habari2md 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: c2fb0d61a3de12d336d24ab9b5d3e47ec7b9a098
+  data.tar.gz: 063bedd9049364662e8e200efbf4a2a1c97a9f46
+SHA512:
+  metadata.gz: 17550bf8808495e447d04fe9f5475e170a28f8119319b89a5099aad97da5569b1d7a95c4cb3e7961c742536ca5851a73939053f25a1ef85c3ca255b6597fe907
+  data.tar.gz: 5c3a0c209e76eceb05092ed68e9fa1ab555f2bb63172357f70d068657f1e6094caa56e637916452f30fd9a0bdaa726708034cc45f6fb20a3d44024b292fa1ecb

checksums.yaml.gz.sig

@@ -0,0 +1,2 @@
+.����-��o�W�Q����<�C�^d3��_�A�垌��{)YT�o.᎝���B�m|��y���ĂDƠ���/�����2%�*��Q��O�nY�+��\!b����ղ
+j'��(s

data/.gitignore

@@ -0,0 +1,17 @@
+*.gem
+*.rbc
+.bundle
+.config
+.yardoc
+Gemfile.lock
+InstalledFiles
+_yardoc
+coverage
+doc/
+lib/bundler/man
+pkg
+rdoc
+spec/reports
+test/tmp
+test/version_tmp
+tmp

data/Gemfile

@@ -0,0 +1,4 @@
+source 'https://rubygems.org'
+
+# Specify your gem's dependencies in habari2md.gemspec
+gemspec

data/LICENSE.txt

@@ -0,0 +1,14 @@
+Copyright (c) 2014 Arnaud Berthomier
+
+habari2md 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 3 of the License, or (at your
+option) any later version.
+
+habari2md 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.
+
+You should have received a copy of the GNU General Public License
+along with this program.  If not, see <http://www.gnu.org/licenses/>.

data/README.md

@@ -0,0 +1,64 @@
+# Habari2md
+
+This is a dirty little Ruby program to export a [Habari][habari] blog to
+markdown format.  I used it to avoid installing PHP on a small VPS in order to
+run a tiny blog of ~2000 posts.
+
+The program makes a few assumptions about your setup, and this conditions what
+you should expect to get from it.
+
+  * It will connect to a MariaDB/MySQL database,
+  * fetch all of its posts and:
+    * dump one file per published post in the `out` directory ;
+    * use a filename like `YYYY-MM-DD-post-slug.md` where `YYYY-MM-DD` are the
+      year, month, and month day when a particular post was published ;
+    * and format a post header with:
+        
+        ```
+        title: The original post title
+        author: The author's username
+        ```
+
+This process can be pretty specific, and if it does not fit your setup, feel
+free to file an issue or, better, send a pull-request. ;)
+
+# Dependencies
+
+ * Ruby >= 1.9
+ * Python >= 2.x
+
+# Installation
+
+`gem install habari2md`
+
+# Usage
+
+```
+$ habari2md -h
+Usage: habari2md [options]
+    -o, --output [DIR]               Output directory
+    -s, --host [HOST]                Database host
+    -d, --db [DB]                    Database name
+    -u, --user [USER]                Database user
+    -p, --password [PASS]            Database password
+    -h, --help                       Show this message
+$ habari2md -o foobar -d my_blog_database -h localhost -u sql_user -p sql_password
+I, [2014-01-08T23:31:20.771303 #74090]  INFO -- : Exporting 12345 posts...
+I, [2014-01-08T23:31:50.618731 #74090]  INFO -- : 50% to go
+I, [2014-01-08T23:32:20.081583 #74090]  INFO -- : We're done.
+D, [2014-01-08T23:32:20.083582 #74090] DEBUG -- : Terminating 6 actors...
+W, [2014-01-08T23:32:20.084398 #74090]  WARN -- : Terminating task: type=:finalizer, meta={:method_name=>:__shutdown__}, status=:callwait
+➜
+```
+
+# License
+
+GPL 3.0
+
+Note: this distribution contains Aaron Swartz [html2text][html2text] GPL
+licensed program. As a matter of fact, we fork one process to convert each post
+from HTML to [Markdown][markdown], yay!
+
+[habari]: http://habariproject.org/
+[html2text]: http://www.aaronsw.com/2002/html2text/
+[markdown]: http://daringfireball.net/projects/markdown/

data/Rakefile

@@ -0,0 +1 @@
+require "bundler/gem_tasks"

data/bin/habari2md

@@ -0,0 +1,44 @@
+#!/usr/bin/env ruby
+
+$:.unshift(File.dirname(__FILE__) + '../lib')
+
+require 'optparse'
+require 'habari2md'
+
+options = {
+  out:      './out',
+  host:     'localhost',
+  user:     'root',
+  password: 'root',
+}
+
+OptionParser.new do |opts|
+  opts.banner = 'Usage: habari2md [options]'
+
+  opts.on("-o", "--output [DIR]", "Output directory") do |dir|
+    options[:out] = dir
+  end
+
+  opts.on("-s", "--host [HOST]", "Database host") do |host|
+    options[:host] = host
+  end
+
+  opts.on("-d", "--db [DB]", "Database name") do |name|
+    options[:db] = name
+  end
+
+  opts.on("-u", "--user [USER]", "Database user") do |user|
+    options[:user] = user
+  end
+
+  opts.on("-p", "--password [PASS]", "Database password") do |pass|
+    options[:password] = pass
+  end
+
+  opts.on("-h", "--help", "Show this message") do
+    puts opts
+    exit
+  end
+end.parse!
+
+Habari2md::Exporter.new(options).export_posts(options[:out])

