From 01fe37dc45c661154e36a394406dfca58d139da0 Mon Sep 17 00:00:00 2001
From: chrispy <chrispy@synopsys.com>
Date: Sun, 26 Jan 2025 21:27:09 -0500
Subject: [PATCH 1/3] remove superfluous leading/trailing whitespace

Signed-off-by: chrispy <chrispy@synopsys.com>
---
 markdownify/__init__.py   |  12 ++-
 tests/test_advanced.py    |   2 +-
 tests/test_conversions.py | 181 +++++++++++++++++++-------------------
 tests/test_lists.py       |  28 +++---
 tests/test_tables.py      |  57 ++++++------
 5 files changed, 144 insertions(+), 136 deletions(-)

diff --git a/markdownify/__init__.py b/markdownify/__init__.py
index ef4e7ca..10ff153 100644
--- a/markdownify/__init__.py
+++ b/markdownify/__init__.py
@@ -180,7 +180,10 @@ def process_tag(self, node, convert_as_inline):
         return text
 
     def convert__document_(self, el, text, convert_as_inline):
-        # for BeautifulSoup objects (where node.name == "[document]"), return content results as-is
+        """Final document-level formatting for BeautifulSoup object (node.name == "[document]")"""
+        # remove all leading newlines
+        text = text.lstrip('\n')
+
         return text
 
     def process_text(self, el):
@@ -454,6 +457,7 @@ def _indent_for_li(match):
     def convert_p(self, el, text, convert_as_inline):
         if convert_as_inline:
             return ' ' + text.strip() + ' '
+        text = text.strip()
         if self.options['wrap']:
             # Preserve newlines (and preceding whitespace) resulting
             # from <br> tags.  Newlines in the input have already been
@@ -500,13 +504,13 @@ def convert_style(self, el, text, convert_as_inline):
     convert_sup = abstract_inline_conversion(lambda self: self.options['sup_symbol'])
 
     def convert_table(self, el, text, convert_as_inline):
-        return '\n\n' + text + '\n'
+        return '\n\n' + text.strip() + '\n\n'
 
     def convert_caption(self, el, text, convert_as_inline):
-        return text + '\n\n'
+        return text.strip() + '\n\n'
 
     def convert_figcaption(self, el, text, convert_as_inline):
-        return '\n\n' + text + '\n\n'
+        return '\n\n' + text.strip() + '\n\n'
 
     def convert_td(self, el, text, convert_as_inline):
         colspan = 1
diff --git a/tests/test_advanced.py b/tests/test_advanced.py
index a3a5fda..14bf3cd 100644
--- a/tests/test_advanced.py
+++ b/tests/test_advanced.py
@@ -14,7 +14,7 @@ def test_chomp():
 
 def test_nested():
     text = md('<p>This is an <a href="http://example.com/">example link</a>.</p>')
-    assert text == '\n\nThis is an [example link](http://example.com/).\n\n'
+    assert text == 'This is an [example link](http://example.com/).\n\n'
 
 
 def test_ignore_comments():
diff --git a/tests/test_conversions.py b/tests/test_conversions.py
index 05c6cd4..4df024e 100644
--- a/tests/test_conversions.py
+++ b/tests/test_conversions.py
@@ -41,7 +41,7 @@ def test_a_no_autolinks():
 
 def test_a_in_code():
     assert md('<code><a href="https://google.com">Google</a></code>') == '`Google`'
-    assert md('<pre><a href="https://google.com">Google</a></pre>') == '\n\n```\nGoogle\n```\n\n'
+    assert md('<pre><a href="https://google.com">Google</a></pre>') == '```\nGoogle\n```\n\n'
 
 
 def test_b():
@@ -56,22 +56,22 @@ def test_b_spaces():
 
 
 def test_blockquote():
-    assert md('<blockquote>Hello</blockquote>') == '\n> Hello\n\n'
-    assert md('<blockquote>\nHello\n</blockquote>') == '\n> Hello\n\n'
+    assert md('<blockquote>Hello</blockquote>') == '> Hello\n\n'
+    assert md('<blockquote>\nHello\n</blockquote>') == '> Hello\n\n'
 
 
 def test_blockquote_with_nested_paragraph():
-    assert md('<blockquote><p>Hello</p></blockquote>') == '\n> Hello\n\n'
-    assert md('<blockquote><p>Hello</p><p>Hello again</p></blockquote>') == '\n> Hello\n>\n> Hello again\n\n'
+    assert md('<blockquote><p>Hello</p></blockquote>') == '> Hello\n\n'
+    assert md('<blockquote><p>Hello</p><p>Hello again</p></blockquote>') == '> Hello\n>\n> Hello again\n\n'
 
 
 def test_blockquote_with_paragraph():
-    assert md('<blockquote>Hello</blockquote><p>handsome</p>') == '\n> Hello\n\nhandsome\n\n'
+    assert md('<blockquote>Hello</blockquote><p>handsome</p>') == '> Hello\n\nhandsome\n\n'
 
 
 def test_blockquote_nested():
     text = md('<blockquote>And she was like <blockquote>Hello</blockquote></blockquote>')
-    assert text == '\n> And she was like\n> > Hello\n\n'
+    assert text == '> And she was like\n> > Hello\n\n'
 
 
 def test_br():
@@ -79,11 +79,6 @@ def test_br():
     assert md('a<br />b<br />c', newline_style=BACKSLASH) == 'a\\\nb\\\nc'
 
