#!/usr/bin/env ruby
# = sisu - SiSU information Structuring Universe
#
# Copyright (c) Ralph Amissah 1997,2004
#
# Ralph Amissah mailto:ralph@amissah.com
#
# * Name: SiSU information Structuring Universe
# * Author: Ralph@Amissah.com
# * Description: document conversion tool, to sisu from other formats
# *  License: GPL 3 or later
# * Notes: word conversion uses wvWare and wvSiSU.xml (a modified/stripped wvHtml.xml)
#   * http://wvware.sourceforge.net/
#   * http://sourceforge.net/projects/wvware
# * <url:sisu.lnk>|sisu.lnk|@|^|
# * <url:sisu>
module CONVERT
  class MyOutput
    def initialize(data, filename, instruct)
      @data=data.compact
      @filename=filename
      @instruct=instruct
    end
    def headerBasic
      <<WOK
% SiSU 2.0

@title:
 :subtitle:

@creator:
 :author:

@classify:
 :topic_register:

@date:
 :published:

@rights:
 :copyright:
 :license:

WOK
    end
    def headerDefault
      <<WOK
% SiSU 2.0

@title:
 :subtitle:

@creator:
 :author:

@classify:
 :topic_register:

@date:
 :published:

@rights:
 :copyright:
 :license:

