#!/usr/local/bin/python2.7 “”“ rest2html - A small wrapper file for parsing ReST files at GitHub.
Written in 2008 by Jannis Leidel <jannis@leidel.info>
Brandon Keepers <bkeepers@github.com> Bryan Veloso <bryan@revyver.com> Chris Wanstrath <chris@ozmm.org> Dave Abrahams <dave@boostpro.com> Garen Torikian <garen@github.com> Gasper Zejn <zejn@kiberpipa.org> Michael Jones <m.pricejones@gmail.com> Sam Whited <sam@samwhited.com> Tyler Chung <zonyitoo@gmail.com> Vicent Marti <tanoku@gmail.com>
To the extent possible under law, the author(s) have dedicated all copyright and related and neighboring rights to this software to the public domain worldwide. This software is distributed without any warranty.
You should have received a copy of the CC0 Public Domain Dedication along with this software. If not, see <creativecommons.org/publicdomain/zero/1.0/>. “”“
__author__ = “Jannis Leidel” __license__ = “CC0” __version__ = “0.2”
import sys import os
# This fixes docutils failing with unicode parameters to CSV-Table. The -S # switch and the following 3 lines can be removed after upgrading to python 3. if sys.version_info < 3:
reload(sys) sys.setdefaultencoding('utf-8')
import site
try:
import locale locale.setlocale(locale.LC_ALL, '')
except:
pass
import codecs import io
from docutils import nodes, statemachine from docutils.parsers.rst import directives, roles from docutils.parsers.rst.directives.body import CodeBlock from docutils.core import publish_parts from docutils.writers.html4css1 import Writer, HTMLTranslator
SETTINGS = {
'cloak_email_addresses': False, 'file_insertion_enabled': False, 'raw_enabled': False, 'strip_comments': True, 'doctitle_xform': True, 'sectsubtitle_xform': True, 'initial_header_level': 2, 'report_level': 5, 'syntax_highlight': 'none', 'math_output': 'latex', 'field_name_limit': 50,
}
class DoctestDirective(CodeBlock):
"""Render Sphinx 'doctest:: [group]' blocks as 'code:: python' """ def run(self): """Discard any doctest group argument, render contents as python code """ self.arguments = ['python'] return super(DoctestDirective, self).run()
class PlantumlDirective(CodeBlock):
""" Render 'plantuml::' blocks as 'code:: plantuml', so they will be picked up in GitLab by the generic Banzai::Filter::PlantumlFilter """ option_spec = CodeBlock.option_spec.copy() option_spec.update({ 'caption': directives.unchanged }) def run(self): self.arguments = ['plantuml'] node = super(PlantumlDirective, self).run() if 'caption' in self.options: # create an anonymous container to parse the caption content container = nodes.Element() content = statemachine.StringList([self.options['caption']]) self.state.nested_parse(content, self.content_offset, container) caption = nodes.caption(self.options['caption'], '', *container) node.append(caption) return node
class GitHubHTMLTranslator(HTMLTranslator, object):
# removes the <div class="document"> tag wrapped around docs # see also: http://bit.ly/1exfq2h (warning! sourceforge link.) def depart_document(self, node): HTMLTranslator.depart_document(self, node) self.html_body.pop(0) # pop the starting <div> off self.html_body.pop() # pop the ending </div> off # technique for visiting sections, without generating additional divs # see also: http://bit.ly/NHtyRx # the a is to support ::contents with ::sectnums: http://git.io/N1yC def visit_section(self, node): id_attribute = node.attributes['ids'][0] self.body.append('<a name="%s"></a>\n' % id_attribute) self.section_level += 1 def depart_section(self, node): self.section_level -= 1 def visit_literal_block(self, node): classes = node.attributes['classes'] if len(classes) >= 2 and classes[0] == 'code': language = classes[1] del classes[:] self.body.append(self.starttag(node, 'pre')) self.body.append(self.starttag(node, 'code', suffix='', lang=language)) else: self.body.append(self.starttag(node, 'pre')) self.body.append(self.starttag(node, 'code', suffix='')) def depart_literal_block(self, node): self.body.append('</code>') return super(GitHubHTMLTranslator, self).depart_literal_block(node) # always wrap two-backtick rst inline literals in <code>, not <tt> # this also avoids the generation of superfluous <span> tags def visit_literal(self, node): self.body.append(self.starttag(node, 'code', suffix='')) def depart_literal(self, node): self.body.append('</code>') def visit_table(self, node): classes = ' '.join(['docutils', self.settings.table_style]).strip() self.body.append( self.starttag(node, 'table', CLASS=classes)) def depart_table(self, node): self.body.append('</table>\n') def depart_image(self, node): uri = node['uri'] ext = os.path.splitext(uri)[1].lower() # we need to swap RST's use of `object` with `img` tags # see http://git.io/5me3dA if ext == ".svg": # preserve essential attributes atts = {} for attribute, value in node.attributes.items(): # we have no time for empty values if value: if attribute == "uri": atts['src'] = value else: atts[attribute] = value # toss off `object` tag self.body.pop() # add on `img` with attributes self.body.append(self.starttag(node, 'img', **atts)) HTMLTranslator.depart_image(self, node)
def kbd(name, rawtext, text, lineno, inliner, options=None, content=None):
return [nodes.raw('', '<kbd>%s</kbd>' % text, format='html')], []
def main():
""" Parses the given ReST file or the redirected string input and returns the HTML body. Usage: rest2html < README.rst rest2html README.rst """ try: text = codecs.open(sys.argv[1], 'r', 'utf-8').read() except IOError: # given filename could not be found return '' except IndexError: # no filename given if sys.version_info[0] < 3: # python 2.x text = sys.stdin.read() else: # python 3 input_stream = io.TextIOWrapper(sys.stdin.buffer, encoding='utf-8') text = input_stream.read() writer = Writer() writer.translator_class = GitHubHTMLTranslator roles.register_canonical_role('kbd', kbd) # Render source code in Sphinx doctest blocks directives.register_directive('doctest', DoctestDirective) # Render source code in Sphinx plantuml blocks # Also support uml:: directive for compatibility with sphinxcontrib-plantuml directives.register_directive('plantuml', PlantumlDirective) directives.register_directive('uml', PlantumlDirective) parts = publish_parts(text, writer=writer, settings_overrides=SETTINGS) if 'html_body' in parts: html = parts['html_body'] # publish_parts() in python 2.x return dict values as Unicode type # in py3k Unicode is unavailable and values are of str type if isinstance(html, str): return html else: return html.encode('utf-8') return ''
if __name__ == '__main__':
if sys.version_info[0] < 3: # python 2.x sys.stdout.write("%s%s" % (main(), "\n")) else: # python 3 output_stream = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8') output_stream.write("%s%s" % (main(), "\n")) sys.stdout.flush()