 
-def test_caption():
-    assert md('TEXT<figure><figcaption>Caption</figcaption><span>SPAN</span></figure>') == 'TEXT\n\nCaption\n\nSPAN'
-    assert md('<figure><span>SPAN</span><figcaption>Caption</figcaption></figure>TEXT') == 'SPAN\n\nCaption\n\nTEXT'
-
-
 def test_code():
     inline_tests('code', '`')
     assert md('<code>*this_should_not_escape*</code>') == '`*this_should_not_escape*`'
@@ -105,13 +100,13 @@ def test_code():
 
 
 def test_dl():
-    assert md('<dl><dt>term</dt><dd>definition</dd></dl>') == '\nterm\n:   definition\n'
-    assert md('<dl><dt><p>te</p><p>rm</p></dt><dd>definition</dd></dl>') == '\nte rm\n:   definition\n'
-    assert md('<dl><dt>term</dt><dd><p>definition-p1</p><p>definition-p2</p></dd></dl>') == '\nterm\n:   definition-p1\n\n    definition-p2\n'
-    assert md('<dl><dt>term</dt><dd><p>definition 1</p></dd><dd><p>definition 2</p></dd></dl>') == '\nterm\n:   definition 1\n:   definition 2\n'
-    assert md('<dl><dt>term 1</dt><dd>definition 1</dd><dt>term 2</dt><dd>definition 2</dd></dl>') == '\nterm 1\n:   definition 1\nterm 2\n:   definition 2\n'
-    assert md('<dl><dt>term</dt><dd><blockquote><p>line 1</p><p>line 2</p></blockquote></dd></dl>') == '\nterm\n:   > line 1\n    >\n    > line 2\n'
-    assert md('<dl><dt>term</dt><dd><ol><li><p>1</p><ul><li>2a</li><li>2b</li></ul></li><li><p>3</p></li></ol></dd></dl>') == '\nterm\n:   1. 1\n\n       * 2a\n       * 2b\n    2. 3\n'
+    assert md('<dl><dt>term</dt><dd>definition</dd></dl>') == 'term\n:   definition\n'
+    assert md('<dl><dt><p>te</p><p>rm</p></dt><dd>definition</dd></dl>') == 'te rm\n:   definition\n'
+    assert md('<dl><dt>term</dt><dd><p>definition-p1</p><p>definition-p2</p></dd></dl>') == 'term\n:   definition-p1\n\n    definition-p2\n'
+    assert md('<dl><dt>term</dt><dd><p>definition 1</p></dd><dd><p>definition 2</p></dd></dl>') == 'term\n:   definition 1\n:   definition 2\n'
+    assert md('<dl><dt>term 1</dt><dd>definition 1</dd><dt>term 2</dt><dd>definition 2</dd></dl>') == 'term 1\n:   definition 1\nterm 2\n:   definition 2\n'
+    assert md('<dl><dt>term</dt><dd><blockquote><p>line 1</p><p>line 2</p></blockquote></dd></dl>') == 'term\n:   > line 1\n    >\n    > line 2\n'
+    assert md('<dl><dt>term</dt><dd><ol><li><p>1</p><ul><li>2a</li><li>2b</li></ul></li><li><p>3</p></li></ol></dd></dl>') == 'term\n:   1. 1\n\n       * 2a\n       * 2b\n    2. 3\n'
 
 
 def test_del():
@@ -126,42 +121,47 @@ def test_em():
     inline_tests('em', '*')
 
 
+def test_figcaption():
+    assert (md("TEXT<figure><figcaption>\nCaption\n</figcaption><span>SPAN</span></figure>") == "TEXT\n\nCaption\n\nSPAN")
+    assert (md("<figure><span>SPAN</span><figcaption>\nCaption\n</figcaption></figure>TEXT") == "SPAN\n\nCaption\n\nTEXT")
+
+
 def test_header_with_space():
-    assert md('<h3>\n\nHello</h3>') == '\n\n### Hello\n\n'
-    assert md('<h3>Hello\n\n\nWorld</h3>') == '\n\n### Hello World\n\n'
-    assert md('<h4>\n\nHello</h4>') == '\n\n#### Hello\n\n'
-    assert md('<h5>\n\nHello</h5>') == '\n\n##### Hello\n\n'
-    assert md('<h5>\n\nHello\n\n</h5>') == '\n\n##### Hello\n\n'
-    assert md('<h5>\n\nHello   \n\n</h5>') == '\n\n##### Hello\n\n'
+    assert md('<h3>\n\nHello</h3>') == '### Hello\n\n'
+    assert md('<h3>Hello\n\n\nWorld</h3>') == '### Hello World\n\n'
+    assert md('<h4>\n\nHello</h4>') == '#### Hello\n\n'
+    assert md('<h5>\n\nHello</h5>') == '##### Hello\n\n'
+    assert md('<h5>\n\nHello\n\n</h5>') == '##### Hello\n\n'
+    assert md('<h5>\n\nHello   \n\n</h5>') == '##### Hello\n\n'
 