WOK
    end
    def hardOutput
      pre = Array.new
      case @instruct
      when /default/
        pre << headerDefault
      else
        pre << headerBasic
      end
      @filename_wv=File.new(%{,,#{@filename}.sst},'w+')
      @filename_wv << pre
      @data.each do |x|
        y = x.split("\n")
        y.each do |z| # cleaner output this way
          z.strip!
          @filename_wv.puts "#{z}\n\n" unless z =~/^$/
        end
      end
    end
  end
  class WareWord97
    def initialize(data, filename, instruct)
      @data=data
      @filename=filename
      @instruct=instruct
    end
    def songsheet
      data=@data
      print "Convert to SiSU file from Word97 << gvim ,,#{@filename}.sst >\n" #: <<#{@@html_title}>>
      data=WareWord97.new(data.collect,@filename,@instruct).strip
      data=WareWord97.new(data.collect,@filename,@instruct).strip
      data=WareWord97.new(data.collect,@filename,@instruct).markup_rules
      data=MyOutput.new(data.collect,@filename,@instruct).hardOutput
    end
    def strip
      data=@data
      tuned_file=Array.new
      endnote_no=1
      data.each do |para|
        para.strip!
        para.gsub!(/<u>\s*<\/u>/,'')
        para.gsub!(/<\/u>\s*<u>/,'')
        para.gsub!(/<b>\s*<\/b>/,'')
        para.gsub!(/<\/b>\s*<b>/,'')
        para.gsub!(/<i>\s*<\/i>/,'')
        para.gsub!(/<\/i>\s*<i>/,'')
        tuned_file << para unless para == nil
      end
      tuned_file
    end
    def markup_rules
      data=@data
      tuned_file=Array.new
      endnote_no=1
      data.each do |para|
      para.strip!
        para.gsub!(/\s+/,' ')
        para.gsub!(/^<b>(Chapter|Article)(.+?)<\/b>/i,'1~ \1 \2') #watch case insensitivity
        para.gsub!(/^<b>(Part|Section|Book)(.+?)<\/b>/i,':C~ \1 \2') #watch case insensitivity
        para.gsub!(/^<b>(\d+\.\d+\.\d+\.?)(.+?)<\/b>/i,'3~ \1 \2') #numeric, decide what to do, can be different
        para.gsub!(/^<b>(\d+\.\d+\.?)(.+?)<\/b>/i,'2~ \1 \2') #numeric, decide what to do, can be different
        para.gsub!(/^<b>(\d.+?)<\/b>/i,'4~ \1 \2') #numeric, decide what to do, can be different
        #para.gsub!(/^<b>([\d.]+?)<\/b>/i,'4~ \1 \2') #numeric, decide what to do, can be different
        para.gsub!(/<u>(.+?)<\/u>/,'_{\1}_')
        para.gsub!(/<b>(.+?)<\/b>/,'!{\1}!')
        para.gsub!(/<i>(.+?)<\/i>/,'/{\1}/')
        tuned_file << para unless para == nil
      end
      tuned_file
    end
  end
  class Html
    def initialize(data, filename, instruct)
      @data=data
      @filename=filename
      @instruct=instruct
    end
    def songsheet
      data=@data
      print "Convert to SiSU file from #{@filename}.html << gvim ,,#{@filename}.sst >\n" #: <<#{@@html_title}>>
      #data=Html.new(data.collect, @filename, @instruct).space_paragraphs
      #data=Html.new(data.split(''), @filename, @instruct).space_paragraphs
      data=Html.new(data.join.split(/\n\n+/), @filename, @instruct).space_paragraphs
      #data=Html.new(data.split("\n"), @filename, @instruct).space_paragraphs
      #data=Html.new(data.collect.join.split("\n"), @filename, @instruct).space_paragraphs
      data=Html.new(data.collect, @filename, @instruct).multiline
      data=Html.new(data.collect.join.split("\n\n"), @filename, @instruct).markup_rules
      data=MyOutput.new(data.collect, @filename, @instruct).hardOutput
    end
    def space_paragraphs
      #data=@data.join.split(/\n/)
      data=@data
      #p data.length
      tuned_file=Array.new
      data.each do |para|
      para.strip!
        para.gsub!(/\r/,'')
        #para.gsub!(/\n/, ' ') #PROBLEM, serious time issues on a few files also for \n (or multiline matches which is less surprising), edit out if necessary
        para.gsub!(/<\/?p>/i,'zZz')
