#!/usr/bin/perl

#
# Perl script to download EDICT dictionaries for gjiten.
#
# It will fetch and unpack dicfiles then configure the gconf settings.
#

use strict;

my $TMPDIR = "/tmp/gjiten-dics";
my $DLDIR = "$TMPDIR/dl";

my $SYSTEMWIDE = ($< == 0) ? 1 : 0; # if we run as root, do a system wide install.

my $UTFDICDIR;
if ($SYSTEMWIDE == 1) {
    $UTFDICDIR = "/usr/share/gjiten/dics";
}
else {
    $UTFDICDIR = "$ENV{HOME}/gjiten";
    die "couldn't get home directory!\n" if "$ENV{HOME}" eq "";
}

if ($#ARGV != 0) {
    printhelp();
    exit 1;
}

my $BASICSETUP;

if (@ARGV[0] eq "-basic") {
    $BASICSETUP = 1;
}
elsif (@ARGV[0] eq "-all") {
    $BASICSETUP = 0;
}
else {
    printhelp();
    exit 1;
}

sub printhelp {

    print <<_HELP_

 This script will download and install all the dictionary files
 known to work with Gjiten.
 You can either install the basic dics (edict + kanjidic) or install
 them all (this requires ~60Mb of disk space!).

 If you run this script as a user, the dictionary files will be
 installed in \$HOME/gjiten.
 If you run it as root, they will be placed under /usr/share/gjiten/dics
 and they will be set up system-wide in gconf to make them automatically
 available to all users without any setup.

 Usage:
	\"$0 -all\"    install all dictionary files.
	\"$0 -basic\"  install basic dictionary files only.

_HELP_
;
}

# binaries
my $WGET = "/usr/bin/wget";
my $GUNZIP = "/bin/gunzip";
my $UNZIP = "/usr/bin/unzip";
my $ICONV = "/usr/bin/iconv";
my $GCONFTOOL = "/usr/bin/gconftool-2";

#my $DELETETMP = 0;  # clean up after finishing?

foreach my $bin ($WGET, $GUNZIP, $ICONV, $GCONFTOOL) {
    if (! -e $bin) {
	die "Couldn't find executable $bin!\n";
    }
}

my @dics = (
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/edict.gz",
		"package"     => "edict.gz",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "edict",
		"targetfile"  => "edict",
		"name"        => "English-main",
		"section"     => "basic",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/kanjidic.gz",
		"package"     => "kanjidic.gz",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "kanjidic",
		"targetfile"  => "kanjidic",
		"name"        => "KanjiDic",
		"section"     => "kanjidic",
	    },

