def html_to_rst(html, indent=0, indent_first=False):
"""
Use bcdoc to convert html to rst.
:type html: string
:param html: Input HTML to be converted
:type indent: int
:param indent: Number of spaces to indent each line
:type indent_first: boolean
:param indent_first: Whether to indent the first line
:rtype: string
"""
doc = ReSTDocument()
# TODO: Remove me, temp workaround to fix doc building
# because of smart quotes that aren't currently supported.
html = html.replace(u'\u2019', "'")
html = html.replace(u'\u2014', '-')
doc.include_doc_string(html)
rst = doc.getvalue().decode('utf-8')
if indent:
rst = '\n'.join([(' ' * indent) + line for line in rst.splitlines()])
if not indent_first:
rst = rst.strip()
return rst
def test_remove_doc_string(self):
doc = ReSTDocument()
doc.writeln('foo')
doc.include_doc_string('<p>this is a <code>test</code></p>')
doc.remove_last_doc_string()
self.assertEqual(doc.getvalue(), six.b('foo\n'))
def test_include_doc_string(self):
doc = ReSTDocument()
doc.include_doc_string('<p>this is a <code>test</code></p>')
self.assertEqual(doc.getvalue(), six.b('\n\nthis is a ``test`` \n\n'))