        para.gsub!(/<\/?\s*p(?:\s+ALIGN=.+?)?>/i,'zZz') #all manner of <p> para.gsub!(/<\/?p>/i, "\n\n")
        para.gsub!(/<p\s+(class|align).+?>/i,'zZz') #
        para.gsub!(/<\/p>/i,'zZz') # repeat actually
        para.gsub!(/<(?:dir|tr|br)>/i,'zZz') #
        #para.gsub!(/<(?:\/\s*)?(?:dir|tr|br)>/i, "zZz") #
        para.gsub!(/(<\/center>)/i,'\1zZz')
        para.gsub!(/(<\/h[1-6]>)/i,'\1zZz')
        para.gsub!(/ \s+/i,' ')
        para.gsub!(/(?:\s*zZz\s*)+/i,'zZz') #
        tuned_file << para unless para == nil
      end
      tuned_file
    end
    def blockquotes(sub='') # SERIOUS PROBLEM INTRODUCED, some blockquotes go missing !, quite unacceptable, debug, for now not used
      res=Array.new
      sub.each do |x|
        if x=~/(<\/blockquote>)/i
          m = $1
          res << x[/(.+?)#{m}/mi,1].gsub!(/zZz/,'zZz_1 ') if x =~/.+?#{m}/mi
          res << x[/#{m}(.+)/mi,1]
        else
          res << x #[/(.+)/mi,1]
        end
      end
      res.join
    end
    def multiline
      data=@data
      tuned_file=Array.new
      data.each do |para|
        para.gsub!(/\n/,' ')
        para.gsub!(/ \s+/mi,' ')
       #ALL HERE could be very time EXPENSIVE but tamed? compromise ... /mi
        para.gsub!(/<([biu]|h[1-6])>(?:zZz)?([^<]+)?zZz(.+?)<\/\1>/i,'zZz<\1>\2 \3</\1>')
        para.gsub!(/<([biu]|h[1-6])>(?:<center>|zZz)+(.+?)(?:<\/center>)?zZz(.+?)?<\/\1>/i,'zZz<\1>\2 \3</\1>')
        #para.gsub!(/<([biu]|h[1-6])>(?:<center>|zZz)+(.+?)<\/center>zZz(.+?)?<\/\1>/i,'zZz<\1>\2 \3</\1>')
        para.gsub!(/<([biu]|h[1-6])>(?:<center>|zZz)+(.+?)<\/\1>/i,'zZz<\1>\2</\1>')
        para.gsub!(/<(h[1-6])>(.+?)(?:<center>|zZz)+<\/\1>/i,'zZz<\1>\2</\1>zZz') #does catch some h1, h2 etc, too expensive to have biu
        #para.gsub!(/<([biu]|h[1-6])>(.+?)(?:<center>|zZz)+<\/\1>/i,'zZz<\1>\2 \3</\1>') #may go too far? useful for h1 h2 etc, remove biu?
        #para.gsub!(/<([biu]|h[1-6])>([^<]+)?zZz(.+?)<\/\1>/i,'zZz<\1>\2 \3</\1>')
        #para.gsub!(/<([biu]|h[1-6])>([^<]+)?zZz(.+?)<\/\1>/i,'zZz<\1>\2 \3</\1>')
       ### SERIOUS PROBLEM INTRODUCED
       # sub = para.split(/<blockquote>/i)
       # para = blockquotes(sub) if sub.length > 0 #check was on >1 could have serious repercussions 2004w29
        para.gsub!(/zZz(\s*zZz)*/,"\n\n")
        tuned_file << para << "\n\n" unless para == nil
      end
      tuned_file
    end
    def markup_rules
      @@flag_blockquote=false
      data=@data
      tuned_file=Array.new
      data.each do |para|
        if para=~/<a href="(http:\/\/.+?)">/i
          #p para.grep(/<a href="(http:\/\/.+?)">/i)
          #m=$1
          #para.gsub!(/(?:&lt;\s*)?<a href="#{m}">#{m}<\/a>(?:\s*&gt;)?\.?/i, "#{m}")
          para.gsub!(/(?:&lt;\s*)?<a href="(http:\/\/.+?)">http:\/\/.+?<\/a>(?:\s*&gt;)?\.?/i,'\1') #risk that url & url are not to match
          #para.gsub!(/(?:&lt;\s*)?<a href="(\w+\.html)">(http:\/\/.+?\/\1)<\/a>(?:\s*&gt;)?\.?/i, "\\2") #does not match
        end
        if para=~/<BLOCKQUOTE>/i
          @@flag_blockquote=true
        end
        if @@flag_blockquote
          para.gsub!(/^/,'_1 ') unless para.empty? or para =~/^\s*<\/?blockquote?>\s*$/i
        end
        if para=~/<\/BLOCKQUOTE>/i
          @@flag_blockquote=false
        end
        para.gsub!(/<\/?blockquote?>/i,'')
        ### clean
        para.gsub!(/^\s+/i,'')
        para.gsub!(/<([bui]|em|su[pb])>\s*<\/\1>/i,'')
        para.gsub!(/<\/?center>/i,'')
        para.gsub!(/\s*<\/dir>/i,'')
        para.gsub!(/<hr>/i,'')
        para.gsub!(/\s*<a href=".+?\.html#(?:[a-z_]+)?(?:[a-z0-9_-]|\*)+">\[(\*+)\]<\/a>/i,'^{[\1]}^ ') #other endnote marker
        para.gsub!(/<a href=".+?\.html#(?:[a-z_$]+)?[0-9_-]+"(?:\s+name=".+?")?>\[[a-z]?\d+\](?:<\/[bi]>)?<\/a>/i,'~^ ') #endnote marker
