Cyclic0007/python-markdownify
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add support for headings that include nested block elements

Browse Source
This commit is contained in:
Igor Dvorkin
2020-11-15 09:04:22 -08:00
parent 25d68b4265
commit d558617cd7
2 changed files with 52 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
markdownify/__init__.py
View File

@@ -61,22 +61,28 @@ class MarkdownConverter(object):
def convert(self, html):
soup = BeautifulSoup(html, 'html.parser')
return self.process_tag(soup, children_only=True)
return self.process_tag(soup, convert_as_inline=False, children_only=True)
def process_tag(self, node, children_only=False):
def process_tag(self, node, convert_as_inline, children_only=False):
text = ''
# markdown headings can't include block elements (elements w/newlines)
isHeading = node.name.startswith('h')
convert_children_as_inline = convert_as_inline
if not children_only and isHeading:
convert_children_as_inline = True
# Convert the children first
for el in node.children:
if isinstance(el, NavigableString):
text += self.process_text(six.text_type(el))
else:
text += self.process_tag(el)
text += self.process_tag(el, convert_children_as_inline)
if not children_only:
convert_fn = getattr(self, 'convert_%s' % node.name, None)
if convert_fn and self.should_convert_tag(node.name):
text = convert_fn(node, text)
text = convert_fn(node, text, convert_as_inline)
return text
@@ -89,8 +95,8 @@ class MarkdownConverter(object):
if m:
n = int(m.group(1))
def convert_tag(el, text):
return self.convert_hn(n, el, text)
def convert_tag(el, text, convert_as_inline):
return self.convert_hn(n, el, text, convert_as_inline)
convert_tag.__name__ = 'convert_h%s' % n
setattr(self, convert_tag.__name__, convert_tag)
@@ -116,10 +122,12 @@ class MarkdownConverter(object):
text = (text or '').rstrip()
return '%s\n%s\n\n' % (text, pad_char * len(text)) if text else ''
def convert_a(self, el, text):
def convert_a(self, el, text, convert_as_inline):
prefix, suffix, text = chomp(text)
if not text:
return ''
if convert_as_inline:
return text
href = el.get('href')
title = el.get('title')
if self.options['autolinks'] and text == href and not title:
@@ -128,22 +136,32 @@ class MarkdownConverter(object):
title_part = ' "%s"' % title.replace('"', r'\"') if title else ''
return '%s[%s](%s%s)%s' % (prefix, text, href, title_part, suffix) if href else text
def convert_b(self, el, text):
return self.convert_strong(el, text)
def convert_b(self, el, text, convert_as_inline):
return self.convert_strong(el, text, convert_as_inline)
def convert_blockquote(self, el, text, convert_as_inline):
if convert_as_inline:
return text
def convert_blockquote(self, el, text):
return '\n' + line_beginning_re.sub('> ', text) if text else ''
def convert_br(self, el, text):
def convert_br(self, el, text, convert_as_inline):
if convert_as_inline:
return ""
return ' \n'
def convert_em(self, el, text):
def convert_em(self, el, text, convert_as_inline):
prefix, suffix, text = chomp(text)
if not text:
return ''
return '%s*%s*%s' % (prefix, text, suffix)
def convert_hn(self, n, el, text):
def convert_hn(self, n, el, text, convert_as_inline):
if convert_as_inline:
return text
style = self.options['heading_style']
text = text.rstrip()
if style == UNDERLINED and n <= 2:
@@ -154,10 +172,14 @@ class MarkdownConverter(object):
return '%s %s %s\n\n' % (hashes, text, hashes)
return '%s %s\n\n' % (hashes, text)
def convert_i(self, el, text):
return self.convert_em(el, text)
def convert_i(self, el, text, convert_as_inline):
return self.convert_em(el, text, convert_as_inline)
def convert_list(self, el, text, convert_as_inline):
# Converting a list to inline is undefined.
# Ignoring convert_to_inline for list.
def convert_list(self, el, text):
nested = False
while el:
if el.name == 'li':
@@ -172,7 +194,7 @@ class MarkdownConverter(object):
convert_ul = convert_list
convert_ol = convert_list
def convert_li(self, el, text):
def convert_li(self, el, text, convert_as_inline):
parent = el.parent
if parent is not None and parent.name == 'ol':
if parent.get("start"):
@@ -190,16 +212,18 @@ class MarkdownConverter(object):
bullet = bullets[depth % len(bullets)]
return '%s %s\n' % (bullet, text or '')
def convert_p(self, el, text):
def convert_p(self, el, text, convert_as_inline):
if convert_as_inline:
return text
return '%s\n\n' % text if text else ''
def convert_strong(self, el, text):
def convert_strong(self, el, text, convert_as_inline):
prefix, suffix, text = chomp(text)
if not text:
return ''
return '%s**%s**%s' % (prefix, text, suffix)
def convert_img(self, el, text):
def convert_img(self, el, text, convert_as_inline):
alt = el.attrs.get('alt', None) or ''
src = el.attrs.get('src', None) or ''
title = el.attrs.get('title', None) or ''

8
tests/test_conversions.py
View File

@@ -107,6 +107,14 @@ def test_hn():
assert md('<h6>Hello</h6>') == '###### Hello\n\n'
def test_hn_nested_tag():
assert md('<h3>A <b>Bold</b> C </h3>') == '### A **Bold** C\n\n'
assert md('<h3>A <p>P</p> C </h3>') == '### A P C\n\n'
assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX_CLOSED) == '# A P C #\n\n'
assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX) == '# A P C\n\n'
assert md('<h3>A <blockquote>BQ</blockquote> C </h3>') == '### A BQ C\n\n'
def test_atx_headings():
assert md('<h1>Hello</h1>', heading_style=ATX) == '# Hello\n\n'
assert md('<h2>Hello</h2>', heading_style=ATX) == '## Hello\n\n'