data/certs/oz.pem

@@ -0,0 +1,20 @@
+-----BEGIN CERTIFICATE-----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-----END CERTIFICATE-----

data/habari2md.gemspec

@@ -0,0 +1,29 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'habari2md/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = "habari2md"
+  spec.version       = Habari2md::VERSION
+  spec.authors       = ["Arnaud Berthomier"]
+  spec.email         = ["oz@cyprio.net"]
+  spec.summary       = %q{Habari to markdown}
+  spec.description   = %q{Dump a Habari blog posts to Markdown format}
+  spec.homepage      = "https://github.com/oz/habari2md"
+  spec.license       = "GPL v3"
+
+  spec.files         = `git ls-files`.split($/)
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = []
+  spec.require_paths = ["lib"]
+  spec.cert_chain    = ["certs/oz.pem"]
+  spec.signing_key   = File.expand_path("~/.ssh/gem-private_key.pem") if $0 =~ /gem\z/
+
+  spec.add_development_dependency "bundler", "~> 1.5"
+  spec.add_development_dependency "rake"
+
+  spec.add_dependency "celluloid", "~> 0.15"
+  spec.add_dependency "sequel",    "~> 4.5"
+  spec.add_dependency "mysql",     "~> 2.9"
+end

data/lib/habari2md.rb

@@ -0,0 +1,164 @@
+#!/usr/bin/env ruby
+require 'fileutils'
+require 'pathname'
+
+require 'celluloid'
+require 'sequel'
+
+module Habari2md
+  # @class Habari2md::Text Text helpers
+  class Text
+    # Shameless snatch from Rails.
+    # @param [String] text
+    # @return [String]
+    def self.simple_format(text)
+      text = '' if text.nil?
+      text = text.dup
+      start_tag = '<p>'
+      text = text.to_str
+      text.gsub!(/\r\n?/, "\n")                    # \r\n and \r -> \n
+      text.gsub!(/\n\n+/, "</p>\n\n#{start_tag}")  # 2+ newline  -> paragraph
+      text.gsub!(/([^\n]\n)(?=[^\n])/, '\1<br />') # 1 newline   -> br
+      text.insert(0, start_tag)
+      text << '</p>'
+      return text
+    end
+
+    # Fork (!) html2text.py to convert form HTML to Markdown.
+    #
+    # @param [String] content
+    # @reutnr [String] Markdown content
+    def self.html2text(content)
+      IO.popen(html2text_script, "r+") do |io|
+        io.write content
+        io.close_write
+        content = io.read
+        io.close_read
+      end
+      content
+    end
+
+    protected
+
+    def self.html2text_script
+      @html2text ||= Pathname.new(File.dirname(__FILE__))
+                             .join('vendor', 'html2text.py').to_s
+    end
+  end
+
+  # @class Habari2md::Exporter
+  # @example Export stuff
+  #   worker = Habari2md::Exporter.new(db: 'foo', user: 'root')
+  #   worker.export_posts("./out")
+  class Exporter
+    attr_reader :db
+    include Celluloid
+    include Celluloid::Logger
+
+    def initialize(opts = {})
+      @db = Sequel.connect(db_uri opts)
+      @counter = 0
+      @halfway = 0
+
+      # Cache users
+      @users = @db[:users].all.inject({}) do |cache, user|
+        cache.merge!([user[:id]] => user)
+      end
+    end
+
+    def posts
+      db[:posts].order(:modified)
+    end
+
+    # @return [Hash]
+    def user(id)
+      @users.fetch(id, {})
+    end
+
+    def export_posts(directory)
+      FileUtils.mkdir_p(directory) unless File.directory?(directory)
+
+      @counter = posts.count
+      @halfway = @counter / 2
+
+      info "Exporting #{@counter} posts..."
+
+      pool = Habari2md::PostExporter.pool(args: [directory, current_actor])
+      posts.each { |post| pool.async.export(post) }
+
+      wait(:done)
+      info "We're done."
+    end
+
+    # Called by PostExport when an export operation has finished.
+    def post_exported(post_id)
+      @counter -= 1
+      info "50% to go" if @counter == @halfway
+      signal(:done) if @counter == 0
+    end
+
+    protected
+
+    def db_uri(opts)
+      "mysql://#{opts[:user]}:#{opts[:password]}@#{opts[:host]}/#{opts[:db]}"
+    end
+  end
+
+  # @class Habari2md::PostExporter Export one post
+  class PostExporter
+    include Celluloid
+
+    # Output directory
+    attr_reader :dir
+
+    # Manager actor
+    attr_reader :manager
+
+    def initialize(dest_dir, manager_actor)
+      @dir = Pathname.new(dest_dir)
+      @manager = manager_actor
+    end
+
+    # Placeholder title for untitled posts
+    def untitled
+      "Untitled"
+    end
+
+    # Signal the managing actor when a post has been exported
+    def done(post = {})
+      manager.post_exported(post[:id])
+    end
+
+    # Export one post to disk
+    # @param [Hash] post
+    def export(post)
+      # Ignore deleted posts and drafts.
+      return done(post) unless published?(post)
+
+      author = manager.user(post[:user_id])[:username]
+      title = post[:title].gsub(/[\r\n]/, '')
+      title = untitled if title == ""
+      date = Time.strptime(post[:pubdate].to_s, "%s").strftime("%Y-%m-%d")
+      filename = dir.join("#{date}-#{post[:slug]}.md")
+      return done(post) if File.exists?(filename) && ENV['FORCE'] == nil
+
+      # Make sure content is at least formatted with <p> tags before
+      # conversion.
+      content = Habari2md::Text.simple_format(post[:content])
+      File.open(filename, 'w+') do |fh|
+        fh << "---\n"
+        fh << "title: #{title}\n"
+        fh << "author: #{author}\n" unless author == nil
+        fh << "---\n\n"
+        fh << Habari2md::Text.html2text(content)
+      end
+
+      done(post)
+    end
+
+    # This actually depends on the values in the poststatus table.
+    def published?(post)
+      post[:status] == 2
+    end
+  end
+end