        para.gsub!(/<a name=".+?"\s+href=".+?\.html#(?:[a-z_$]+)?[0-9_-]+"?>\[[a-z]?\d+\](?:<\/[bi]>)?<\/a>/i,'~^ ') #endnote marker
        para.gsub!(/<a name="(?:[a-z$]+)?[0-9_-]+">\s*(<\/a>)?\s*\d+\.?\s*(<\/a>)?\s*/i,'^~ ') #endnote
        #para.gsub!(/<h([1-6])>\s*(.+?)\s*<\/h\1>\s*/i,'\1~ \2') #
        para.gsub!(/<h([1-6])(?: align=.+?)?>\s*(.+?)\s*<\/h\1>\s*/i,'\1~ \2') #
        para.gsub!(/^<b>(Chapter|Article)(.+?)<\/b>/i,'4~ \1 \2') #watch case insensitivity
        para.gsub!(/^<b>(Part|Section|Book)(.+?)<\/b>/i,'3~ \1 \2') #watch case insensitivity
        para.gsub!(/^<b>(\d+\.\d+\.\d+\.?)(.+?)<\/b>/i,'6~ \1 \2') #numeric, decide what to do, can be different
        para.gsub!(/^<b>(\d+\.\d+\.?)(.+?)<\/b>/i,'5~ \1 \2') #numeric, decide what to do, can be different
        para.gsub!(/^<b>(\d+\.?)(.+?)<\/b>/i,'4~ \1 \2') #numeric, decide what to do, can be different
        #<a name="ii"></a><B>
        para.gsub!(/^(<a name=".+?">)(?:<small>)?<(?:b|strong)>\s*(.+?)\s*<\/(?:b|strong)>/i,'5~ \2 \1') #watch
        para.gsub!(/^(<(a name|A NAME)=".+?">)(\s*|<\/[aA]>)?([A-Z][A-Z])+/,'5~ \2 \1') #watch
        para.gsub!(/^(\s+|<p>)?(<a name=".+?">)(\s*|<\/a>)?<b>/i,'5~ \2 \1') #watch
        para.gsub!(/<h([1-6])>\s*(.+?)\s*<\/h\1>\s*/i,'\1~ \2') #
        para.gsub!(/^<b>\s*(.+?)<\/b>\s*(<\/i>\s*)?$/i,'4~ \1\2') # wish it all were less messy
        para.gsub!(/^<i>\s*([^"(].+?)<\/i>\s*(<\/b>\s*)?$/i,'5~ \1\2') # wish it all were less messy
        para.gsub!(/<\/?[biu]>/i,'') if para =~/[1-6]\{/
        para.gsub!(/<u>\s*(.+?)\s*<\/u>/i,'_{\1}_')
        para.gsub!(/<(b|strong)>\s*(.+?)\s*<\/\1>/i,'*{\2}*')
        para.gsub!(/<(i|em)>\s*(.+?)\s*<\/\1>/i,'/{\2}/')
        para.gsub!(/<sup>\s*(.+?)\s*<\/sup>/i,'^{\1}^')
        para.gsub!(/(([\/\*!_])\{.+?\}\2)\s\s+/i,'\1 ')
        para.gsub!(/(([\/\*!_])\{.+?\}\2)\s+([.,;?\)])\s+/i,'\1\3 ')
        para.gsub!(/(([\/\*!_])\{.+?\}\2)(["'])\s+/i,'\1\3 ')
        para.gsub!(/(([\/\*!_])\{.+?\}\2)\s*([a-z0-9])/i,'\1 \3')
        para.gsub!(/(([\/\*_])\{.+?\}\2)\s*([a-z0-9])/i,'\1 \3')
        para.gsub!(/([a-z0-9])(([\/\*_])\{.+?\}\3)/i,' \1 \2') #eg this/{problem}/
        para.gsub!(/([\/\*_])\{([,.;; ]+)\}\1/i,'\2') #eg /{,}/ or *{ }* etc.
        para.gsub!(/ \s+/i,' ')
        #para.gsub!(/\/\{\*\{/i, '*{/{')
        #para.gsub!(/\}\*\}\//i, '}/}*')
        para.gsub!(/&quot;/i,'"')
        para.gsub!(/&amp;/i,'and')
        para.gsub!(/<!doctype html public .+/i,'')
        para.gsub!(/<\/?(?:html|head|body|font|small)>/i,'')
        para.gsub!(/<\/(?:title)>/i,'')
        para.gsub!(/<title>/i,'#{~title? ')
        para.gsub!(/<blockquote>(.+?)<\/blockquote>/mi,"\n\n_1 \\1\n\n")
        para.gsub!(/<div align=.+?>|<\/div>|<font size=.+?>|<\/a><\/em><\/strong>/i,'')
        para.gsub!(/~^\s+\.\s*/i,'.~^ ') #check vim equiv # %s/\~e\s\+\.\s*/.\~e /c
        para.gsub!(/\s+~^\s+/i,'~^ ')
        para.gsub!(/ \s+/i,' ')
        para.gsub!(/\s+$/i,'')
        para.gsub!(/^(?:<\/[bi]>)+$/i,'')
        para.gsub!(/^(?:(?:<i>)+<b>|(?:<b>)+<i>)\s*([^"(].+?)/i,'5~ \1\2') # wish it all were less messy
        para.gsub!(/^(?:<\/?(?:[ib]|em)>\s*)+$/i,'') # cleaning up left over <i> etc.
        para.gsub!(/<(?:i|em)>\s*(.+)/i,'/{\1}/') # using up left over <i>
        para.gsub!(/<b>\s*(.+)/i,'*{\1}*') # using up left over <b>
        para.gsub!(/<dd>([\d.]+)/i,'5~ \1')
        para.gsub!(/<dd>(?:&nbsp;)+([\d.]+)/i,'6~ \1')
        para.gsub!(/<dd>(\([a-z]\))/i,'7~ \1')
        para.gsub!(/^([1-9]~)( .+?)<a name="(\S+?)">(.+?)(<\/a>)/i,'\1\3\2\4')
        para.gsub!(/^([1-9]~)( .+?)<a name="(\S+?)">/i,'\1\3\2')
        para.gsub!(/http\/\/(\S+)/i,'http:\/\/\1')
        para.gsub!(/\s*<a href="\S+?">(http:\/\/\S+?)<\/a>\s*/i,' \1 ')
        para.gsub!(/([a-zA-Z.,!?;:])([*\/_-]\{)/,'\1 \2')