 
 def test_h1():
-    assert md('<h1>Hello</h1>') == '\n\nHello\n=====\n\n'
+    assert md('<h1>Hello</h1>') == 'Hello\n=====\n\n'
 
 
 def test_h2():
-    assert md('<h2>Hello</h2>') == '\n\nHello\n-----\n\n'
+    assert md('<h2>Hello</h2>') == 'Hello\n-----\n\n'
 
 
 def test_hn():
-    assert md('<h3>Hello</h3>') == '\n\n### Hello\n\n'
-    assert md('<h4>Hello</h4>') == '\n\n#### Hello\n\n'
-    assert md('<h5>Hello</h5>') == '\n\n##### Hello\n\n'
-    assert md('<h6>Hello</h6>') == '\n\n###### Hello\n\n'
+    assert md('<h3>Hello</h3>') == '### Hello\n\n'
+    assert md('<h4>Hello</h4>') == '#### Hello\n\n'
+    assert md('<h5>Hello</h5>') == '##### Hello\n\n'
+    assert md('<h6>Hello</h6>') == '###### Hello\n\n'
     assert md('<h10>Hello</h10>') == md('<h6>Hello</h6>')
     assert md('<hn>Hello</hn>') == md('Hello')
 
 
 def test_hn_chained():
-    assert md('<h1>First</h1>\n<h2>Second</h2>\n<h3>Third</h3>', heading_style=ATX) == '\n\n# First\n\n## Second\n\n### Third\n\n'
+    assert md('<h1>First</h1>\n<h2>Second</h2>\n<h3>Third</h3>', heading_style=ATX) == '# First\n\n## Second\n\n### Third\n\n'
     assert md('X<h1>First</h1>', heading_style=ATX) == 'X\n\n# First\n\n'
     assert md('X<h1>First</h1>', heading_style=ATX_CLOSED) == 'X\n\n# First #\n\n'
     assert md('X<h1>First</h1>') == 'X\n\nFirst\n=====\n\n'
 
 
 def test_hn_nested_tag_heading_style():
-    assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX_CLOSED) == '\n\n# A P C #\n\n'
-    assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX) == '\n\n# A P C\n\n'
+    assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX_CLOSED) == '# A P C #\n\n'
+    assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX) == '# A P C\n\n'
 
 
 def test_hn_nested_simple_tag():
@@ -177,9 +177,9 @@ def test_hn_nested_simple_tag():
     ]
 
     for tag, markdown in tag_to_markdown:
-        assert md('<h3>A <' + tag + '>' + tag + '</' + tag + '> B</h3>') == '\n\n### A ' + markdown + ' B\n\n'
+        assert md('<h3>A <' + tag + '>' + tag + '</' + tag + '> B</h3>') == '### A ' + markdown + ' B\n\n'
 
-    assert md('<h3>A <br>B</h3>', heading_style=ATX) == '\n\n### A B\n\n'
+    assert md('<h3>A <br>B</h3>', heading_style=ATX) == '### A B\n\n'
 
     # Nested lists not supported
     # assert md('<h3>A <ul><li>li1</i><li>l2</li></ul></h3>', heading_style=ATX) == '\n### A li1 li2 B\n\n'
@@ -192,23 +192,23 @@ def test_hn_nested_img():
         ("alt='Alt Text' title='Optional title'", "Alt Text", " \"Optional title\""),
     ]
     for image_attributes, markdown, title in image_attributes_to_markdown:
-        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>') == '\n\n### A' + (' ' + markdown + ' ' if markdown else ' ') + 'B\n\n'
-        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>', keep_inline_images_in=['h3']) == '\n\n### A ![' + markdown + '](/path/to/img.jpg' + title + ') B\n\n'
+        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>') == '### A' + (' ' + markdown + ' ' if markdown else ' ') + 'B\n\n'
+        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>', keep_inline_images_in=['h3']) == '### A ![' + markdown + '](/path/to/img.jpg' + title + ') B\n\n'
 
 
 def test_hn_atx_headings():
-    assert md('<h1>Hello</h1>', heading_style=ATX) == '\n\n# Hello\n\n'
-    assert md('<h2>Hello</h2>', heading_style=ATX) == '\n\n## Hello\n\n'
+    assert md('<h1>Hello</h1>', heading_style=ATX) == '# Hello\n\n'
+    assert md('<h2>Hello</h2>', heading_style=ATX) == '## Hello\n\n'
 
 
 def test_hn_atx_closed_headings():
-    assert md('<h1>Hello</h1>', heading_style=ATX_CLOSED) == '\n\n# Hello #\n\n'
-    assert md('<h2>Hello</h2>', heading_style=ATX_CLOSED) == '\n\n## Hello ##\n\n'
+    assert md('<h1>Hello</h1>', heading_style=ATX_CLOSED) == '# Hello #\n\n'
+    assert md('<h2>Hello</h2>', heading_style=ATX_CLOSED) == '## Hello ##\n\n'
 