data/lib/habari2md/version.rb

@@ -0,0 +1,3 @@
+module Habari2md
+  VERSION = "0.0.1"
+end

data/lib/vendor/html2text.py

@@ -0,0 +1,914 @@
+#!/usr/bin/env python
+"""html2text: Turn HTML into equivalent Markdown-structured text."""
+__version__ = "3.200.3"
+__author__ = "Aaron Swartz (me@aaronsw.com)"
+__copyright__ = "(C) 2004-2008 Aaron Swartz. GNU GPL 3."
+__contributors__ = ["Martin 'Joey' Schulze", "Ricardo Reyes", "Kevin Jay North"]
+
+# TODO:
+#   Support decoded entities with unifiable.
+
+try:
+    True
+except NameError:
+    setattr(__builtins__, 'True', 1)
+    setattr(__builtins__, 'False', 0)
+
+def has_key(x, y):
+    if hasattr(x, 'has_key'): return x.has_key(y)
+    else: return y in x
+
+try:
+    import htmlentitydefs
+    import urlparse
+    import HTMLParser
+except ImportError: #Python3
+    import html.entities as htmlentitydefs
+    import urllib.parse as urlparse
+    import html.parser as HTMLParser
+try: #Python3
+    import urllib.request as urllib
+except:
+    import urllib
+import optparse, re, sys, codecs, types
+
+try: from textwrap import wrap
+except: pass
+
+# Use Unicode characters instead of their ascii psuedo-replacements
+UNICODE_SNOB = 0
+
+# Escape all special characters.  Output is less readable, but avoids corner case formatting issues.
+ESCAPE_SNOB = 0
+
+# Put the links after each paragraph instead of at the end.
+LINKS_EACH_PARAGRAPH = 0
+
+# Wrap long lines at position. 0 for no wrapping. (Requires Python 2.3.)
+BODY_WIDTH = 78
+
+# Don't show internal links (href="#local-anchor") -- corresponding link targets
+# won't be visible in the plain text file anyway.
+SKIP_INTERNAL_LINKS = True
+
+# Use inline, rather than reference, formatting for images and links
+INLINE_LINKS = True
+
+# Number of pixels Google indents nested lists
+GOOGLE_LIST_INDENT = 36
+
+IGNORE_ANCHORS = False
+IGNORE_IMAGES = False
+IGNORE_EMPHASIS = False
+
+### Entity Nonsense ###
+
+def name2cp(k):
+    if k == 'apos': return ord("'")
+    if hasattr(htmlentitydefs, "name2codepoint"): # requires Python 2.3
+        return htmlentitydefs.name2codepoint[k]
+    else:
+        k = htmlentitydefs.entitydefs[k]
+        if k.startswith("&#") and k.endswith(";"): return int(k[2:-1]) # not in latin-1
+        return ord(codecs.latin_1_decode(k)[0])
+
+unifiable = {'rsquo':"'", 'lsquo':"'", 'rdquo':'"', 'ldquo':'"',
+'copy':'(C)', 'mdash':'--', 'nbsp':' ', 'rarr':'->', 'larr':'<-', 'middot':'*',
+'ndash':'-', 'oelig':'oe', 'aelig':'ae',
+'agrave':'a', 'aacute':'a', 'acirc':'a', 'atilde':'a', 'auml':'a', 'aring':'a',
+'egrave':'e', 'eacute':'e', 'ecirc':'e', 'euml':'e',
+'igrave':'i', 'iacute':'i', 'icirc':'i', 'iuml':'i',
+'ograve':'o', 'oacute':'o', 'ocirc':'o', 'otilde':'o', 'ouml':'o',
+'ugrave':'u', 'uacute':'u', 'ucirc':'u', 'uuml':'u',
+'lrm':'', 'rlm':''}
+
+unifiable_n = {}
+
+for k in unifiable.keys():
+    unifiable_n[name2cp(k)] = unifiable[k]
+
+### End Entity Nonsense ###
+
+def onlywhite(line):
+    """Return true if the line does only consist of whitespace characters."""
+    for c in line:
+        if c is not ' ' and c is not '  ':
+            return c is ' '
+    return line
+
+def hn(tag):
+    if tag[0] == 'h' and len(tag) == 2:
+        try:
+            n = int(tag[1])
+            if n in range(1, 10): return n
+        except ValueError: return 0
+
+def dumb_property_dict(style):
+    """returns a hash of css attributes"""
+    return dict([(x.strip(), y.strip()) for x, y in [z.split(':', 1) for z in style.split(';') if ':' in z]]);
+
+def dumb_css_parser(data):
+    """returns a hash of css selectors, each of which contains a hash of css attributes"""
+    # remove @import sentences
+    data += ';'
+    importIndex = data.find('@import')
+    while importIndex != -1:
+        data = data[0:importIndex] + data[data.find(';', importIndex) + 1:]
+        importIndex = data.find('@import')
+
+    # parse the css. reverted from dictionary compehension in order to support older pythons
+    elements =  [x.split('{') for x in data.split('}') if '{' in x.strip()]
+    try:
+        elements = dict([(a.strip(), dumb_property_dict(b)) for a, b in elements])
+    except ValueError:
+        elements = {} # not that important
+
+    return elements
+
+def element_style(attrs, style_def, parent_style):
+    """returns a hash of the 'final' style attributes of the element"""
+    style = parent_style.copy()
+    if 'class' in attrs:
+        for css_class in attrs['class'].split():
+            css_style = style_def['.' + css_class]
+            style.update(css_style)
+    if 'style' in attrs:
+        immediate_style = dumb_property_dict(attrs['style'])
+        style.update(immediate_style)
+    return style
+
+def google_list_style(style):
+    """finds out whether this is an ordered or unordered list"""
+    if 'list-style-type' in style:
+        list_style = style['list-style-type']
+        if list_style in ['disc', 'circle', 'square', 'none']:
+            return 'ul'
+    return 'ol'
+
+def google_has_height(style):
+    """check if the style of the element has the 'height' attribute explicitly defined"""
+    if 'height' in style:
+        return True
+    return False
+
+def google_text_emphasis(style):
+    """return a list of all emphasis modifiers of the element"""
+    emphasis = []
+    if 'text-decoration' in style:
+        emphasis.append(style['text-decoration'])
+    if 'font-style' in style:
+        emphasis.append(style['font-style'])
+    if 'font-weight' in style:
+        emphasis.append(style['font-weight'])
+    return emphasis
+
+def google_fixed_width_font(style):
+    """check if the css of the current element defines a fixed width font"""
+    font_family = ''
+    if 'font-family' in style:
+        font_family = style['font-family']
+    if 'Courier New' == font_family or 'Consolas' == font_family:
+        return True
+    return False
+
+def list_numbering_start(attrs):
+    """extract numbering from list element attributes"""
+    if 'start' in attrs:
+        return int(attrs['start']) - 1
+    else:
+        return 0
+
+class HTML2Text(HTMLParser.HTMLParser):
+    def __init__(self, out=None, baseurl=''):
+        HTMLParser.HTMLParser.__init__(self)
+
+        # Config options
+        self.unicode_snob = UNICODE_SNOB
+        self.escape_snob = ESCAPE_SNOB
+        self.links_each_paragraph = LINKS_EACH_PARAGRAPH
+        self.body_width = BODY_WIDTH
+        self.skip_internal_links = SKIP_INTERNAL_LINKS
+        self.inline_links = INLINE_LINKS
+        self.google_list_indent = GOOGLE_LIST_INDENT
+        self.ignore_links = IGNORE_ANCHORS
+        self.ignore_images = IGNORE_IMAGES
+        self.ignore_emphasis = IGNORE_EMPHASIS
+        self.google_doc = False
+        self.ul_item_mark = '*'
+        self.emphasis_mark = '_'
+        self.strong_mark = '**'
+
+        if out is None:
+            self.out = self.outtextf
+        else:
+            self.out = out
+
+        self.outtextlist = []  # empty list to store output characters before they are "joined"
+
+        try:
+            self.outtext = unicode()
+        except NameError:  # Python3
+            self.outtext = str()
+
+        self.quiet = 0
+        self.p_p = 0  # number of newline character to print before next output
+        self.outcount = 0
+        self.start = 1
+        self.space = 0
+        self.a = []
+        self.astack = []
+        self.maybe_automatic_link = None
+        self.absolute_url_matcher = re.compile(r'^[a-zA-Z+]+://')
+        self.acount = 0
+        self.list = []
+        self.blockquote = 0
+        self.pre = 0
+        self.startpre = 0
+        self.code = False
+        self.br_toggle = ''
+        self.lastWasNL = 0
+        self.lastWasList = False
+        self.style = 0
+        self.style_def = {}
+        self.tag_stack = []
+        self.emphasis = 0
+        self.drop_white_space = 0
+        self.inheader = False
+        self.abbr_title = None  # current abbreviation definition
+        self.abbr_data = None  # last inner HTML (for abbr being defined)
+        self.abbr_list = {}  # stack of abbreviations to write later
+        self.baseurl = baseurl
+
+        try: del unifiable_n[name2cp('nbsp')]
+        except KeyError: pass
+        unifiable['nbsp'] = '&nbsp_place_holder;'
+
+
+    def feed(self, data):
+        data = data.replace("</' + 'script>", "</ignore>")
+        HTMLParser.HTMLParser.feed(self, data)
+
+    def handle(self, data):
+        self.feed(data)
+        self.feed("")
+        return self.optwrap(self.close())
+
+    def outtextf(self, s):
+        self.outtextlist.append(s)
+        if s: self.lastWasNL = s[-1] == '\n'
+
+    def close(self):
+        HTMLParser.HTMLParser.close(self)
+
+        self.pbr()
+        self.o('', 0, 'end')
+
+        self.outtext = self.outtext.join(self.outtextlist)
+        if self.unicode_snob:
+            nbsp = unichr(name2cp('nbsp'))
+        else:
+            nbsp = u' '
+        self.outtext = self.outtext.replace(u'&nbsp_place_holder;', nbsp)
+
+        return self.outtext
+
+    def handle_charref(self, c):
+        self.o(self.charref(c), 1)
+
+    def handle_entityref(self, c):
+        self.o(self.entityref(c), 1)
+
+    def handle_starttag(self, tag, attrs):
+        self.handle_tag(tag, attrs, 1)
+
+    def handle_endtag(self, tag):
+        self.handle_tag(tag, None, 0)
+
+    def previousIndex(self, attrs):
+        """ returns the index of certain set of attributes (of a link) in the
+            self.a list
+
+            If the set of attributes is not found, returns None
+        """
+        if not has_key(attrs, 'href'): return None
+
+        i = -1
+        for a in self.a:
+            i += 1
+            match = 0
+
+            if has_key(a, 'href') and a['href'] == attrs['href']:
+                if has_key(a, 'title') or has_key(attrs, 'title'):
+                        if (has_key(a, 'title') and has_key(attrs, 'title') and
+                            a['title'] == attrs['title']):
+                            match = True
+                else:
+                    match = True
+
+            if match: return i
+
+    def drop_last(self, nLetters):
+        if not self.quiet:
+            self.outtext = self.outtext[:-nLetters]
+
+    def handle_emphasis(self, start, tag_style, parent_style):
+        """handles various text emphases"""
+        tag_emphasis = google_text_emphasis(tag_style)
+        parent_emphasis = google_text_emphasis(parent_style)
+
+        # handle Google's text emphasis
+        strikethrough =  'line-through' in tag_emphasis and self.hide_strikethrough
+        bold = 'bold' in tag_emphasis and not 'bold' in parent_emphasis
+        italic = 'italic' in tag_emphasis and not 'italic' in parent_emphasis
+        fixed = google_fixed_width_font(tag_style) and not \
+                google_fixed_width_font(parent_style) and not self.pre
+
+        if start:
+            # crossed-out text must be handled before other attributes
