Partial fix for #253.

This actually fixes the immediate problem. The method on the inlinePattern returns the correct subtree. However, for some reason, the parent element of that subtree is later stripped of all content (perhaps by the inlineplaceholder stuff). So, for example, given this input: ***foo** bar* The inlinepattern returns: <em><strong>foo</strong> bar</em> However, later it is reduced to: <em></em> Interestingly, this still works: ***foo*** Correctly becomes: <strong><em>foo</em></strong> I'm guessing that the tail on the `em` tag in the first instance is tripping things up some how.

Author: Waylan Limberg

markdown/inlinepatterns.py

@@ -75,7 +75,8 @@ def build_inlinepatterns(md_instance, **kwargs):
  inlinePatterns["html"] = HtmlPattern(HTML_RE, md_instance)
  inlinePatterns["entity"] = HtmlPattern(ENTITY_RE, md_instance)
  inlinePatterns["not_strong"] = SimpleTextPattern(NOT_STRONG_RE)
- inlinePatterns["strong_em"] = DoubleTagPattern(STRONG_EM_RE, 'strong,em')
+ inlinePatterns["strong_em"] = DoubleTagPattern(A_STRONG_EM_RE, 'strong,em')
+ inlinePatterns["strong_em_"] = DoubleTagPattern(U_STRONG_EM_RE, 'strong,em')
  inlinePatterns["strong"] = SimpleTagPattern(STRONG_RE, 'strong')
  inlinePatterns["emphasis"] = SimpleTagPattern(EMPHASIS_RE, 'em')
  if md_instance.smart_emphasis:
@@ -99,8 +100,9 @@ def build_inlinepatterns(md_instance, **kwargs):
 BACKTICK_RE = r'(?<!\\)(`+)(.+?)(?<!`)\2(?!`)' # `e=f()` or ``e=f("`")``
 ESCAPE_RE = r'\\(.)' # \<
 EMPHASIS_RE = r'(\*)([^\*]+)\2' # *emphasis*
-STRONG_RE = r'(\*{2}|_{2})(.+?)\2' # **strong**
-STRONG_EM_RE = r'(\*{3}|_{3})(.+?)\2' # ***strong***
+STRONG_RE = r'(\*{2}|_{2})(.+?)\2' # **strong**
+A_STRONG_EM_RE = r'\*{3}(.+?)\*{2}([^*]*)\*' # ***strongem***
+U_STRONG_EM_RE = r'_{3}(.+?)_{2}([^_]*)_' # ___strongem___
 SMART_EMPHASIS_RE = r'(?<!\w)(_)(?!_)(.+?)(?<!_)\2(?!\w)' # _smart_emphasis_
 EMPHASIS_2_RE = r'(_)(.+?)\2' # _emphasis_
 LINK_RE = NOIMG + BRK + \
@@ -276,9 +278,17 @@ class DoubleTagPattern(SimpleTagPattern):
  """
  def handleMatch(self, m):
  tag1, tag2 = self.tag.split(",")
- el1 = util.etree.Element(tag1)
- el2 = util.etree.SubElement(el1, tag2)
- el2.text = m.group(3)
+ if m.group(3):
+ # This is a match for `***foo** bar*`. Group 3 contains ` bar`.
+ el1 = util.etree.Element(tag2)
+ el2 = util.etree.SubElement(el1, tag1)
+ el2.text = m.group(2)
+ el2.tail = m.group(3)
+ else:
+ # This is a match for `***foo***`. We don't switch tag order here.
+ el1 = util.etree.Element(tag1)
+ el2 = util.etree.SubElement(el1, tag2)
+ el2.text = m.group(2)
  return el1