waylan/html_tidy.py

## html_tidy.py
# HTML Tidy Extension for Python-Markdown
# =======================================
#
# Runs [HTML Tidy][] on the output of Python-Markdown using the [uTidylib][]
# Python wrapper. Both libtidy and uTidylib must be installed on your system.
#
# [HTML Tidy]: http://tidy.sourceforge.net/
# [uTidylib]: http://utidylib.berlios.de/
#
# Note than any Tidy [options][] can be passed in as [extension configs][]. So,
# for example, to output HTML rather than XHTML, set ``output_xhtml=0``. To
# indent the output, set ``indent=auto`` and to have Tidy wrap the output in
# ``<html>`` and ``<body>`` tags, set ``show_body_only=0``. See Tidy's
# [options][] for a full list of the available options. The defaults are set to
# most closely match Markdowns defaults with the exception that you get much
# better pretty-printing.
#
# [options]: http://tidy.sourceforge.net/docs/quickref.html
# [extension configs]: ../reference.html#extension_configs
#
# Note that options set in this extension will override most any other settings
# passed on to Markdown (such as "output_format"). Unlike Markdown, this extension
# will also treat raw HTML no different than that output by Markdown. In other
# words, it may munge a document authors carefully crafted HTML. Of course, it
# may also transform poorly formed raw HTML into nice, valid HTML. Take these
# things into consideration when electing to use this extension.
#
# Copyright (c)2008 [Waylan Limberg](http://achinghead.com)
#
# License: [BSD](http://www.opensource.org/licenses/bsd-license.php)

from __future__ import absolute_import
from . import Extension
from ..postprocessors import Postprocessor
from ..util import text_type
try:
    import tidy
except ImportError:
    tidy = None

class TidyExtension(Extension):

    def __init__(self, configs):
        # Set defaults to match typical markdown behavior.
        self.config = dict(output_xhtml=1,
                           show_body_only=1,
                           char_encoding='utf8'
                          )
        # Merge in user defined configs overriding any present if nessecary.
        for c in configs:
            self.config[c[0]] = c[1]

    def extendMarkdown(self, md, md_globals):
        # Save options to markdown instance
        md.tidy_options = self.config
        # Add TidyProcessor to postprocessors
        if tidy:
            md.postprocessors['tidy'] = TidyProcessor(md)


class TidyProcessor(Postprocessor):

    def run(self, text):
        # Pass text to Tidy. As Tidy does not accept unicode we need to encode
        # it and decode its return value.
        enc = self.markdown.tidy_options.get('char_encoding', 'utf8')
        return text_type(tidy.parseString(text.encode(enc),
                                        **self.markdown.tidy_options),
                       encoding=enc)


def makeExtension(configs=None):
    return TidyExtension(configs=configs)
	# HTML Tidy Extension for Python-Markdown
	# =======================================
	#
	# Runs [HTML Tidy][] on the output of Python-Markdown using the [uTidylib][]
	# Python wrapper. Both libtidy and uTidylib must be installed on your system.
	#
	# [HTML Tidy]: http://tidy.sourceforge.net/
	# [uTidylib]: http://utidylib.berlios.de/
	#
	# Note than any Tidy [options][] can be passed in as [extension configs][]. So,
	# for example, to output HTML rather than XHTML, set ``output_xhtml=0``. To
	# indent the output, set ``indent=auto`` and to have Tidy wrap the output in
	# ``<html>`` and ``<body>`` tags, set ``show_body_only=0``. See Tidy's
	# [options][] for a full list of the available options. The defaults are set to
	# most closely match Markdowns defaults with the exception that you get much
	# better pretty-printing.
	#
	# [options]: http://tidy.sourceforge.net/docs/quickref.html
	# [extension configs]: ../reference.html#extension_configs
	#
	# Note that options set in this extension will override most any other settings
	# passed on to Markdown (such as "output_format"). Unlike Markdown, this extension
	# will also treat raw HTML no different than that output by Markdown. In other
	# words, it may munge a document authors carefully crafted HTML. Of course, it
	# may also transform poorly formed raw HTML into nice, valid HTML. Take these
	# things into consideration when electing to use this extension.
	#
	# Copyright (c)2008 [Waylan Limberg](http://achinghead.com)
	#
	# License: [BSD](http://www.opensource.org/licenses/bsd-license.php)

	from __future__ import absolute_import
	from . import Extension
	from ..postprocessors import Postprocessor
	from ..util import text_type
	try:
	import tidy
	except ImportError:
	tidy = None

	class TidyExtension(Extension):

	def __init__(self, configs):
	# Set defaults to match typical markdown behavior.
	self.config = dict(output_xhtml=1,
	show_body_only=1,
	char_encoding='utf8'
	)
	# Merge in user defined configs overriding any present if nessecary.
	for c in configs:
	self.config[c[0]] = c[1]

	def extendMarkdown(self, md, md_globals):
	# Save options to markdown instance
	md.tidy_options = self.config
	# Add TidyProcessor to postprocessors
	if tidy:
	md.postprocessors['tidy'] = TidyProcessor(md)


	class TidyProcessor(Postprocessor):

	def run(self, text):
	# Pass text to Tidy. As Tidy does not accept unicode we need to encode
	# it and decode its return value.
	enc = self.markdown.tidy_options.get('char_encoding', 'utf8')
	return text_type(tidy.parseString(text.encode(enc),
	**self.markdown.tidy_options),
	encoding=enc)


	def makeExtension(configs=None):
	return TidyExtension(configs=configs)