#!/usr/bin/perl -w
#
# Kwiki to Foswiki text converter
#
# Copyright (c) 2009 Herbert Liechti, herbert.liechti@thinx.ch
#
# Acknowledgements:
#    Thanks to http://www.thinx.ch/ for giving me
#    permission to distribute this.
#
# Adapted from:
# Copyright (c) 2003 Fred Morris, m3047 at inwa d0t net
#    Thanks to http://www.graysoft.com/ for giving me
#    permission to distribute this.
#
#
# Dolphin to TWiki text converter
#
# Copyright (C) 2001 Peter Thoeny, peter@thoeny.com
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details, published at
# http://www.gnu.org/copyleft/gpl.html
#

# 
#
#
# !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! important !!!!!!!!!!!!!!!!!!!!!!
# steps before and after using this tool. This tool is quick and dirty hack
# for migration kwiki wikis to foswiki. It assumes, that the archive
# kwiki plugin is active and checks wether attachments exists for
# a topic
#
# Kwiki::VimMode plugin of kwiki (.vim) is also rewritten in this script
# you need the SyntaxHighlightingPlugin in foswiki
#
# STEPS:
#
# * of course you have a backup of both systems
# * Go to the kwiki repository and cd to the database directory
# * Delete unwanted wiki pages (i.e. all Kwiki* pages)
# * to a recode of all files from utf8 to the charset defined in the foswiki
#   unless you use utf8 in foswiki. Important!!
#   use the recode program
#      for i in * 
#      do
#         recode utf8..latin1 $i
#      done
# * Go to the plugin/archive directory of the kwiki
# * repeat the recode for all rcs files (,v)
# * open foswiki admin in your browser and add a new web 
# * place this script in the data directory of your foswiki installation
#   (debian /var/lib/foswiki/data), set the users variables below in this script to the correct values and start
#   the script
# * after migration unlock all rcs files (kwiki files are always locked) in the foswiki web directory
#   rcs files are named as topic.txt,v
#   do this as the user who owns the lock
#
#   ex. shellscript
#   for i in *.txt,v
#   do
#     rcs -u $i  # unlock stuff
#   done
# * be sure that the ownership ist correct for the migrated wiki. 
# * kwiki allows any word to be a wikiword. if you have kwiki pages
#   starting with lowercase or are not wikiwords you can use
#   the next tool wikify.pl which replace interactively the content
#   in your wiki with correct WikiWords
# * if you have attachments change configuration of your web so that attachments
#   are shown automatically

use lib ( '.' );
use lib ( '../lib' );
use Regexp::Common qw /URI/;
use File::Copy;

# -------------------------  User variables settings -----------------------------------------
$outDir = '/var/lib/foswiki/data/ContentX/'; # path of foswiki web
$inDir = '/var/www/cxWiki/database/';  # path of origin kwiki 
$arDir  = '/var/www/cxWiki/plugin/archive/';  # path of origin kwiki rcs files
$attachDir  = '/var/www/cxWiki/plugin/attachments/'; # path of origin kwiki attachments
$attachOutDir  = '/var/lib/foswiki/pub/ContentX';  # path of target attachment dir in foswiki
# -------------------------  End User variables settings -------------------------------------
# these variables will be set by TWiki.cfg:
$revCiDateCmd = "";
$defaultUserName = "";

# read the configuration part
do "TWiki.cfg";

# Not quite, appears some things have changed in Twiki.cfg since
# Thoeny wrote the original proggie. -- FWM

%settingsHash = @storeSettings;
$revCiDateCmd = $settingsHash{ciDateCmd};

# On with the show..


#if( @ARGV ) {
    #$inDir  = $ARGV[0] if $ARGV[0];
    #$outDir = $ARGV[1] if $ARGV[1];
    #$inDir  = "" unless -d $inDir;
    #$outDir = "" unless -d $outDir;
#
#}

&main();

sub main
{
    print "Kwiki to foswiki text converter. Copyright (C) 2009, Herbert Liechti\n";
    print "Kwiki to TWiki   text converter. Copyright (C) 2003, Fred Morris, m3047 at inwa d0t net\n";
    print "                               Copyright (C) 2001, Peter Thoeny, http://TWiki.org/\n";
    if( ! $inDir || ! $outDir ) {
        print "example% kwiki2twiki fromDir toDir\n";
        print "fromDir:   Directory containing Kwiki text files\n";
        print "toDir:     Directory where converted .txt and .txt,v files are stored\n";
        print "Attention: toDir is assumed to be empty. Existing files will be overwritten!\n";
        return;
    }
    if( $inDir eq $outDir ) {
        print "Error: toDir and fromDir must be different!\n";
        return;
    }

    opendir( DIR, "$inDir" ) or die "could not open $inDir";
    my @inList = grep /^\w/, readdir DIR;
    closedir DIR;
    my $text = "";
    my $time = 0;
    foreach( @inList ) {
    #foreach( qw/FranklinGothic PerlJournals YuiDnd HomePage/ ) {
        print "- $_: read";
        $text = readFile( "$inDir/$_" );
        print ", convert";
        ( $text, $time ) = convertText( $text, $_ );
        print ", copy rcs";
        copyrcsFile( $_ );
	
	$outf = "$outDir/$_" . '.txt';

        print ", save";
        saveFile( $outf, $text );
        print ", attachments";
        attachFiles( $_ );
        print "\n";
    }
}