        para.gsub!(/^\s*(&nbsp;){10,12}/i,'_2 ')
        para.gsub!(/^\s*(&nbsp;){4,5}/i,'_1 ')
        para.gsub!(/&#9;/,' ') #check
        ## glyphs & tildes
        para.gsub!(/&iexcl;/,  '¡') #'Inverted exclamation
        para.gsub!(/&cent;/,   '¢') #'Cent sign               ¢
        para.gsub!(/&pound;/,  '£') #'Pound sign              £
        para.gsub!(/&curren;/, '¤') #'General currency sign
        para.gsub!(/&yen;/,    '¥') #'Yen sign                ¥
        para.gsub!(/&brvbar;/, '¦') #'Broken vertical bar
        para.gsub!(/&sect;/,   '§') #'Section sign            §
        para.gsub!(/&uml;/,    '¨') #'Umlaut
        para.gsub!(/&copy;/,   '©') #'Copyright               ©
        para.gsub!(/&ordf;/,   'ª') #'Feminine ordinal        ª
        para.gsub!(/&laquo;/,  '«') #'Left angle quote        «
        para.gsub!(/&not;/,    '¬') #'Not sign
        para.gsub!(/&shy;/,    '­') #'Soft hyphen
        para.gsub!(/&reg;/,    '®') #'Registered trademark    ®
        para.gsub!(/&macr;/,   '¯') #'Macron accent
        para.gsub!(/&deg;/,    '°') #'Degree sign             °
        para.gsub!(/&plusmin;/,'±') #'Plus or minus           ±
        para.gsub!(/&sup2;/,   '²') #'Superscript 2           ²
        para.gsub!(/&sup3;/,   '³') #'Superscript 3           ³
        para.gsub!(/&acute;/,  '') #'Acute accent
        para.gsub!(/&micro;/,  'µ') #'Micro sign (Greek mu)   µ
        para.gsub!(/&para;/,   '¶') #'Paragraph sign          ¶
        para.gsub!(/&middot;/, '·') #'Middle dot
        para.gsub!(/&cedil;/,  '¸') #'Cedilla
        para.gsub!(/&sup1;/,   '¹') #'Superscript 1           ¹
        para.gsub!(/&ordm;/,   'º') #'Masculine ordinal       º
        para.gsub!(/&raquo;/,  '»') #'Right angle quote
        para.gsub!(/&frac14;/, '¼') #'Fraction one quarter    ¼
        para.gsub!(/&frac12;/, '½') #'Fraction on half        ½
        para.gsub!(/&frac34;/, '¾') #'Fraction three quarters ¾
        para.gsub!(/&iquest;/, '¿') #'Inverted question mark  ¿
        para.gsub!(/&Agrave;/, 'À') #'Capital A, grave accent À
        para.gsub!(/&Aacute;/, 'Á') #'Capital A, acute accent Á
        para.gsub!(/&Acirc;/,  'Â') #'Capital A, circumflex accent Â
        para.gsub!(/&Atilde;/, 'Ã') #'Capital A, tilde        Ã
        para.gsub!(/&Auml;/,   'Ä') #'Capital A, umlaut       Ä
        para.gsub!(/&Aring;/,  'Å') #'Capital A, ring         Å
        para.gsub!(/&AElig;/,  'Æ') #'Capital AE ligature     Æ
        para.gsub!(/&Ccedil;/, 'Ç') #'Capital C, cedilla      Ç
        para.gsub!(/&Egrave;/, 'È') #'Capital E, grave accent È
        para.gsub!(/&Eacute;/, 'É') #'Capital E, acute accent É
        para.gsub!(/&Ecirc;/,  'Ê') #'Capital E, circumflex accent Ê
        para.gsub!(/&Euml;/,   'Ë') #'Capital E, umlaut       Ë
        para.gsub!(/&Igrave;/, 'Ì') #'Capital I, grave accent Ì
        para.gsub!(/&Iacute;/, 'Í') #'Capital I, acute accent Í
        para.gsub!(/&Icirc;/,  'Î') #'Capital I, circumflex accent Î
        para.gsub!(/&Iuml;/,   'Ï') #'Capital I, umlaut       Ï
        para.gsub!(/&ETH;/,    'Ð') #'Capital eth, Icelandic
        para.gsub!(/&Ntilde;/, 'Ñ') #'Capital N, tilde        Ñ
        para.gsub!(/&Ograve;/, 'Ò') #'Capital O, grave accent Ò
        para.gsub!(/&Oacute;/, 'Ó') #'Capital O, acute accent Ó
        para.gsub!(/&Ocirc;/,  'Ô') #'Capital O, circumflex accent Ô
        para.gsub!(/&Otilde;/, 'Õ') #'Capital O, tilde        Õ
        para.gsub!(/&Ouml;/,   'Ö') #'Capital O, umlaut       Ö
        para.gsub!(/&times;/,  '×') #'Multiply sign           ×
        para.gsub!(/&Oslash;/, 'Ø') #'Capital O, slash        Ø
        para.gsub!(/&Ugrave;/, 'Ù') #'Capital U, grave accent Ù
        para.gsub!(/&Uacute;/, 'Ú') #'Capital U, acute accent Ú