+            # in order not to output qualifiers unnecessarily
+            if bold or italic or fixed:
+                self.emphasis += 1
+            if strikethrough:
+                self.quiet += 1
+            if italic:
+                self.o(self.emphasis_mark)
+                self.drop_white_space += 1
+            if bold:
+                self.o(self.strong_mark)
+                self.drop_white_space += 1
+            if fixed:
+                self.o('`')
+                self.drop_white_space += 1
+                self.code = True
+        else:
+            if bold or italic or fixed:
+                # there must not be whitespace before closing emphasis mark
+                self.emphasis -= 1
+                self.space = 0
+                self.outtext = self.outtext.rstrip()
+            if fixed:
+                if self.drop_white_space:
+                    # empty emphasis, drop it
+                    self.drop_last(1)
+                    self.drop_white_space -= 1
+                else:
+                    self.o('`')
+                self.code = False
+            if bold:
+                if self.drop_white_space:
+                    # empty emphasis, drop it
+                    self.drop_last(2)
+                    self.drop_white_space -= 1
+                else:
+                    self.o(self.strong_mark)
+            if italic:
+                if self.drop_white_space:
+                    # empty emphasis, drop it
+                    self.drop_last(1)
+                    self.drop_white_space -= 1
+                else:
+                    self.o(self.emphasis_mark)
+            # space is only allowed after *all* emphasis marks
+            if (bold or italic) and not self.emphasis:
+                    self.o(" ")
+            if strikethrough:
+                self.quiet -= 1
+
+    def handle_tag(self, tag, attrs, start):
+        #attrs = fixattrs(attrs)
+        if attrs is None:
+            attrs = {}
+        else:
+            attrs = dict(attrs)
+
+        if self.google_doc:
+            # the attrs parameter is empty for a closing tag. in addition, we
+            # need the attributes of the parent nodes in order to get a
+            # complete style description for the current element. we assume
+            # that google docs export well formed html.
+            parent_style = {}
+            if start:
+                if self.tag_stack:
+                  parent_style = self.tag_stack[-1][2]
+                tag_style = element_style(attrs, self.style_def, parent_style)
+                self.tag_stack.append((tag, attrs, tag_style))
+            else:
+                dummy, attrs, tag_style = self.tag_stack.pop()
+                if self.tag_stack:
+                    parent_style = self.tag_stack[-1][2]
+
+        if hn(tag):
+            self.p()
+            if start:
+                self.inheader = True
+                self.o(hn(tag)*"#" + ' ')
+            else:
+                self.inheader = False
+                return # prevent redundant emphasis marks on headers
+
+        if tag in ['p', 'div']:
+            if self.google_doc:
+                if start and google_has_height(tag_style):
+                    self.p()
+                else:
+                    self.soft_br()
+            else:
+                self.p()
+
+        if tag == "br" and start: self.o("  \n")
+
+        if tag == "hr" and start:
+            self.p()
+            self.o("* * *")
+            self.p()
+
+        if tag in ["head", "style", 'script']:
+            if start: self.quiet += 1
+            else: self.quiet -= 1
+
+        if tag == "style":
+            if start: self.style += 1
+            else: self.style -= 1
+
+        if tag in ["body"]:
+            self.quiet = 0 # sites like 9rules.com never close <head>
+
+        if tag == "blockquote":
+            if start:
+                self.p(); self.o('> ', 0, 1); self.start = 1
+                self.blockquote += 1
+            else:
+                self.blockquote -= 1
+                self.p()
+
+        if tag in ['em', 'i', 'u'] and not self.ignore_emphasis: self.o(self.emphasis_mark)
+        if tag in ['strong', 'b'] and not self.ignore_emphasis: self.o(self.strong_mark)
+        if tag in ['del', 'strike', 's']:
+            if start:
+                self.o("<"+tag+">")
+            else:
+                self.o("</"+tag+">")
+
+        if self.google_doc:
+            if not self.inheader:
+                # handle some font attributes, but leave headers clean
+                self.handle_emphasis(start, tag_style, parent_style)
+
+        if tag in ["code", "tt"] and not self.pre: self.o('`') #TODO: `` `this` ``
+        if tag == "abbr":
+            if start:
+                self.abbr_title = None
+                self.abbr_data = ''
+                if has_key(attrs, 'title'):
+                    self.abbr_title = attrs['title']
+            else:
+                if self.abbr_title != None:
+                    self.abbr_list[self.abbr_data] = self.abbr_title
+                    self.abbr_title = None
+                self.abbr_data = ''
+
+        if tag == "a" and not self.ignore_links:
+            if start:
+                if has_key(attrs, 'href') and not (self.skip_internal_links and attrs['href'].startswith('#')):
+                    self.astack.append(attrs)
+                    self.maybe_automatic_link = attrs['href']
+                else:
+                    self.astack.append(None)