 
 def test_hn_newlines():
-    assert md("<h1>H1-1</h1>TEXT<h2>H2-2</h2>TEXT<h1>H1-2</h1>TEXT", heading_style=ATX) == '\n\n# H1-1\n\nTEXT\n\n## H2-2\n\nTEXT\n\n# H1-2\n\nTEXT'
-    assert md('<h1>H1-1</h1>\n<p>TEXT</p>\n<h2>H2-2</h2>\n<p>TEXT</p>\n<h1>H1-2</h1>\n<p>TEXT</p>', heading_style=ATX) == '\n\n# H1-1\n\nTEXT\n\n## H2-2\n\nTEXT\n\n# H1-2\n\nTEXT\n\n'
+    assert md("<h1>H1-1</h1>TEXT<h2>H2-2</h2>TEXT<h1>H1-2</h1>TEXT", heading_style=ATX) == '# H1-1\n\nTEXT\n\n## H2-2\n\nTEXT\n\n# H1-2\n\nTEXT'
+    assert md('<h1>H1-1</h1>\n<p>TEXT</p>\n<h2>H2-2</h2>\n<p>TEXT</p>\n<h1>H1-2</h1>\n<p>TEXT</p>', heading_style=ATX) == '# H1-1\n\nTEXT\n\n## H2-2\n\nTEXT\n\n# H1-2\n\nTEXT\n\n'
 
 
 def test_head():
@@ -218,7 +218,7 @@ def test_head():
 def test_hr():
     assert md('Hello<hr>World') == 'Hello\n\n---\n\nWorld'
     assert md('Hello<hr />World') == 'Hello\n\n---\n\nWorld'
-    assert md('<p>Hello</p>\n<hr>\n<p>World</p>') == '\n\nHello\n\n---\n\nWorld\n\n'
+    assert md('<p>Hello</p>\n<hr>\n<p>World</p>') == 'Hello\n\n---\n\nWorld\n\n'
 
 
 def test_i():
@@ -235,48 +235,49 @@ def test_kbd():
 
 
 def test_p():
-    assert md('<p>hello</p>') == '\n\nhello\n\n'
-    assert md('<p>123456789 123456789</p>') == '\n\n123456789 123456789\n\n'
-    assert md('<p>123456789\n\n\n123456789</p>') == '\n\n123456789\n123456789\n\n'
-    assert md('<p>123456789\n\n\n123456789</p>', wrap=True, wrap_width=80) == '\n\n123456789 123456789\n\n'
-    assert md('<p>123456789\n\n\n123456789</p>', wrap=True, wrap_width=None) == '\n\n123456789 123456789\n\n'
-    assert md('<p>123456789 123456789</p>', wrap=True, wrap_width=10) == '\n\n123456789\n123456789\n\n'
-    assert md('<p><a href="https://example.com">Some long link</a></p>', wrap=True, wrap_width=10) == '\n\n[Some long\nlink](https://example.com)\n\n'
-    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '\n\n12345\\\n67890\n\n'
-    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=50, newline_style=BACKSLASH) == '\n\n12345\\\n67890\n\n'
-    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '\n\n12345  \n67890\n\n'
-    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=50, newline_style=SPACES) == '\n\n12345  \n67890\n\n'
-    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '\n\n12345678901\\\n12345\n\n'
-    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=50, newline_style=BACKSLASH) == '\n\n12345678901\\\n12345\n\n'
-    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '\n\n12345678901  \n12345\n\n'
-    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=50, newline_style=SPACES) == '\n\n12345678901  \n12345\n\n'
-    assert md('<p>1234 5678 9012<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '\n\n1234 5678\n9012\\\n67890\n\n'
-    assert md('<p>1234 5678 9012<br />67890</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '\n\n1234 5678\n9012  \n67890\n\n'
+    assert md('<p>hello</p>') == 'hello\n\n'
+    assert md("<p><p>hello</p></p>") == "hello\n\n"
+    assert md('<p>123456789 123456789</p>') == '123456789 123456789\n\n'
+    assert md('<p>123456789\n\n\n123456789</p>') == '123456789\n123456789\n\n'
+    assert md('<p>123456789\n\n\n123456789</p>', wrap=True, wrap_width=80) == '123456789 123456789\n\n'
+    assert md('<p>123456789\n\n\n123456789</p>', wrap=True, wrap_width=None) == '123456789 123456789\n\n'
+    assert md('<p>123456789 123456789</p>', wrap=True, wrap_width=10) == '123456789\n123456789\n\n'
+    assert md('<p><a href="https://example.com">Some long link</a></p>', wrap=True, wrap_width=10) == '[Some long\nlink](https://example.com)\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '12345\\\n67890\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=50, newline_style=BACKSLASH) == '12345\\\n67890\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '12345  \n67890\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=50, newline_style=SPACES) == '12345  \n67890\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '12345678901\\\n12345\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=50, newline_style=BACKSLASH) == '12345678901\\\n12345\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '12345678901  \n12345\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=50, newline_style=SPACES) == '12345678901  \n12345\n\n'
+    assert md('<p>1234 5678 9012<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '1234 5678\n9012\\\n67890\n\n'
+    assert md('<p>1234 5678 9012<br />67890</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '1234 5678\n9012  \n67890\n\n'
     assert md('First<p>Second</p><p>Third</p>Fourth') == 'First\n\nSecond\n\nThird\n\nFourth'
 
