From 94962cb644b3117e7dd46f16b1a4341609ad6bf2 Mon Sep 17 00:00:00 2001
From: Isaac Muse <faceless.shop@gmail.com>
Date: Tue, 24 Jan 2017 08:36:37 -0700
Subject: Fix HTML parse with empty lines (#537)

If both open and close was not found in first block, additional blocks
were evaluated without context of previous blocks.  The algorithm needs
to evaluate a buffer with the left bracket present.  So feed in all
items and get the right bracket, then adjust the data_index to be
relative to the last block. Fixes #452.
---
 markdown/preprocessors.py |  8 +++++++-
 tests/misc/html.html      | 11 +++++++++++
 tests/misc/html.txt       | 11 +++++++++++
 3 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/markdown/preprocessors.py b/markdown/preprocessors.py
index 7ea4fcf..94f9830 100644
--- a/markdown/preprocessors.py
+++ b/markdown/preprocessors.py
@@ -258,7 +258,13 @@ class HtmlBlockPreprocessor(Preprocessor):
             else:
                 items.append(block)
 
-                right_tag, data_index = self._get_right_tag(left_tag, 0, block)
+                # Need to evaluate all items so we can calculate relative to the left index.
+                right_tag, data_index = self._get_right_tag(left_tag, left_index, ''.join(items))
+                # Adjust data_index: relative to items -> relative to last block
+                prev_block_length = 0
+                for item in items[:-1]:
+                    prev_block_length += len(item)
+                data_index -= prev_block_length
 
                 if self._equal_tags(left_tag, right_tag):
                     # if find closing tag
diff --git a/tests/misc/html.html b/tests/misc/html.html
index 1eb6a97..5380bbd 100644
--- a/tests/misc/html.html
+++ b/tests/misc/html.html
@@ -8,6 +8,17 @@
 Html with various attributes.
 </div>
 
+<div>
+    <div>
+        Div with a blank line
+
+        in the middle.
+    </div>
+    <div>
+        This gets treated as HTML.
+    </div>
+</div>
+
 <p>And of course <script>blah</script>.</p>
 <p><a href="script&gt;stuff&lt;/script">this <script>link</a></p>
 <p>Some funky <x\]> inline stuff with markdown escaping syntax.</p>
diff --git a/tests/misc/html.txt b/tests/misc/html.txt
index dfee68d..8f18fa7 100644
--- a/tests/misc/html.txt
+++ b/tests/misc/html.txt
@@ -11,6 +11,17 @@ Now some <arbitrary>arbitrary tags</arbitrary>.
 Html with various attributes.
 </div>
 
+<div>
+    <div>
+        Div with a blank line
+
+        in the middle.
+    </div>
+    <div>
+        This gets treated as HTML.
+    </div>
+</div>
+
 And of course <script>blah</script>.
 
 [this <script>link](<script>stuff</script>)
-- 
cgit v1.2.3