+            else:
+                if self.astack:
+                    a = self.astack.pop()
+                    if self.maybe_automatic_link:
+                        self.maybe_automatic_link = None
+                    elif a:
+                        if self.inline_links:
+                            self.o("](" + escape_md(a['href']) + ")")
+                        else:
+                            i = self.previousIndex(a)
+                            if i is not None:
+                                a = self.a[i]
+                            else:
+                                self.acount += 1
+                                a['count'] = self.acount
+                                a['outcount'] = self.outcount
+                                self.a.append(a)
+                            self.o("][" + str(a['count']) + "]")
+
+        if tag == "img" and start and not self.ignore_images:
+            if has_key(attrs, 'src'):
+                attrs['href'] = attrs['src']
+                alt = attrs.get('alt', '')
+                self.o("![" + escape_md(alt) + "]")
+
+                if self.inline_links:
+                    self.o("(" + escape_md(attrs['href']) + ")")
+                else:
+                    i = self.previousIndex(attrs)
+                    if i is not None:
+                        attrs = self.a[i]
+                    else:
+                        self.acount += 1
+                        attrs['count'] = self.acount
+                        attrs['outcount'] = self.outcount
+                        self.a.append(attrs)
+                    self.o("[" + str(attrs['count']) + "]")
+
+        if tag == 'dl' and start: self.p()
+        if tag == 'dt' and not start: self.pbr()
+        if tag == 'dd' and start: self.o('    ')
+        if tag == 'dd' and not start: self.pbr()
+
+        if tag in ["ol", "ul"]:
+            # Google Docs create sub lists as top level lists
+            if (not self.list) and (not self.lastWasList):
+                self.p()
+            if start:
+                if self.google_doc:
+                    list_style = google_list_style(tag_style)
+                else:
+                    list_style = tag
+                numbering_start = list_numbering_start(attrs)
+                self.list.append({'name':list_style, 'num':numbering_start})
+            else:
+                if self.list: self.list.pop()
+            self.lastWasList = True
+        else:
+            self.lastWasList = False
+
+        if tag == 'li':
+            self.pbr()
+            if start:
+                if self.list: li = self.list[-1]
+                else: li = {'name':'ul', 'num':0}
+                if self.google_doc:
+                    nest_count = self.google_nest_count(tag_style)
+                else:
+                    nest_count = len(self.list)
+                self.o("  " * nest_count) #TODO: line up <ol><li>s > 9 correctly.
+                if li['name'] == "ul": self.o(self.ul_item_mark + " ")
+                elif li['name'] == "ol":
+                    li['num'] += 1
+                    self.o(str(li['num'])+". ")
+                self.start = 1
+
+        if tag in ["table", "tr"] and start: self.p()
+        if tag == 'td': self.pbr()
+
+        if tag == "pre":
+            if start:
+                self.startpre = 1
+                self.pre = 1
+            else:
+                self.pre = 0
+            self.p()
+
+    def pbr(self):
+        if self.p_p == 0:
+            self.p_p = 1
+
+    def p(self):
+        self.p_p = 2
+
+    def soft_br(self):
+        self.pbr()
+        self.br_toggle = '  '
+
+    def o(self, data, puredata=0, force=0):
+        if self.abbr_data is not None:
+            self.abbr_data += data
+
+        if not self.quiet:
+            if self.google_doc:
+                # prevent white space immediately after 'begin emphasis' marks ('**' and '_')
+                lstripped_data = data.lstrip()
+                if self.drop_white_space and not (self.pre or self.code):
+                    data = lstripped_data
+                if lstripped_data != '':
+                    self.drop_white_space = 0
+
+            if puredata and not self.pre:
+                data = re.sub('\s+', ' ', data)
+                if data and data[0] == ' ':
+                    self.space = 1
+                    data = data[1:]
+            if not data and not force: return
+
+            if self.startpre:
+                #self.out(" :") #TODO: not output when already one there
+                if not data.startswith("\n"):  # <pre>stuff...
+                    data = "\n" + data
+
+            bq = (">" * self.blockquote)
+            if not (force and data and data[0] == ">") and self.blockquote: bq += " "
+
+            if self.pre:
+                if not self.list:
+                    bq += "    "
+                #else: list content is already partially indented
+                for i in xrange(len(self.list)):
+                    bq += "    "
+                data = data.replace("\n", "\n"+bq)
+
+            if self.startpre:
+                self.startpre = 0
+                if self.list:
+                    data = data.lstrip("\n") # use existing initial indentation
+
+            if self.start:
+                self.space = 0
+                self.p_p = 0
+                self.start = 0
+
+            if force == 'end':
+                # It's the end.
+                self.p_p = 0
+                self.out("\n")