 
 def test_pre():
-    assert md('<pre>test\n    foo\nbar</pre>') == '\n\n```\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre><code>test\n    foo\nbar</code></pre>') == '\n\n```\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre>*this_should_not_escape*</pre>') == '\n\n```\n*this_should_not_escape*\n```\n\n'
-    assert md('<pre><span>*this_should_not_escape*</span></pre>') == '\n\n```\n*this_should_not_escape*\n```\n\n'
-    assert md('<pre>\t\tthis  should\t\tnot  normalize</pre>') == '\n\n```\n\t\tthis  should\t\tnot  normalize\n```\n\n'
-    assert md('<pre><span>\t\tthis  should\t\tnot  normalize</span></pre>') == '\n\n```\n\t\tthis  should\t\tnot  normalize\n```\n\n'
-    assert md('<pre>foo<b>\nbar\n</b>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<i>\nbar\n</i>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo\n<i>bar</i>\nbaz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<i>\n</i>baz</pre>') == '\n\n```\nfoo\nbaz\n```\n\n'
-    assert md('<pre>foo<del>\nbar\n</del>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<em>\nbar\n</em>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<code>\nbar\n</code>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<strong>\nbar\n</strong>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<s>\nbar\n</s>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<sup>\nbar\n</sup>baz</pre>', sup_symbol='^') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>test\n    foo\nbar</pre>') == '```\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre><code>test\n    foo\nbar</code></pre>') == '```\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre>*this_should_not_escape*</pre>') == '```\n*this_should_not_escape*\n```\n\n'
+    assert md('<pre><span>*this_should_not_escape*</span></pre>') == '```\n*this_should_not_escape*\n```\n\n'
+    assert md('<pre>\t\tthis  should\t\tnot  normalize</pre>') == '```\n\t\tthis  should\t\tnot  normalize\n```\n\n'
+    assert md('<pre><span>\t\tthis  should\t\tnot  normalize</span></pre>') == '```\n\t\tthis  should\t\tnot  normalize\n```\n\n'
+    assert md('<pre>foo<b>\nbar\n</b>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<i>\nbar\n</i>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo\n<i>bar</i>\nbaz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<i>\n</i>baz</pre>') == '```\nfoo\nbaz\n```\n\n'
+    assert md('<pre>foo<del>\nbar\n</del>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<em>\nbar\n</em>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<code>\nbar\n</code>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<strong>\nbar\n</strong>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<s>\nbar\n</s>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<sup>\nbar\n</sup>baz</pre>', sup_symbol='^') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '```\nfoo\nbar\nbaz\n```\n\n'
 
     assert md('foo<pre>bar</pre>baz', sub_symbol='^') == 'foo\n\n```\nbar\n```\n\nbaz'
-    assert md("<p>foo</p>\n<pre>bar</pre>\n</p>baz</p>", sub_symbol="^") == "\n\nfoo\n\n```\nbar\n```\n\nbaz"
+    assert md("<p>foo</p>\n<pre>bar</pre>\n</p>baz</p>", sub_symbol="^") == 'foo\n\n```\nbar\n```\n\nbaz'
 
 
 def test_script():
@@ -319,24 +320,24 @@ def test_sup():
 
 
 def test_lang():
-    assert md('<pre>test\n    foo\nbar</pre>', code_language='python') == '\n\n```python\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre><code>test\n    foo\nbar</code></pre>', code_language='javascript') == '\n\n```javascript\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre>test\n    foo\nbar</pre>', code_language='python') == '```python\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre><code>test\n    foo\nbar</code></pre>', code_language='javascript') == '```javascript\ntest\n    foo\nbar\n```\n\n'
 
 
 def test_lang_callback():
     def callback(el):
         return el['class'][0] if el.has_attr('class') else None
 