        para.gsub!(/&Ucirc;/,  'Û') #'Capital U, circumflex accent Û
        para.gsub!(/&Uuml;/,   'Ü') #'Capital U, umlaut       Ü
        para.gsub!(/&Yacute;/, 'Ý') #'Capital Y, acute accent Ý
        para.gsub!(/&THORN;/,  'Þ') #'Capital thorn, Icelandic     Þ
        para.gsub!(/&szlig;/,  'ß') #'Small sz ligature, German    ß
        para.gsub!(/&agrave;/, 'à') #'Small a, grave accent   à
        para.gsub!(/&aacute;/, 'á') #'Small a, acute accent   á
        para.gsub!(/&acirc;/,  'â') #'Small a, circumflex accent   â
        para.gsub!(/&atilde;/, 'ã') #'Small a, tilde          ã
        para.gsub!(/&auml;/,   'ä') #'Small a, umlaut         ä
        para.gsub!(/&aring;/,  'å') #'Small a, ring           å
        para.gsub!(/&aelig;/,  'æ') #'Small ae ligature       æ
        para.gsub!(/&ccedil;/, 'ç') #'Small c, cedilla        ç
        para.gsub!(/&egrave;/, 'è') #'Small e, grave accent   è
        para.gsub!(/&eacute;/, 'é') #'Small e, acute accent   é
        para.gsub!(/&ecirc;/,  'ê') #'Small e, circumflex accent   ê
        para.gsub!(/&euml;/,   'ë') #'Small e, umlaut         ë
        para.gsub!(/&igrave;/, 'ì') #'Small i, grave accent   ì
        para.gsub!(/&iacute;/, 'í') #'Small i, acute accent   í
        para.gsub!(/&icirc;/,  'î') #'Small i, circumflex accent   î
        para.gsub!(/&iuml;/,   'ï') #'Small i, umlaut         ï
        para.gsub!(/&eth;/,    'ð') #'Small eth, Icelandic    ð
        para.gsub!(/&ntilde;/, 'ñ') #'Small n, tilde          ñ
        para.gsub!(/&ograve;/, 'ò') #'Small o, grave accent   ò
        para.gsub!(/&oacute;/, 'ó') #'Small o, acute accent   ó
        para.gsub!(/&ocirc;/,  'ô') #'Small o, circumflex accent   ô
        para.gsub!(/&otilde;/, 'õ') #'Small o, tilde          õ
        para.gsub!(/&ouml;/,   'ö') #'Small o, umlaut         ö
        para.gsub!(/&divide;/, '÷') #'Divide sign             ÷
        para.gsub!(/&oslash;/, 'ø') #'Small o, slash          ø
        para.gsub!(/&ugrave;/, 'ù') #'Small u, grave accent   ù
        para.gsub!(/&uacute;/, 'ú') #'Small u, acute accent   ú
        para.gsub!(/&ucirc;/,  'û') #'Small u, circumflex accent   û
        para.gsub!(/&uuml;/,   'ü') #'Small u, umlaut         ü
        para.gsub!(/&yacute;/, 'ý') #'Small y, acute accent   ý
        para.gsub!(/&thorn;/,  'þ') #'Small thorn, Icelandic  þ
        para.gsub!(/&yuml;/,   'ÿ') #'Smally y, umlaut        ÿ
        ##
        para.gsub!(/\s\s+/,' ')
        para.gsub!(/\t+/,' ')
        #para.gsub!(/	+/,' ')
        #para.gsub!(/^(?:<(?:\/)?[bi]>)+$/i, '')
        tuned_file << para unless para == nil
      end
      tuned_file
    end
  end
  class Default < Html
    def initialize(data, filename, instruct)
      @data=data
      @filename=filename
      @instruct=instruct
    end
    def songsheet
      data=@data
      print "Convert to SiSU file from #{@filename}.html << gvim ,,#{@filename}.sst >\n" #: <<#{@@html_title}>>
      data=Default.new(data.collect, @filename, @instruct).space_paragraphs
      data=Default.new(data.collect, @filename, @instruct).multiline
      data=Default.new(data.collect.join.split("\n\n"), @filename, @instruct).markup_rules
      data=Default.new(data.collect, @filename, @instruct).markup_default
      data=MyOutput.new(data.collect, @filename, @instruct).hardOutput
    end
    def markup_default
      data=@data
      tuned_file=Array.new
      data.each do |para|
        para.gsub!(/<i>(Id\.?)(\s|$)/i,'/\{\1\}\2/')
        para.gsub!(/^(~\{\{ .+?)(<\/LI>\s*|<\/OL>\s*)+$/i,'\1')
        para.gsub!(/\/\{Id\.\s*<\/LI>\s*\}\//i,'/{Id.}/')
        tuned_file << para unless para == nil
      end
      tuned_file
    end
  end
end
def help
      puts <<WOK
conversion program
initial SiSU markup from other file formats

