Better nested STRONG EM support.

Fixes #253. Thanks to @facelessuser for the tests. Although I removed
a bunch of weird ones (even some that passed) from his PR (#342). For
the most part, there is no definitive way for those to be parsed. So
there is no point of testing for them. In most of those situations,
authors should be mixing underscores and astericks so it is clear
what is intended.

Author: waylan

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["em_strong"] = DoubleTagPattern(EM_STRONG_RE, 'strong,em')
+    inlinePatterns["strong_em"] = DoubleTagPattern(STRONG_EM_RE, 'em,strong')
     inlinePatterns["strong"] = SimpleTagPattern(STRONG_RE, 'strong')
     inlinePatterns["emphasis"] = SimpleTagPattern(EMPHASIS_RE, 'em')
     if md_instance.smart_emphasis:
@@ -100,7 +101,8 @@ def build_inlinepatterns(md_instance, **kwargs):
 ESCAPE_RE = r'\\(.)'                             # \<
 EMPHASIS_RE = r'(\*)([^\*]+)\2'                    # *emphasis*
 STRONG_RE = r'(\*{2}|_{2})(.+?)\2'                      # **strong**
-STRONG_EM_RE = r'(\*{3}|_{3})(.+?)\2'            # ***strong***
+EM_STRONG_RE = r'(\*|_){3}(.+?)\2(.*?)\2{2}'            # ***strongem*** or ***em*strong**
+STRONG_EM_RE = r'(\*|_){3}(.+?)\2{2}(.*?)\2'            #  ***strong**em*
 SMART_EMPHASIS_RE = r'(?<!\w)(_)(?!_)(.+?)(?<!_)\2(?!\w)'  # _smart_emphasis_
 EMPHASIS_2_RE = r'(_)(.+?)\2'                 # _emphasis_
 LINK_RE = NOIMG + BRK + \
@@ -276,6 +278,8 @@ def handleMatch(self, m):
         el1 = util.etree.Element(tag1)
         el2 = util.etree.SubElement(el1, tag2)
         el2.text = m.group(3)
+        if len(m.groups())==5:
+            el2.tail = m.group(4)
         return el1
 
 

tests/misc/em_strong_complex.html

@@ -0,0 +1,14 @@
+<p><em><strong>test test</strong> test test</em></p>
+<p><strong><em>test test</em> test test</strong></p>
+<p><strong><em>test</em></strong></p>
+<p><strong>test</strong>_</p>
+<p><strong><em>test</em> test</strong>_</p>
+<p><strong><em>test</em> test</strong></p>
+<p><em>test_test test_test</em></p>
+<p><em><strong>test test</strong> test test</em></p>
+<p><strong><em>test test</em> test test</strong></p>
+<p>*<em>test</em></p>
+<p><strong><em>test</em></strong></p>
+<p><strong>test</strong>*</p>
+<p><strong><em>test</em> test</strong></p>
+<p><em>test</em>test test<em>test</em></p>

tests/misc/em_strong_complex.txt

@@ -0,0 +1,27 @@
+___test test__ test test_
+
+___test test_ test test__
+
+___test___
+
+__test___
+
+___test_ test___
+
+___test_ test__
+
+_test_test test_test_
+
+***test test** test test*
+
+***test test* test test**
+
+**test*
+
+***test***
+
+**test***
+
+***test* test**
+
+*test*test test*test*