-    assert md('<pre class="python">test\n    foo\nbar</pre>', code_language_callback=callback) == '\n\n```python\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre class="javascript"><code>test\n    foo\nbar</code></pre>', code_language_callback=callback) == '\n\n```javascript\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre class="javascript"><code class="javascript">test\n    foo\nbar</code></pre>', code_language_callback=callback) == '\n\n```javascript\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre class="python">test\n    foo\nbar</pre>', code_language_callback=callback) == '```python\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre class="javascript"><code>test\n    foo\nbar</code></pre>', code_language_callback=callback) == '```javascript\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre class="javascript"><code class="javascript">test\n    foo\nbar</code></pre>', code_language_callback=callback) == '```javascript\ntest\n    foo\nbar\n```\n\n'
 
 
 def test_spaces():
-    assert md('<p> a b </p> <p> c d </p>') == '\n\na b\n\nc d\n\n'
-    assert md('<p> <i>a</i> </p>') == '\n\n*a*\n\n'
+    assert md('<p> a b </p> <p> c d </p>') == 'a b\n\nc d\n\n'
+    assert md('<p> <i>a</i> </p>') == '*a*\n\n'
     assert md('test <p> again </p>') == 'test\n\nagain\n\n'
     assert md('test <blockquote> text </blockquote> after') == 'test\n> text\n\nafter'
-    assert md(' <ol> <li> x </li> <li> y </li> </ol> ') == '\n\n1. x\n2. y\n'
-    assert md(' <ul> <li> x </li> <li> y </li> </ol> ') == '\n\n* x\n* y\n'
+    assert md(' <ol> <li> x </li> <li> y </li> </ol> ') == '1. x\n2. y\n'
+    assert md(' <ul> <li> x </li> <li> y </li> </ol> ') == '* x\n* y\n'
     assert md('test <pre> foo </pre> bar') == 'test\n\n```\n foo \n```\n\nbar'
diff --git a/tests/test_lists.py b/tests/test_lists.py
index ce54a87..2d7b570 100644
--- a/tests/test_lists.py
+++ b/tests/test_lists.py
@@ -41,21 +41,21 @@
 
 
 def test_ol():
-    assert md('<ol><li>a</li><li>b</li></ol>') == '\n\n1. a\n2. b\n'
-    assert md('<ol start="3"><li>a</li><li>b</li></ol>') == '\n\n3. a\n4. b\n'
+    assert md('<ol><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
+    assert md('<ol start="3"><li>a</li><li>b</li></ol>') == '3. a\n4. b\n'
     assert md('foo<ol start="3"><li>a</li><li>b</li></ol>bar') == 'foo\n\n3. a\n4. b\n\nbar'
-    assert md('<ol start="-1"><li>a</li><li>b</li></ol>') == '\n\n1. a\n2. b\n'
-    assert md('<ol start="foo"><li>a</li><li>b</li></ol>') == '\n\n1. a\n2. b\n'
-    assert md('<ol start="1.5"><li>a</li><li>b</li></ol>') == '\n\n1. a\n2. b\n'
-    assert md('<ol start="1234"><li><p>first para</p><p>second para</p></li><li><p>third para</p><p>fourth para</p></li></ol>') == '\n\n1234. first para\n\n      second para\n1235. third para\n\n      fourth para\n'
+    assert md('<ol start="-1"><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
+    assert md('<ol start="foo"><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
+    assert md('<ol start="1.5"><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
+    assert md('<ol start="1234"><li><p>first para</p><p>second para</p></li><li><p>third para</p><p>fourth para</p></li></ol>') == '1234. first para\n\n      second para\n1235. third para\n\n      fourth para\n'
 
 
 def test_nested_ols():
-    assert md(nested_ols) == '\n\n1. 1\n   1. a\n      1. I\n      2. II\n      3. III\n   2. b\n   3. c\n2. 2\n3. 3\n'
+    assert md(nested_ols) == '1. 1\n   1. a\n      1. I\n      2. II\n      3. III\n   2. b\n   3. c\n2. 2\n3. 3\n'
 
 
 def test_ul():
-    assert md('<ul><li>a</li><li>b</li></ul>') == '\n\n* a\n* b\n'
+    assert md('<ul><li>a</li><li>b</li></ul>') == '* a\n* b\n'
     assert md("""<ul>
      <li>
              a
@@ -63,12 +63,12 @@ def test_ul():
      <li> b </li>
      <li>   c
      </li>
- </ul>""") == '\n\n* a\n* b\n* c\n'
-    assert md('<ul><li><p>first para</p><p>second para</p></li><li><p>third para</p><p>fourth para</p></li></ul>') == '\n\n* first para\n\n  second para\n* third para\n\n  fourth para\n'
+ </ul>""") == '* a\n* b\n* c\n'
+    assert md('<ul><li><p>first para</p><p>second para</p></li><li><p>third para</p><p>fourth para</p></li></ul>') == '* first para\n\n  second para\n* third para\n\n  fourth para\n'
 
 
 def test_inline_ul():
-    assert md('<p>foo</p><ul><li>a</li><li>b</li></ul><p>bar</p>') == '\n\nfoo\n\n* a\n* b\n\nbar\n\n'
+    assert md('<p>foo</p><ul><li>a</li><li>b</li></ul><p>bar</p>') == 'foo\n\n* a\n* b\n\nbar\n\n'
     assert md('foo<ul><li>bar</li></ul>baz') == 'foo\n\n* bar\n\nbaz'
 
 
@@ -77,12 +77,12 @@ def test_nested_uls():
     Nested ULs should alternate bullet characters.
 