# EXTRA
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/enamdict.gz",
		"package"     => "enamdict.gz",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "enamdict",
		"targetfile"  => "enamdict",
		"name"        => "Japanese Names",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/j_places.gz",
		"package"     => "j_places.gz",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "j_places",
		"targetfile"  => "j_places",
		"name"        => "Japanese Places",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/lifscdic.Z",
		"package"     => "lifscdic.Z",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "lifscdic",
		"targetfile"  => "lifescidic",
		"name"        => "Science",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/engscidic.gz",
		"package"     => "engscidic.gz",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "engscidic",
		"targetfile"  => "engscidic",
		"name"        => "Science2",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/compdic.gz",
		"package"     => "compdic.gz",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "compdic",
		"targetfile"  => "compdic",
		"name"        => "Computers",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/geodic.gz",
		"package"     => "geodic.gz",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "geodic",
		"targetfile"  => "geodic",
		"name"        => "Geology",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/compverb.gz",
		"package"     => "compverb.gz",
		"format"      => "gzip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "compverb",
		"targetfile"  => "compverb",
		"name"        => "Compound Verbs",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/riverwater.zip",
		"package"     => "riverwater.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "riverwater",
		"targetfile"  => "riverwater",
		"name"        => "RiverWater",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/4jword3_edict.zip",
		"package"     => "4jword3_edict.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "4jword3_edict.euc",
		"targetfile"  => "4jword3",
		"name"        => "4-kanji expressions",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/classical.zip",
		"package"     => "classical.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "classical",
		"targetfile"  => "classical",
		"name"        => "Classical Japanese",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/lingdic.zip",
		"package"     => "lingdic.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "lingdic",
		"targetfile"  => "lingdic",
		"name"        => "Linguistics",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/forsdic.zip",
		"package"     => "forsdic.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "forsdic_e",
		"targetfile"  => "forsdic",
		"name"        => "Forestry",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/aviation.zip",
		"package"     => "aviation.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "aviation",
		"targetfile"  => "aviation",
		"name"        => "Aviation",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/concrete.zip",
		"package"     => "concrete.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "concrete",
		"targetfile"  => "concrete",
		"name"        => "Concrete",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/findic.zip",
		"package"     => "findic.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "findic",
		"targetfile"  => "findic",
		"name"        => "Financial",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/mktdic.zip",
		"package"     => "mktdic.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "mktdic",
		"targetfile"  => "mktdic",
		"name"        => "Marketing",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/stardict.zip",
		"package"     => "stardict.zip",
		"format"      => "zip",
		"encoding"    => "EUC-JP",
		"sourcefile"  => "stardict",
		"targetfile"  => "stardict",
		"name"        => "Stars-Constellations",
		"section"     => "extra",
	    },
	    { 
		"url"         => "http://ftp.cc.monash.edu.au/pub/nihongo/buddhdic.zip",
		"package"     => "buddhdic.zip",
		"format"      => "zip",
		"encoding"    => "UTF-8",
		"sourcefile"  => "buddhdicu",
		"targetfile"  => "buddhdic",
		"name"        => "Buddhism",
		"section"     => "extra",
	    },
# GERMAN
	    { 
		"url"         => "http://bibiko.de/WdJTUTF.zip",
		"package"     => "WdJTUTF.zip",
		"format"      => "zip",
		"encoding"    => "UTF-8",
		"sourcefile"  => "WdJTUTF.txt",
		"targetfile"  => "wadoku",
		"name"        => "Japanese-German",
		"section"     => "extra",
	    },
# CHINESE
	    { 
		"url"         => "http://www.mandarintools.com/download/cedictu8.zip",
		"package"     => "cedictu8.zip",
		"format"      => "zip",
		"encoding"    => "UTF-8",
		"sourcefile"  => "cedict_ts.u8",
		"targetfile"  => "cedict",
		"name"        => "Chinese-English",
		"section"     => "extra",
	    },

	    );


print "Doing a user-only installation into $UTFDICDIR.\n" if ($SYSTEMWIDE == 0);
print "Doing a system-wide installation into $UTFDICDIR.\n" if ($SYSTEMWIDE == 1);


if (! -d $TMPDIR) {
    mkdir $TMPDIR or die "couldn't create $TMPDIR";
}
if (! -d $DLDIR) {
    mkdir $DLDIR or die "couldn't create $DLDIR";
}
if (! -d $UTFDICDIR) {
    mkdir $UTFDICDIR or die "couldn't create $UTFDICDIR";
}

my $OKDIC = 0;

foreach my $dic (@dics) {

    next if (($BASICSETUP == 1) && ($dic->{section} eq "extra"));

# DOWNLOAD    
#wget -c http://ftp.cc.monash.edu.au/pub/nihongo/edict.gz -O dl/edict.gz
    if (! -f "$DLDIR/$dic->{package}") {
	print "Downloading \"$dic->{name}\" from $dic->{url}\n";
	my @wget = ("$WGET -q -c $dic->{url} -O $DLDIR/$dic->{package}");
	if (system(@wget) != 0) {
	    $dic->{status} = "download failed";
	    next;
	}
    }
    $dic->{status} = "downloaded";


# UNPACK
#gunzip   <dl/edict.gz >tmp/edict
#unzip -o  dl/classical.zip -d tmp
    if (! -f "$DLDIR/$dic->{sourcefile}") {
	my @unpack;
	if ($dic->{format} eq "gzip") {
	    @unpack = ("$GUNZIP <$DLDIR/$dic->{package} >$DLDIR/$dic->{sourcefile}");
	    #print join(" ", @unpack) . "\n";
	    if (system(@unpack) != 0) {
		$dic->{status} = "unpack failed";
		next;
	    }
	}
	elsif ($dic->{format} eq "zip") {
	    @unpack = ("$UNZIP -o $DLDIR/$dic->{package} -d $DLDIR");
	    if (system(@unpack) != 0) {
		$dic->{status} = "unpack failed";
		next;
	    }
	}
	elsif ($dic->{format} ne "uncompressed") {
	    $dic->{status} = "unpack failed";
	    next;
	}
    }
    $dic->{status} = "unpacked";