+                self.space = 0
+
+            if self.p_p:
+                self.out((self.br_toggle+'\n'+bq)*self.p_p)
+                self.space = 0
+                self.br_toggle = ''
+
+            if self.space:
+                if not self.lastWasNL: self.out(' ')
+                self.space = 0
+
+            if self.a and ((self.p_p == 2 and self.links_each_paragraph) or force == "end"):
+                if force == "end": self.out("\n")
+
+                newa = []
+                for link in self.a:
+                    if self.outcount > link['outcount']:
+                        self.out("   ["+ str(link['count']) +"]: " + urlparse.urljoin(self.baseurl, link['href']))
+                        if has_key(link, 'title'): self.out(" ("+link['title']+")")
+                        self.out("\n")
+                    else:
+                        newa.append(link)
+
+                if self.a != newa: self.out("\n") # Don't need an extra line when nothing was done.
+
+                self.a = newa
+
+            if self.abbr_list and force == "end":
+                for abbr, definition in self.abbr_list.items():
+                    self.out("  *[" + abbr + "]: " + definition + "\n")
+
+            self.p_p = 0
+            self.out(data)
+            self.outcount += 1
+
+    def handle_data(self, data):
+        if r'\/script>' in data: self.quiet -= 1
+
+        if self.style:
+            self.style_def.update(dumb_css_parser(data))
+
+        if not self.maybe_automatic_link is None:
+            href = self.maybe_automatic_link
+            if href == data and self.absolute_url_matcher.match(href):
+                self.o("<" + data + ">")
+                return
+            else:
+                self.o("[")
+                self.maybe_automatic_link = None
+
+        if not self.code and not self.pre:
+            data = escape_md_section(data, snob=self.escape_snob)
+        self.o(data, 1)
+
+    def unknown_decl(self, data): pass
+
+    def charref(self, name):
+        if name[0] in ['x','X']:
+            c = int(name[1:], 16)
+        else:
+            c = int(name)
+
+        if not self.unicode_snob and c in unifiable_n.keys():
+            return unifiable_n[c]
+        else:
+            try:
+                return unichr(c)
+            except NameError: #Python3
+                return chr(c)
+
+    def entityref(self, c):
+        if not self.unicode_snob and c in unifiable.keys():
+            return unifiable[c]
+        else:
+            try: name2cp(c)
+            except KeyError: return "&" + c + ';'
+            else:
+                try:
+                    return unichr(name2cp(c))
+                except NameError: #Python3
+                    return chr(name2cp(c))
+
+    def replaceEntities(self, s):
+        s = s.group(1)
+        if s[0] == "#":
+            return self.charref(s[1:])
+        else: return self.entityref(s)
+
+    r_unescape = re.compile(r"&(#?[xX]?(?:[0-9a-fA-F]+|\w{1,8}));")
+    def unescape(self, s):
+        return self.r_unescape.sub(self.replaceEntities, s)
+
+    def google_nest_count(self, style):
+        """calculate the nesting count of google doc lists"""
+        nest_count = 0
+        if 'margin-left' in style:
+            nest_count = int(style['margin-left'][:-2]) / self.google_list_indent
+        return nest_count
+
+
+    def optwrap(self, text):
+        """Wrap all paragraphs in the provided text."""
+        if not self.body_width:
+            return text
+
+        assert wrap, "Requires Python 2.3."
+        result = ''
+        newlines = 0
+        for para in text.split("\n"):
+            if len(para) > 0:
+                if not skipwrap(para):
+                    result += "\n".join(wrap(para, self.body_width))
+                    if para.endswith('  '):
+                        result += "  \n"
+                        newlines = 1
+                    else:
+                        result += "\n\n"
+                        newlines = 2
+                else:
+                    if not onlywhite(para):
+                        result += para + "\n"
+                        newlines = 1
+            else:
+                if newlines < 2:
+                    result += "\n"
+                    newlines += 1
+        return result
+
+ordered_list_matcher = re.compile(r'\d+\.\s')
+unordered_list_matcher = re.compile(r'[-\*\+]\s')
+md_chars_matcher = re.compile(r"([\\\[\]\(\)])")
+md_chars_matcher_all = re.compile(r"([`\*_{}\[\]\(\)#!])")
+md_dot_matcher = re.compile(r"""
+    ^             # start of line
+    (\s*\d+)      # optional whitespace and a number
+    (\.)          # dot
+    (?=\s)        # lookahead assert whitespace
+    """, re.MULTILINE | re.VERBOSE)
+md_plus_matcher = re.compile(r"""
+    ^
+    (\s*)
+    (\+)
+    (?=\s)
+    """, flags=re.MULTILINE | re.VERBOSE)
+md_dash_matcher = re.compile(r"""
+    ^
+    (\s*)
+    (-)
+    (?=\s|\-)     # followed by whitespace (bullet list, or spaced out hr)
+                  # or another dash (header or hr)
+    """, flags=re.MULTILINE | re.VERBOSE)
+slash_chars = r'\`*_{}[]()#+-.!'
+md_backslash_matcher = re.compile(r'''
+    (\\)          # match one slash
+    (?=[%s])      # followed by a char that requires escaping
+    ''' % re.escape(slash_chars),
+    flags=re.VERBOSE)
+
+def skipwrap(para):
+    # If the text begins with four spaces or one tab, it's a code block; don't wrap
+    if para[0:4] == '    ' or para[0] == '\t':
+        return True