     """
-    assert md(nested_uls) == '\n\n* 1\n  + a\n    - I\n    - II\n    - III\n  + b\n  + c\n* 2\n* 3\n'
+    assert md(nested_uls) == '* 1\n  + a\n    - I\n    - II\n    - III\n  + b\n  + c\n* 2\n* 3\n'
 
 
 def test_bullets():
-    assert md(nested_uls, bullets='-') == '\n\n- 1\n  - a\n    - I\n    - II\n    - III\n  - b\n  - c\n- 2\n- 3\n'
+    assert md(nested_uls, bullets='-') == '- 1\n  - a\n    - I\n    - II\n    - III\n  - b\n  - c\n- 2\n- 3\n'
 
 
 def test_li_text():
-    assert md('<ul><li>foo <a href="#">bar</a></li><li>foo bar  </li><li>foo <b>bar</b>   <i>space</i>.</ul>') == '\n\n* foo [bar](#)\n* foo bar\n* foo **bar** *space*.\n'
+    assert md('<ul><li>foo <a href="#">bar</a></li><li>foo bar  </li><li>foo <b>bar</b>   <i>space</i>.</ul>') == '* foo [bar](#)\n* foo bar\n* foo **bar** *space*.\n'
diff --git a/tests/test_tables.py b/tests/test_tables.py
index da4bf53..dee0960 100644
--- a/tests/test_tables.py
+++ b/tests/test_tables.py
@@ -228,7 +228,10 @@
     </tbody>
 </table>"""
 
-table_with_caption = """TEXT<table><caption>Caption</caption>
+table_with_caption = """TEXT<table>
+    <caption>
+        Caption
+    </caption>
     <tbody><tr><td>Firstname</td>
             <td>Lastname</td>
             <td>Age</td>
@@ -266,34 +269,34 @@
 
 
 def test_table():
-    assert md(table) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_html_content) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_paragraphs) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_linebreaks) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith Jackson | 50 |\n| Eve | Jackson Smith | 94 |\n\n'
-    assert md(table_with_header_column) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_head_body) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_head_body_multiple_head) == '\n\n|  |  |  |\n| --- | --- | --- |\n| Creator | Editor | Server |\n| Operator | Manager | Engineer |\n| Bob | Oliver | Tom |\n| Thomas | Lucas | Ethan |\n\n'
-    assert md(table_head_body_missing_head) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_missing_text) == '\n\n|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_missing_head) == '\n\n|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_body) == '\n\n|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_html_content) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_paragraphs) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_linebreaks) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith Jackson | 50 |\n| Eve | Jackson Smith | 94 |\n\n'
+    assert md(table_with_header_column) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_head_body) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_head_body_multiple_head) == '|  |  |  |\n| --- | --- | --- |\n| Creator | Editor | Server |\n| Operator | Manager | Engineer |\n| Bob | Oliver | Tom |\n| Thomas | Lucas | Ethan |\n\n'
+    assert md(table_head_body_missing_head) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_missing_text) == '|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_missing_head) == '|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_body) == '|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
     assert md(table_with_caption) == 'TEXT\n\nCaption\n\n|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n\n'
-    assert md(table_with_colspan) == '\n\n| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_undefined_colspan) == '\n\n| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'
+    assert md(table_with_colspan) == '| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_undefined_colspan) == '| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'
 
 
 def test_table_infer_header():
-    assert md(table, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_html_content, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_paragraphs, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_linebreaks, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith Jackson | 50 |\n| Eve | Jackson Smith | 94 |\n\n'
-    assert md(table_with_header_column, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_head_body, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_head_body_multiple_head, table_infer_header=True) == '\n\n| Creator | Editor | Server |\n| --- | --- | --- |\n| Operator | Manager | Engineer |\n| Bob | Oliver | Tom |\n| Thomas | Lucas | Ethan |\n\n'
-    assert md(table_head_body_missing_head, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_missing_text, table_infer_header=True) == '\n\n|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_missing_head, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_body, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_html_content, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_paragraphs, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_linebreaks, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith Jackson | 50 |\n| Eve | Jackson Smith | 94 |\n\n'
+    assert md(table_with_header_column, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_head_body, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_head_body_multiple_head, table_infer_header=True) == '| Creator | Editor | Server |\n| --- | --- | --- |\n| Operator | Manager | Engineer |\n| Bob | Oliver | Tom |\n| Thomas | Lucas | Ethan |\n\n'
+    assert md(table_head_body_missing_head, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_missing_text, table_infer_header=True) == '|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_missing_head, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_body, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
     assert md(table_with_caption, table_infer_header=True) == 'TEXT\n\nCaption\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n\n'
-    assert md(table_with_colspan, table_infer_header=True) == '\n\n| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_undefined_colspan, table_infer_header=True) == '\n\n| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'
+    assert md(table_with_colspan, table_infer_header=True) == '| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_undefined_colspan, table_infer_header=True) == '| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'

From dd251c6b8931cfe6fab31e0eaf7be9cd61eff1ce Mon Sep 17 00:00:00 2001
From: chrispy <chrispy@synopsys.com>
Date: Mon, 27 Jan 2025 09:04:47 -0500
Subject: [PATCH 2/3] disable document stripping for unit tests

