gunzip...\n");
  $archtype=1;
  system("gunzip $ARGV[0]");
}elsif($ARGV[0] =~ /\.zip/){  
  print("Uncompressing $ARGV[0] with unzip...\n");
  $archtype=2;
  system("unzip $ARGV[0]");
}

$ARGV[0]=~s/\.(gz|Z|zip)//;		# strip .gz, .Z or .zip extension
#print("$ARGV[0]\n");

open(FP, "<$ARGV[0]") || die "Could not open $ARGV[0] for reading\n";

# find the <title> tag in the file and extract the title.  Or if it's
# a plain text file, look for "Title:"


while(<FP>){
  chop;

  if(/<(title|TITLE)>/){
    s{<(title|TITLE)>}{}g; s{</(title|TITLE)>}{}g; 
    $DOC_TITLE=$_; close FP; 
  }elsif(/[Tt]itle[ :]/){
    s{[Tt]itle[ :]}{}g;
    $DOC_TITLE=$_; close FP;
  }

}

# If title cannot be extracted, use the filename without the extension
if(!$DOC_TITLE){
  print "Could not get title. ";
  $DOC_TITLE=$ARGV[0];
  $DOC_TITLE=~s/\.html//g;
}

print "Plucking with document title: $DOC_TITLE\n";

system("$PLUCKER -q --stayonhost --pluckerdir=$PLUCKDIR --home-url=\"$ARGV[0]\" -f $ARGV[2] -N \"$DOC_TITLE\" --category Books --maxdepth=$ARGV[1]");

if($archtype==1){
  print "Re-compressing $ARGV[0]\n";
  system("gzip $ARGV[0]\n");
}elsif($archtype==2){
  print "Re-compressing $ARGV[0]\n";
  system("zip $ARGV[0]\n");
}