+    # If the text begins with only two "--", possibly preceded by whitespace, that's
+    # an emdash; so wrap.
+    stripped = para.lstrip()
+    if stripped[0:2] == "--" and len(stripped) > 2 and stripped[2] != "-":
+        return False
+    # I'm not sure what this is for; I thought it was to detect lists, but there's
+    # a <br>-inside-<span> case in one of the tests that also depends upon it.
+    if stripped[0:1] == '-' or stripped[0:1] == '*':
+        return True
+    # If the text begins with a single -, *, or +, followed by a space, or an integer,
+    # followed by a ., followed by a space (in either case optionally preceeded by
+    # whitespace), it's a list; don't wrap.
+    if ordered_list_matcher.match(stripped) or unordered_list_matcher.match(stripped):
+        return True
+    return False
+
+def wrapwrite(text):
+    text = text.encode('utf-8')
+    try: #Python3
+        sys.stdout.buffer.write(text)
+    except AttributeError:
+        sys.stdout.write(text)
+
+def html2text(html, baseurl=''):
+    h = HTML2Text(baseurl=baseurl)
+    return h.handle(html)
+
+def unescape(s, unicode_snob=False):
+    h = HTML2Text()
+    h.unicode_snob = unicode_snob
+    return h.unescape(s)
+
+def escape_md(text):
+    """Escapes markdown-sensitive characters within other markdown constructs."""
+    return md_chars_matcher.sub(r"\\\1", text)
+
+def escape_md_section(text, snob=False):
+    """Escapes markdown-sensitive characters across whole document sections."""
+    text = md_backslash_matcher.sub(r"\\\1", text)
+    if snob:
+        text = md_chars_matcher_all.sub(r"\\\1", text)
+    text = md_dot_matcher.sub(r"\1\\\2", text)
+    text = md_plus_matcher.sub(r"\1\\\2", text)
+    text = md_dash_matcher.sub(r"\1\\\2", text)
+    return text
+
+
+def main():
+    baseurl = ''
+
+    p = optparse.OptionParser('%prog [(filename|url) [encoding]]',
+                              version='%prog ' + __version__)
+    p.add_option("--ignore-emphasis", dest="ignore_emphasis", action="store_true",
+        default=IGNORE_EMPHASIS, help="don't include any formatting for emphasis")
+    p.add_option("--ignore-links", dest="ignore_links", action="store_true",
+        default=IGNORE_ANCHORS, help="don't include any formatting for links")
+    p.add_option("--ignore-images", dest="ignore_images", action="store_true",
+        default=IGNORE_IMAGES, help="don't include any formatting for images")
+    p.add_option("-g", "--google-doc", action="store_true", dest="google_doc",
+        default=False, help="convert an html-exported Google Document")
+    p.add_option("-d", "--dash-unordered-list", action="store_true", dest="ul_style_dash",
+        default=False, help="use a dash rather than a star for unordered list items")
+    p.add_option("-e", "--asterisk-emphasis", action="store_true", dest="em_style_asterisk",
+        default=False, help="use an asterisk rather than an underscore for emphasized text")
+    p.add_option("-b", "--body-width", dest="body_width", action="store", type="int",
+        default=BODY_WIDTH, help="number of characters per output line, 0 for no wrap")
+    p.add_option("-i", "--google-list-indent", dest="list_indent", action="store", type="int",
+        default=GOOGLE_LIST_INDENT, help="number of pixels Google indents nested lists")
+    p.add_option("-s", "--hide-strikethrough", action="store_true", dest="hide_strikethrough",
+        default=False, help="hide strike-through text. only relevant when -g is specified as well")
+    p.add_option("--escape-all", action="store_true", dest="escape_snob",
+        default=False, help="Escape all special characters.  Output is less readable, but avoids corner case formatting issues.")
+    (options, args) = p.parse_args()
+
+    # process input
+    encoding = "utf-8"
+    if len(args) > 0:
+        file_ = args[0]
+        if len(args) == 2:
+            encoding = args[1]
+        if len(args) > 2:
+            p.error('Too many arguments')
+
+        if file_.startswith('http://') or file_.startswith('https://'):
+            baseurl = file_
+            j = urllib.urlopen(baseurl)
+            data = j.read()
+            if encoding is None:
+                try:
+                    from feedparser import _getCharacterEncoding as enc
+                except ImportError:
+                    enc = lambda x, y: ('utf-8', 1)
+                encoding = enc(j.headers, data)[0]
+                if encoding == 'us-ascii':
+                    encoding = 'utf-8'
+        else:
+            data = open(file_, 'rb').read()
+            if encoding is None:
+                try:
+                    from chardet import detect
+                except ImportError:
+                    detect = lambda x: {'encoding': 'utf-8'}
+                encoding = detect(data)['encoding']
+    else:
+        data = sys.stdin.read()
+
+    data = data.decode(encoding)
+    h = HTML2Text(baseurl=baseurl)
+    # handle options
+    if options.ul_style_dash: h.ul_item_mark = '-'
+    if options.em_style_asterisk:
+        h.emphasis_mark = '*'
+        h.strong_mark = '__'
+
+    h.body_width = options.body_width
+    h.list_indent = options.list_indent
+    h.ignore_emphasis = options.ignore_emphasis
+    h.ignore_links = options.ignore_links
+    h.ignore_images = options.ignore_images
+    h.google_doc = options.google_doc
+    h.hide_strikethrough = options.hide_strikethrough
+    h.escape_snob = options.escape_snob
+
+    wrapwrite(h.handle(data))
+
+
+if __name__ == "__main__":
+    main()