# CONVERT
#iconv -c -f EUC-JP -t UTF-8 tmp/edict -o dics/edict

    if (! -f "$UTFDICDIR/$dic->{targetfile}") {
	my @iconv;
	if ($dic->{encoding} ne "UTF-8") {
	    #print "$ICONV -c -f $dic->{encoding} -t UTF-8 $DLDIR/$dic->{sourcefile} -o $DLDIR/$dic->{targetfile}.utf8\n";
	    @iconv = ("$ICONV -c -f $dic->{encoding} -t UTF-8 $DLDIR/$dic->{sourcefile} -o $DLDIR/$dic->{targetfile}.utf8");
	    if (system(@iconv) != 0) {
		$dic->{status} = "iconv failed";
		next;
	    }
	    else {
		$dic->{sourcefile} = "$dic->{targetfile}.utf8";
	    }
	}
    }
    $dic->{status} = "converted";
    

# MOVE

    if (! -f "$UTFDICDIR/$dic->{targetfile}") {
	if (system("/bin/mv $DLDIR/$dic->{sourcefile} $UTFDICDIR/$dic->{targetfile}") != 0) {
	    $dic->{status} = "move failed";
	    next;
	}
    }
    $dic->{status} = "OK";
    $OKDIC++;
}

die "All dicfiles failed. Please run \"rm -rf $TMPDIR\"\n" if ($OKDIC == 0);

print "Dicfiles successfully installed into $UTFDICDIR.\n";

my $gconfstrg = "[";
my $kanjidicgconf = "";

foreach my $dic (@dics) {
    next if (($BASICSETUP == 1) && ($dic->{section} eq "extra"));

    print "[$dic->{name}] $dic->{status}\n";
    if ($dic->{status} eq "OK") {
	if ($dic->{section} eq "kanjidic") {
	    $kanjidicgconf = "$UTFDICDIR/$dic->{targetfile}";
	}
	else {
	    $gconfstrg .= "$UTFDICDIR/$dic->{targetfile}\n$dic->{name},";
	}
    }
}
chop($gconfstrg);
    $gconfstrg .= "]";

if ($SYSTEMWIDE == 0) {
    if (system("$GCONFTOOL --type list --list-type=string --set /apps/gjiten/general/dictionary_list \"$gconfstrg\"") != 0) {
	print "Setting up dicfiles for gjiten failed.\n";
    }

    if ($kanjidicgconf ne "") {
#    print "$GCONFTOOL --type string --set /apps/gjiten/kanjidic/kanjidicfile \"$kanjidicgconf\"\n";
	if (system("$GCONFTOOL --type string --set /apps/gjiten/kanjidic/kanjidicfile \"$kanjidicgconf\"") != 0) {
	    print "Setting up kanjidic for gjiten failed.\n";
	}
    }
}
else {
    my $SCHEMAFILE = "$TMPDIR/gjiten-dicfiles.schema";
    open(SCHEMA, ">$SCHEMAFILE") || die "couldn't create $SCHEMAFILE\n";

    print SCHEMA <<__SCHEMA__
<gconfschemafile>
  <schemalist>

    <schema>
      <key>/schemas/apps/gjiten/general/dictionary_list</key>
      <applyto>/apps/gjiten/general/dictionary_list</applyto>
      <owner>gjiten</owner>
      <type>list</type>
      <list_type>string</list_type>
      <default>$gconfstrg</default>
      <locale name=\"C\">
         <short>List of dictionary files</short>
         <long>
          List of dictionary files. Format is: 
          [/path/to/dicfile1\ndictionary_name1,/path/to/dicfile2\ndictionary_name2]
          Don\'t put a space after the comma!
         </long>
      </locale>
    </schema>

    <schema>
      <key>/schemas/apps/gjiten/kanjidic/kanjidicfile</key>
      <applyto>/apps/gjiten/kanjidic/kanjidicfile</applyto>
      <owner>gjiten</owner>
      <type>string</type>
      <default>$kanjidicgconf</default>
      <locale name=\"C\">
       <short>Kanjidic dictionary file</short>
       <long>Kanjidic dictionary file</long>
      </locale>
    </schema>

  </schemalist>
</gconfschemafile>

__SCHEMA__
;
    close(SCHEMA);

    if (system("GCONF_CONFIG_SOURCE=xml::/etc/gconf/gconf.xml.defaults $GCONFTOOL --makefile-install-rule $SCHEMAFILE") != 0) {
	print "Gconf schema installation failed!\n";
    }
    else {
	system("/usr/bin/killall gconfd-2"); #restart gconfd
    }
}

print "\nPlease run \"rm -rf $TMPDIR\" if everything is set up correctly!\n";