class Typeout

# File lib/typeout.rb, line 9
def self.convert(text, sanitize = true)
  self.new(text.to_s).to_html(sanitize)
end

# File lib/typeout.rb, line 76
def archive(text)
  @archive << text
  "!a!r!c!h!i!v!e!#{@archive.length-1}!a!r!c!h!i!v!e!" # nobody's going to type that!
end

# File lib/typeout.rb, line 69
def archive_blockquotes(text)
  text.gsub(/^~{3,}\n(.+?)\n~{3,}$/m) do
    content = remove_excess_newlines(make_paragraphs(make_lists("\n\n#{$1}\n\n"))) # blockquotes support paragraphs and lists
    archive("<blockquote>#{content}</blockquote>")
  end
end

# File lib/typeout.rb, line 64
def archive_code(text)
  text = text.gsub(/^-{3,}\n(.+?)\n-{3,}$/m) { archive("<pre><code>#{$1}</code></pre>") }
  text.gsub(/`(?!\s)((?:\s*\S)+?)`/) { archive("<code>#{$1}</code>") }
end

# File lib/typeout.rb, line 60
def archive_html(text)
  text.gsub(/\[html\](.*?)\[\/html\]/mi) { archive(sanitize_html($1)) }
end

# File lib/typeout.rb, line 50
def clean_whitespace(text)
  text.gsub(/\r\n/, "\n").
    gsub(/\r/, "\n").
    gsub(/^ +$/, '')
end

# File lib/typeout.rb, line 122
def make_breaks(text)
  text.gsub(/([^\n])\n(?!\n)/m, "\\1<br />\n")
end

# File lib/typeout.rb, line 88
def make_headings(text)
  text.gsub(/^(={1,6})(.+?)=*?$/) do
    depth = $1.length
    "\n\n<h#{depth}>#{$2.strip}</h#{depth}>\n\n"
  end
end

# File lib/typeout.rb, line 126
def make_inlines(text)
  text = text.
    gsub(/^((?: ?[\w\(\)']){3,}:) (?!\s*$)/, '<b>\1</b> '). # spaces are not allowed directly in front of the colon, there must be a space after it, and the rest of the line can't be blank
    gsub(/!\((.+?)\)(?:\:([a-zA-Z0-9\_]+))?/) do
      if $2
        "<img src=\"#{archive($1)}\" class=\"#{$2}\" />"
      else
        "<img src=\"#{archive($1)}\" />"
      end
    end.
    gsub(/\[(.+?)\]\((.+?)\)(?:\:([a-zA-Z0-9\_]+))?/) do
      if $3
        "<a href=\"#{archive($2)}\" class=\"#{$3}\">#{archive($1)}</a>"
      else
        "<a href=\"#{archive($2)}\">#{archive($1)}</a>"
      end
    end.
    gsub(/([a-z0-9\.\-\_\+]+)@((?:[a-z0-9\-]{2,}\.)*)([a-z0-9\-]+\.)(com|org|net|biz|edu|info|gov|co\.uk|co\.us)/i) do
      "<a href=\"mailto:#{archive($1 + '@' + $2 + $3 + $4)}\">#{archive($1 + '@' + $2 + $3 + $4)}</a>"
    end.
    gsub(/(https?:\/\/)?((?:[a-z0-9\-]{2,}\.)*)([a-z0-9\-]+\.)(com|org|net|biz|edu|info|gov|co\.uk|co\.us)((?:\/(?:\.?[a-z0-9\-_=\?&;\%#]+)+)*\/?)/i) do # some of the magic in here is to not grab periods at the end of URLs
      if $1.nil?
        protocol = 'http://'
      else
        protocol = $1
      end
      "<a href=\"#{archive(protocol + $2 + $3 + $4 + $5)}\">#{archive(protocol + $2 + $3 + $4 + $5)}</a>"
    end

  inlines = [
    ['*', 'strong'],
    ['_', 'em'],
    ['`', 'code'],
    ['^', 'sup'],
    ['~', 'sub']
  ]

  inlines.each do |char, tag|
    char = Regexp.escape(char)
    re = /#{char}         # the opening character
         (?!\s)           # no spaces directly after the char
         (                # main capture group
         (?:\s*\S)+?      # because of no lookbehinds, every space must be followed by a non-space
         )                # with lookbehinds, the whole regex is: \*(?! )(.+?)(?<! )\*
         #{char}/x        # closing character, no spaces directly before it
    #re = /#{char}(.+?)#{char}/
    text.gsub!(re, "<#{tag}>\\1</#{tag}>")
  end

  text
end

# File lib/typeout.rb, line 95
def make_lists(text, base_level = nil)
  text.gsub(/^(#{Regexp.escape(base_level.to_s)}[\*\#])[\*\#]* [^\n]+?\n(?:\1[\*\#]* (?:[^\n])+?\n)*/m) do |content|
    level = $1

    content = make_lists(content, level)
    content = content.gsub(/^#{Regexp.escape(level)} (.+)$/, '<li>\1</li>')
    content = content.gsub(/<\/li>\n<([uo])l>/m, '<\1l>')

    if level.last == "*" # its an unordered list
      content = "<ul>\n#{content}</ul>"
    else
      content = "<ol>\n#{content}</ol>"
    end

    if base_level
      content += "</li>\n"
    else
      content += "\n"
    end
    content
  end
end

# File lib/typeout.rb, line 118
def make_paragraphs(text)
  text.gsub(/^(?!<|!a!r!c!h!i!v!e!)([^\n]+\n)+\n/m) { |content| "\n\n<p>\n#{make_breaks(content.strip)}</p>\n\n" }
end

# File lib/typeout.rb, line 56
def remove_excess_newlines(text)
  text.gsub(/\n{3,}/, "\n\n")
end

# File lib/typeout.rb, line 81
def retrieve_archive(text)
  @archive.each_with_index do |content, index|
    text = text.sub("!a!r!c!h!i!v!e!#{index}!a!r!c!h!i!v!e!") { content } # the block is so gsub won't substitute \&
  end
  text
end

# File lib/typeout.rb, line 42
def sanitize_html(text)
  if @sanitize
    Sanitize.clean(text, Sanitize::Config::RELAXED)
  else
    text
  end
end

# File lib/typeout.rb, line 13
def to_html(sanitize = true)
  @sanitize = sanitize

  text = self.dup

  text = clean_whitespace text
  text = "\n\n#{text}\n\n" # a few guaranteed newlines

  @archive = []

  text = archive_html text

  text = html_escape text

  text = archive_code text
  text = archive_blockquotes text

  text = make_headings text
  text = make_lists text
  text = make_paragraphs text
  text = make_inlines text

  text = remove_excess_newlines text

  text = retrieve_archive text

  text
end