Render restructured text as html in pygtk webkit view

app.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

""" Render restructured text as html in a pygtk webkit view.
"""

import gobject
import gtk
import gtksourceview2 as gtksourceview
import io
import os.path
import pango
import webkit
import sys

import publishrst

class GtkApplication(object):
    """ Application class.
    """
    def __init__(self, rst_file):
        #build main gtk window
        self.rst_file = rst_file
        self.win = gtk.Window()
        self.win.set_default_size(800, 600)
        self.win.connect("destroy", lambda w: gtk.main_quit())
        self.lm = gtksourceview.language_manager_get_default()
        self.lm.set_search_path(self.lm.get_search_path() + [os.getcwd()])
        self.lang = self.lm.get_language('restructuredtext')

        #build notebook
        notebook = gtk.Notebook()

        #build webkit widget
        tab1 = gtk.ScrolledWindow()
        browser = webkit.WebView()
        with io.open(self.rst_file, 'rb') as fp:
            data = publishrst.html_parts(unicode(fp.read()))['whole']
            browser.load_html_string(data, 'file:///')

        #build source view
        tab2 = gtk.ScrolledWindow()
        buf = gtksourceview.Buffer()
        buf.set_language(self.lang)
        buf.set_highlight_syntax(True)
        with io.open(self.rst_file, 'rb') as fp:
            buf.set_text(fp.read())
            buf.set_data('filename', self.rst_file)
        buf.set_modified(True)
        buf.place_cursor(buf.get_start_iter())
        sview = gtksourceview.View(buf)
        sview.set_editable(False)
        font_desc = pango.FontDescription('monospace')
        if font_desc:
            sview.modify_font(font_desc)

        #add widgets and show window
        tab1.add_with_viewport(browser)
        tab2.add_with_viewport(sview)
        notebook.append_page(tab1, gtk.Label("Html view"))
        notebook.append_page(tab2, gtk.Label("Source view"))
        self.win.add(notebook)
        self.win.show_all()

    def open_file(self, buf, filename):
        return buf

    def run(self):
        gtk.main()

if __name__ == '__main__':
    if len(sys.argv) > 1:
        gobject.threads_init()
        app = GtkApplication(sys.argv[1])
        app.run()
    else:
        print("usage: %s <rst file>" % os.path.basename(sys.argv[0]))

example.rst

big title

Summary

• Section 1
  • Section 1.1
    • Section 1.1.1
• Section 2
  • Section 2.1
    • Section 2.1.1

Section 1

Section 1.1

Section 1.1.1

Section 2

Section 2.1

Section 2.1.1

publishrst.py

# Author: David Goodger <goodger@python.org>
# Copyright: This module has been placed in the public domain.

"""
This module contains practical examples of Docutils client code.

Importing this module from client code is not recommended; its contents are
subject to change in future Docutils releases.  Instead, it is recommended
that you copy and paste the parts you need into your own code, modifying as
necessary.
"""

from docutils import core, io


def html_parts(input_string, source_path=None, destination_path=None,
               input_encoding='unicode', doctitle=True,
               initial_header_level=1):
    """
    Given an input string, returns a dictionary of HTML document parts.

    Dictionary keys are the names of parts, and values are Unicode strings;
    encoding is up to the client.

    Parameters:

    - `input_string`: A multi-line text string; required.
    - `source_path`: Path to the source file or object.  Optional, but useful
      for diagnostic output (system messages).
    - `destination_path`: Path to the file or object which will receive the
      output; optional.  Used for determining relative paths (stylesheets,
      source links, etc.).
    - `input_encoding`: The encoding of `input_string`.  If it is an encoded
      8-bit string, provide the correct encoding.  If it is a Unicode string,
      use "unicode", the default.
    - `doctitle`: Disable the promotion of a lone top-level section title to
      document title (and subsequent section title to document subtitle
      promotion); enabled by default.
    - `initial_header_level`: The initial level for header elements (e.g. 1
      for "<h1>").
    """
    overrides = {'input_encoding': input_encoding,
                 'doctitle_xform': doctitle,
                 'initial_header_level': initial_header_level}
    parts = core.publish_parts(
        source=input_string, source_path=source_path,
        destination_path=destination_path,
        writer_name='html', settings_overrides=overrides)
    return parts

def html_body(input_string, source_path=None, destination_path=None,
              input_encoding='unicode', output_encoding='unicode',
              doctitle=True, initial_header_level=1):
    """
    Given an input string, returns an HTML fragment as a string.

    The return value is the contents of the <body> element.

    Parameters (see `html_parts()` for the remainder):

    - `output_encoding`: The desired encoding of the output.  If a Unicode
      string is desired, use the default value of "unicode" .
    """
    parts = html_parts(
        input_string=input_string, source_path=source_path,
        destination_path=destination_path,
        input_encoding=input_encoding, doctitle=doctitle,
        initial_header_level=initial_header_level)
    fragment = parts['html_body']
    if output_encoding != 'unicode':
        fragment = fragment.encode(output_encoding)
    return fragment

def internals(input_string, source_path=None, destination_path=None,
              input_encoding='unicode', settings_overrides=None):
    """
    Return the document tree and publisher, for exploring Docutils internals.

    Parameters: see `html_parts()`.
    """
    if settings_overrides:
        overrides = settings_overrides.copy()
    else:
        overrides = {}
    overrides['input_encoding'] = input_encoding
    output, pub = core.publish_programmatically(
        source_class=io.StringInput, source=input_string,
        source_path=source_path,
        destination_class=io.NullOutput, destination=None,
        destination_path=destination_path,
        reader=None, reader_name='standalone',
        parser=None, parser_name='restructuredtext',
        writer=None, writer_name='null',
        settings=None, settings_spec=None, settings_overrides=overrides,
        config_section=None, enable_exit_status=None)
    return pub.writer.document, pub

rst.lang

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE language SYSTEM "language.dtd">
<language _name="reStructuredText" version="1.0" _section="Markup" mimetypes="text/restructured">

	<pattern-item _name="Emphasis" style="Preprocessor">
		<regex>(\*[^\*\n]*[^\*]\*)</regex>
	</pattern-item>
	
	<pattern-item _name="Strong Emphasis" style="Data Type">
		<regex>(\*\*[^\*\n]*\*\*)</regex>
	</pattern-item>

	<pattern-item _name="Link" style="String">
		<regex>`[^`]*`_</regex>
	</pattern-item>
	
	<pattern-item _name="WikiLink" style="Link">
		<regex>\[\[[^\n]+\]\]</regex>
	</pattern-item>
	
	<pattern-item _name="List item" style="Keyword">
		<regex>(^ *\*[ \t]+)|(^ *[0-9]+\.[ \t]+)</regex>
	</pattern-item>
	
	<pattern-item _name="Line" style="Data Type">
    <regex>^[ \t]*[-='`"~`]+[ \t]*$</regex>
	</pattern-item>
	
	<pattern-item _name="Automatic link" style="String" end-at-line-end="TRUE">
		<regex>(&lt;[A-Za-z._0-9]+@[A-Za-z0-9]+.[A-Za-z]+&gt;)|(&lt;[a-z]+://[^ \n]+&gt;)</regex>
	</pattern-item>

	<block-comment _name = "Code block" style= "Comment">
    <start-regex>^::</start-regex>
	    <end-regex>^$</end-regex>
	</block-comment>
	
	<pattern-item _name="Inline Code" style="Comment">
		<regex>``[^``]*``</regex>
	</pattern-item>

	<string _name="Directive" style="Comment">
	    <start-regex>^\.\. </start-regex>
	    <end-regex>$</end-regex>
	</string>


</language>