From 85ac70859d5c995058d4509179bc484b6cd9904c Mon Sep 17 00:00:00 2001
From: Trent Mick <trentm@gmail.com>
Date: Wed, 16 Dec 2009 07:21:33 +0000
Subject: [PATCH] Implement issue 33: "cuddled-lists" extra. Still have doc and
 version updates to come.

---
 lib/markdown2.py                         | 40 ++++++++++++++++++------
 test/tm-cases/cuddled_para_and_list.html | 25 +++++++++++++++
 test/tm-cases/cuddled_para_and_list.tags |  2 +-
 test/tm-cases/cuddled_para_and_list.text | 20 ++++++++++++
 4 files changed, 76 insertions(+), 11 deletions(-)

diff --git a/lib/markdown2.py b/lib/markdown2.py
index 46778060..1ad6736d 100755
--- a/lib/markdown2.py
+++ b/lib/markdown2.py
@@ -1129,12 +1129,12 @@ def _do_lists(self, text):
         return text
     
     _list_item_re = re.compile(r'''
-        (\n)?               # leading line = \1
-        (^[ \t]*)           # leading whitespace = \2
-        (%s) [ \t]+         # list marker = \3
-        ((?:.+?)            # list item text = \4
-         (\n{1,2}))         # eols = \5
-        (?= \n* (\Z | \2 (%s) [ \t]+))
+        (\n)?                   # leading line = \1
+        (^[ \t]*)               # leading whitespace = \2
+        (?P<marker>%s) [ \t]+   # list marker = \3
+        ((?:.+?)                # list item text = \4
+         (\n{1,2}))             # eols = \5
+        (?= \n* (\Z | \2 (?P<next_marker>%s) [ \t]+))
         ''' % (_marker_any, _marker_any),
         re.M | re.X | re.S)
 
@@ -1380,15 +1380,35 @@ def _form_paragraphs(self, text):
         text = text.strip('\n')
 
         # Wrap <p> tags.
-        grafs = re.split(r"\n{2,}", text)
-        for i, graf in enumerate(grafs):
+        grafs = []
+        for i, graf in enumerate(re.split(r"\n{2,}", text)):
             if graf in self.html_blocks:
                 # Unhashify HTML blocks
-                grafs[i] = self.html_blocks[graf]
+                grafs.append(self.html_blocks[graf])
             else:
+                cuddled_list = None
+                if "cuddled-lists" in self.extras:
+                    # Need to put back trailing '\n' for `_list_item_re`
+                    # match at the end of the paragraph.
+                    li = self._list_item_re.search(graf + '\n')
+                    # Two of the same list marker in this paragraph: a likely
+                    # candidate for a list cuddled to preceding paragraph
+                    # text (issue 33). Note the `[-1]` is a quick way to
+                    # consider numeric bullets (e.g. "1." and "2.") to be
+                    # equal.
+                    if (li and li.group("next_marker")
+                        and li.group("marker")[-1] == li.group("next_marker")[-1]):
+                        start = li.start()
+                        cuddled_list = self._do_lists(graf[start:]).rstrip("\n")
+                        assert cuddled_list.startswith("<ul>") or cuddled_list.startswith("<ol>")
+                        graf = graf[:start]
+                    
                 # Wrap <p> tags.
                 graf = self._run_span_gamut(graf)
-                grafs[i] = "<p>" + graf.lstrip(" \t") + "</p>"
+                grafs.append("<p>" + graf.lstrip(" \t") + "</p>")
+                
+                if cuddled_list:
+                    grafs.append(cuddled_list)
 
         return "\n\n".join(grafs)
 
diff --git a/test/tm-cases/cuddled_para_and_list.html b/test/tm-cases/cuddled_para_and_list.html
index da6947fa..2402667c 100644
--- a/test/tm-cases/cuddled_para_and_list.html
+++ b/test/tm-cases/cuddled_para_and_list.html
@@ -5,3 +5,28 @@
 <li>bullet2</li>
 <li>bullet3</li>
 </ul>
+
+<p>I did these too:</p>
+
+<ul>
+<li>bullet1</li>
+<li>bullet2</li>
+<li>bullet3 this is
+a longer one.</li>
+</ul>
+
+<p>And these in order:</p>
+
+<ol>
+<li>bullet1</li>
+<li>bullet2</li>
+<li>bullet3</li>
+</ol>
+
+<p>Here is an asterisk,
+* What do you think
+about it?</p>
+
+<p>I suggest using version
+8. Oops, now this line is treated
+as a sub-list.</p>
diff --git a/test/tm-cases/cuddled_para_and_list.tags b/test/tm-cases/cuddled_para_and_list.tags
index e6433144..d66f4f5f 100644
--- a/test/tm-cases/cuddled_para_and_list.tags
+++ b/test/tm-cases/cuddled_para_and_list.tags
@@ -1 +1 @@
-smedberg issue33 knownfailure
+smedberg issue33
diff --git a/test/tm-cases/cuddled_para_and_list.text b/test/tm-cases/cuddled_para_and_list.text
index 0dffc389..198d2b7b 100644
--- a/test/tm-cases/cuddled_para_and_list.text
+++ b/test/tm-cases/cuddled_para_and_list.text
@@ -2,3 +2,23 @@ I did these things:
 * bullet1
 * bullet2
 * bullet3
+
+I did these too:
+* bullet1
+* bullet2
+* bullet3 this is
+a longer one.
+
+And these in order:
+1. bullet1
+2. bullet2
+3. bullet3
+
+Here is an asterisk,
+* What do you think
+about it?
+
+I suggest using version
+8. Oops, now this line is treated
+as a sub-list.
+