Merge pull request #439 from aidantwoods/patch-4

Improve CommonMark mixed-marker list compliance

Parsedown.php

@@ -524,13 +524,31 @@ class Parsedown
 
     protected function blockList($Line)
     {
-        list($name, $pattern) = $Line['text'][0] <= '-' ? array('ul', '[*+-]') : array('ol', '[0-9]+[.]');
+        list($name, $pattern) = $Line['text'][0] <= '-' ? array('ul', '[*+-]') : array('ol', '[0-9]{1,9}[.\)]');
 
-        if (preg_match('/^('.$pattern.'[ ]+)(.*)/', $Line['text'], $matches))
+        if (preg_match('/^('.$pattern.'([ ]+|$))(.*)/', $Line['text'], $matches))
         {
+            $contentIndent = strlen($matches[2]);
+
+            if ($contentIndent >= 5)
+            {
+                $contentIndent -= 1;
+                $matches[1] = substr($matches[1], 0, -$contentIndent);
+                $matches[3] = str_repeat(' ', $contentIndent) . $matches[3];
+            }
+            elseif ($contentIndent === 0)
+            {
+                $matches[1] .= ' ';
+            }
+
             $Block = array(
                 'indent' => $Line['indent'],
                 'pattern' => $pattern,
+                'data' => array(
+                    'type' => $name,
+                    'marker' => $matches[1],
+                    'markerType' => ($name === 'ul' ? strstr($matches[1], ' ', true) : substr(strstr($matches[1], ' ', true), -1)),
+                ),
                 'element' => array(
                     'name' => $name,
                     'handler' => 'elements',
@@ -539,7 +557,7 @@ class Parsedown
 
             if($name === 'ol')
             {
-                $listStart = stristr($matches[0], '.', true);
+                $listStart = ltrim(strstr($matches[1], $Block['data']['markerType'], true), '0') ?: '0';
 
                 if($listStart !== '1')
                 {
@@ -550,9 +568,7 @@ class Parsedown
             $Block['li'] = array(
                 'name' => 'li',
                 'handler' => 'li',
-                'text' => array(
-                    $matches[2],
-                ),
+                'text' => !empty($matches[3]) ? array($matches[3]) : array(),
             );
 
             $Block['element']['text'] []= & $Block['li'];
@@ -563,7 +579,28 @@ class Parsedown
 
     protected function blockListContinue($Line, array $Block)
     {
-        if ($Block['indent'] === $Line['indent'] and preg_match('/^'.$Block['pattern'].'(?:[ ]+(.*)|$)/', $Line['text'], $matches))
+        if (isset($Block['interrupted']) and empty($Block['li']['text']))
+        {
+            return null;
+        }
+
+        $requiredIndent = ($Block['indent'] + strlen($Block['data']['marker']));
+
+        if (
+            $Line['indent'] < $requiredIndent
+            and
+            (
+                (
+                    $Block['data']['type'] === 'ol'
+                    and preg_match('/^[0-9]+'.preg_quote($Block['data']['markerType']).'(?:[ ]+(.*)|$)/', $Line['text'], $matches)
+                )
+                or
+                (
+                    $Block['data']['type'] === 'ul'
+                    and preg_match('/^'.preg_quote($Block['data']['markerType']).'(?:[ ]+(.*)|$)/', $Line['text'], $matches)
+                )
+            )
+        )
         {
             if (isset($Block['interrupted']))
             {
@@ -578,6 +615,8 @@ class Parsedown
 
             $text = isset($matches[1]) ? $matches[1] : '';
 
+            $Block['indent'] = $Line['indent'];
+
             $Block['li'] = array(
                 'name' => 'li',
                 'handler' => 'li',
@@ -590,31 +629,38 @@ class Parsedown
 
             return $Block;
         }
+        elseif ($Line['indent'] < $requiredIndent and $this->blockList($Line))
+        {
+            return null;
+        }
 
         if ($Line['text'][0] === '[' and $this->blockReference($Line))
         {
             return $Block;
         }
 
-        if ( ! isset($Block['interrupted']))
+        if ($Line['indent'] >= $requiredIndent)
         {
-            $text = preg_replace('/^[ ]{0,4}/', '', $Line['body']);
+            if (isset($Block['interrupted']))
+            {
+                $Block['li']['text'] []= '';
+
+                unset($Block['interrupted']);
+            }
+
+            $text = substr($Line['body'], $requiredIndent);
 
             $Block['li']['text'] []= $text;
 
             return $Block;
         }
 
-        if ($Line['indent'] > 0)
+        if ( ! isset($Block['interrupted']))
         {
-            $Block['li']['text'] []= '';
-
-            $text = preg_replace('/^[ ]{0,4}/', '', $Line['body']);
+            $text = preg_replace('/^[ ]{0,'.$requiredIndent.'}/', '', $Line['body']);
 
             $Block['li']['text'] []= $text;
 
-            unset($Block['interrupted']);
-
             return $Block;
         }
     }

test/data/deeply_nested_list.html

@@ -10,3 +10,31 @@
 </ul></li>
 <li>li</li>
 </ul>
+<hr />
+<ul>
+<li>level 1
+<ul>
+<li>level 2
+<ul>
+<li>level 3
+<ul>
+<li>level 4
+<ul>
+<li>level 5</li>
+</ul></li>
+</ul></li>
+</ul></li>
+</ul></li>
+</ul>
+<hr />
+<ul>
+<li>a</li>
+<li>b</li>
+<li>c</li>
+<li>d</li>
+<li>e</li>
+<li>f</li>
+<li>g</li>
+<li>h</li>
+<li>i</li>
+</ul>

test/data/deeply_nested_list.md

@@ -4,3 +4,23 @@
         - li
     - li
 - li
+
+---
+
+- level 1
+  - level 2
+    - level 3
+      - level 4
+        - level 5
+
+---
+
+- a
+ - b
+  - c
+   - d
+    - e
+   - f
+  - g
+ - h
+- i

test/data/unordered_list.html

@@ -2,9 +2,21 @@
 <li>li</li>
 <li>li</li>
 </ul>
-<p>mixed markers:</p>
+<p>mixed unordered markers:</p>
 <ul>
 <li>li</li>
-<li>li</li>
+</ul>
+<ul>
 <li>li</li>
 </ul>
+<ul>
+<li>li</li>
+</ul>
+<p>mixed ordered markers:</p>
+<ol>
+<li>starting at 1, list one</li>
+<li>number 2, list one</li>
+</ol>
+<ol start="3">
+<li>starting at 3, list two</li>
+</ol>

test/data/unordered_list.md

@@ -1,8 +1,14 @@
 - li
 - li
 
-mixed markers:
+mixed unordered markers:
 
 * li
 + li
 - li
+
+mixed ordered markers:
+
+1. starting at 1, list one
+2. number 2, list one
+3) starting at 3, list two