  zxy_convert --word    does initial conversion from word97 to sisu markup, expects [filename].doc (can also use --doc)
  zxy_convert --html    does initial conversion from html to sisu markup, expects [filename].html
  zxy_convert --default does initial conversion from defalt html to sisu markup, expects [filename].html

WOK
end
def do_word(argv, instruct)
  argv.each do  |f|
    if f =~/.+?\.doc$/
      @argv << f[/(.+?)\.doc$/, 1]
    else
      print "not .doc? << #{f} >> "
    end
  end
  @argv.each do |filename|
    system(%{wvWare -x #{@dir.path.home}/.sisu/convert/wvSiSU.xml #{filename}.doc > #{filename}.wv})
    file_array=IO.readlines("#{filename}.wv", "")
    CONVERT::WareWord97.new(file_array, filename, instruct).songsheet # metaverse created here
  end
end
def do_html(argv, instruct)
  argv.each do  |f|
    if f =~/.+?\.html?$/
      @argv << f[/(.+?)\.html?$/, 1]
    else
      print "not .html? << #{f} >> "
    end
  end
  @argv.each do |filename|
    file_end=if FileTest.file?("#{filename}.html")
      'html'
    elsif  FileTest.file?("#{filename}.htm")
      'htm'
    end
    file_array=IO.readlines("#{filename}.#{file_end}","\n\r")
    CONVERT::Html.new(file_array,filename,instruct).songsheet # metaverse created here
  end
end
def do_default(argv, instruct)
  argv.each do  |f|
    if f =~/.+?\.html$/
      @argv << f[/(.+?)\.html$/, 1]
    else
      print "not .html? << #{f} >> "
    end
  end
  @argv.each do |filename|
    file_array=IO.readlines("#{filename}.html", "\n\r")
    CONVERT::Default.new(file_array, filename, instruct).songsheet # metaverse created here
  end
end
def cases(argv, instruct)
  case instruct
  when/^--(word(97)?|doc)$/i #creates minimal sisu_small.gz package to send
      do_word(argv, instruct)
  when/^--(html)$/i #creates sisu.gz package to send
      do_html(argv, instruct)
  when/^--(default)$/i #creates sisu.gz package to send
      do_default(argv, instruct)
  else
      help
  end
end
$KCODE='u'
branch='v2'
@argv=Array.new
argv=$*
SiSU_version_dir=(argv.inspect=~/--v1/) ? 'v1' : 'v2'
SiSU_lib="sisu/#{SiSU_version_dir}"
require "#{SiSU_lib}/sysenv"
include SiSU_Env
@dir=SiSU_Env::Info_env.new
instruct = "#{argv[0].to_s}"
argv.shift
instruct.chomp!
instruct = "help" if instruct.nil? or instruct == "";
cases(argv, instruct)