# =========================
sub convertText
{
    my( $theText, $file ) = @_;
    my $utime = (stat("$inDir$file"))[9];
print ", utime $inDir$file $utime";
    my $revision = '1.1';
    if ( -f "$arDir/${file},v") {
       my $pid = open (PH, "rlog -r $arDir/${file},v |") or die $!;
       print ", rcs rlog PID $pid";
       for my $l (<PH>) {
          if ( $l =~ m/^head: (.*)$/ ) {
	      $revision = $1;
	      next;
	  }
       }
       close PH;
    }

    my $text = '%META:TOPICINFO{author="BaseUserMapping_333" '
	     . ' date="' . $utime . '" format="1.0" version="'. $revision . '"}%' . "\n";
    my $time = 0;
    my $verbatim = 0;
    my $verbatimblock = 0;
    my $vim = 0;
    $theText =~ s/\r//gos;
    foreach( split( /\n/, $theText ) ) {

        # in a verbatim block
    	if    ($verbatim) {

	    if ( m/^ /o ) {

	        $text .= "$_\n";
		next;
	    }
	    else {

	    	$text .= "</verbatim>\n";
		$verbatim = 0;
	    }
	}
        # in a verbatim block
    	if    ($verbatimblock) {

	    if ( m/^\.pre/ ) {
	    	$text .= "</verbatim>\n";
		$verbatimblock = 0;
                next;

	    }
	    else {
	        $text .= "$_\n";
		next;

	    }
	}
        # in a vim block
    	if    ($vim) {
            if ($vim == 1) {
               if ( m/filetype:\s+(.*)/ ) {
                   $text .= "%CODE{\"$1\"}%\n";
               } else {
                   $text .= "%CODE{\"...\"}%\n$_";
               }
               $vim++;
               next;
            }


	    if ( m/^\.vim\s*$/ ) {
	    	$text .= "%ENDCODE%\n";
		$vim = 0;
                next;

	    }
	    else {
	        $text .= "$_\n";
		next;

	    }
	}

	# start a verbatim block
	if    ( m/^\.pre/ ) {

	    $text .= "<verbatim>\n";
	    $verbatimblock = 1;

	    next;
	}
	# start a vim block
	if    ( m/^\.vim\s*$/ ) {

	    $vim = 1;

	    next;
	}
	# start a verbatim block
	if    ( m/^ /o ) {

	    $text .= "<verbatim>\n$_\n";
	    $verbatim = 1;

	    next;
	}

	# = Head 		--> ---+ Head
	if    ( m/^(=+)\s+(.*)/o ) {

	    my $indent = $1;
	    my $header = $2;

	    $indent =~ s/=/+/go;
	    $header =~ s/=+$//o;
	    $_ = "---$indent $header";
	}
	# [=code]		--> =code=
	s/\[=(.*?)]/=$1=/go;
	# [Wikiword]		--> [[Wikiwor]]
	s/\[(\w+)\]/##$1££/go;
        s/\[($RE{URI})\s+([^\]]*)\]/[[$1][$2]]/g;
        s/\[(.*)\s+($RE{URI})\]/[[$2][$1]]/g;

	# /*bolditalic*/ 	--> __bolditalic__
	s/(^|\s)\/\*(\S.*?\S)\*\/(\W|$)/$1__$2__$3/go;
	# /italic/ 		--> _italic_
	s/(^|\s)\/(\S.*?\S)\/(\W|$)/$1_$2_$3/go;
	# &			--> &amp;
	s/&/&amp;/go;
	# <			--> &lt;
	s/</&lt;/gi;
	# >			--> &gt;
	s/>/&gt;/gi;
	# !WikiWord		--> <nop>WikiWord
	s/!([A-Z]\S+)/<nop>$1/go;
	# * list		-->   * list
	if    ( m/^([*]+)\s*(.*)/go ) {

	    my $indent = $1;
	    my $item = $2;

	    $indent =~ s/./\t/go;

	    $_ = $indent . '* ' . $item;
	}
	# 0 list		-->   1 list
	if    ( m/^([0-9]+)\s*(.*)/go ) {

	    my $indent = $1;
	    my $item = $2;

	    $indent =~ s/./\t/go;

	    $_ = $indent . '1 ' . $item;
	}
	s/##/[[/go;
	s/££/]]/go;

	$text .= "$_\n";
    }

    $text =~ s/\n\n+/\n\n/gos;
    return ( $text, $time );
}

# =========================
sub readFile
{
    my( $theName ) = @_;
    my $data = "";
    undef $/; # set to read to EOF
    open( IN_FILE, "<$theName" ) || return "";
    $data = <IN_FILE>;
    $/ = "\n";
    close( IN_FILE );
    return $data;
}

# =========================
sub saveFile
{
    my( $theName, $theText ) = @_;
    open( FILE, ">$theName" ) or warn "Can't create file $theName";
    print FILE $theText;
    close( FILE);
}

# =========================

sub copyrcsFile {
   my $file = shift;
   if ( -f "$arDir/${file},v") {
      copy( "$arDir/${file},v", "$outDir/${file}.txt,v" ) or die $!;
   }
}
sub attachFiles {
   my $file = shift;
   if ( -d "$attachDir/${file}") {
      unless ( -d "$attachOutDir/$file" ) {
         mkdir "$attachOutDir/$file" or die $! ;
      }
      opendir( DIR, "$attachDir/${file}" ) or die "could not open $attachDir/$file";
      my @inList = grep /^\w/, readdir DIR;
      closedir DIR;
      for (@inList) {
         print "\n       attaching file $_";
         copy( "$attachDir/$file/$_", "$attachOutDir/${file}/$_" ) or die $!;
      }
   }
}