Signed-off-by: chrispy <chrispy@synopsys.com>
---
 README.rst                |   7 ++
 markdownify/__init__.py   |  18 +++-
 tests/test_advanced.py    |  11 ++-
 tests/test_args.py        |  10 ++-
 tests/test_basic.py       |   9 +-
 tests/test_conversions.py | 181 ++++++++++++++++++++------------------
 tests/test_escaping.py    |   9 +-
 tests/test_lists.py       |  37 ++++----
 tests/test_tables.py      |  61 +++++++------
 9 files changed, 207 insertions(+), 136 deletions(-)

diff --git a/README.rst b/README.rst
index 34ed7e0..5c655d2 100644
--- a/README.rst
+++ b/README.rst
@@ -150,6 +150,13 @@ wrap, wrap_width
   Use with ``newline_style=BACKSLASH`` to keep line breaks in paragraphs.
   A `wrap_width` value of `None` reflows lines to unlimited line length.
 
+strip_document
+  Controls whether leading and/or trailing separation newlines are removed from
+  the final converted document. Supported values are ``LSTRIP`` (leading),
+  ``RSTRIP`` (trailing), ``STRIP`` (both), and ``None`` (no removal). Newlines
+  within the document are unaffected.
+  Defaults to ``LSTRIP``.
+
 Options may be specified as kwargs to the ``markdownify`` function, or as a
 nested ``Options`` class in ``MarkdownConverter`` subclasses.
 
diff --git a/markdownify/__init__.py b/markdownify/__init__.py
index 10ff153..6f77d59 100644
--- a/markdownify/__init__.py
+++ b/markdownify/__init__.py
@@ -26,6 +26,11 @@
 ASTERISK = '*'
 UNDERSCORE = '_'
 
+# Document strip styles
+LSTRIP = 'lstrip'
+RSTRIP = 'rstrip'
+STRIP = 'strip'
+
 
 def chomp(text):
     """
@@ -99,6 +104,7 @@ class DefaultOptions:
         keep_inline_images_in = []
         newline_style = SPACES
         strip = None
+        strip_document = LSTRIP
         strong_em_symbol = ASTERISK
         sub_symbol = ''
         sup_symbol = ''
@@ -181,8 +187,16 @@ def process_tag(self, node, convert_as_inline):
 
     def convert__document_(self, el, text, convert_as_inline):
         """Final document-level formatting for BeautifulSoup object (node.name == "[document]")"""
-        # remove all leading newlines
-        text = text.lstrip('\n')
+        if self.options['strip_document'] == LSTRIP:
+            text = text.lstrip('\n')  # remove leading separation newlines
+        elif self.options['strip_document'] == RSTRIP:
+            text = text.rstrip('\n')  # remove trailing separation newlines
+        elif self.options['strip_document'] == STRIP:
+            text = text.strip('\n')  # remove leading and trailing separation newlines
+        elif self.options['strip_document'] is None:
+            pass  # leave leading and trailing separation newlines as-is
+        else:
+            raise ValueError('Invalid value for strip_document: %s' % self.options['strip_document'])
 
         return text
 
diff --git a/tests/test_advanced.py b/tests/test_advanced.py
index 14bf3cd..286b354 100644
--- a/tests/test_advanced.py
+++ b/tests/test_advanced.py
@@ -1,4 +1,11 @@
-from markdownify import markdownify as md
+from markdownify import MarkdownConverter
+
+
+def md(html, **options):
+    # disable document-level stripping so separation newlines are included in testing
+    options = {**options, "strip_document": None}
+
+    return MarkdownConverter(**options).convert(html)
 
 
 def test_chomp():
@@ -14,7 +21,7 @@ def test_chomp():
 
 def test_nested():
     text = md('<p>This is an <a href="http://example.com/">example link</a>.</p>')
-    assert text == 'This is an [example link](http://example.com/).\n\n'
+    assert text == '\n\nThis is an [example link](http://example.com/).\n\n'
 
 
 def test_ignore_comments():
diff --git a/tests/test_args.py b/tests/test_args.py
index ebce4a8..30324ad 100644
--- a/tests/test_args.py
+++ b/tests/test_args.py
@@ -2,7 +2,7 @@
 Test whitelisting/blacklisting of specific tags.
 
 """
-from markdownify import markdownify as md
+from markdownify import markdownify as md, LSTRIP, RSTRIP, STRIP
 
 
 def test_strip():
@@ -23,3 +23,11 @@ def test_convert():
 def test_do_not_convert():
     text = md('<a href="https://github.com/matthewwithanm">Some Text</a>', convert=[])
     assert text == 'Some Text'
+
+
+def test_strip_document():
+    assert md("<p>Hello</p>") == "Hello\n\n"  # defaults to LSTRIP
+    assert md("<p>Hello</p>", strip_document=LSTRIP) == "Hello\n\n"
+    assert md("<p>Hello</p>", strip_document=RSTRIP) == "\n\nHello"
+    assert md("<p>Hello</p>", strip_document=STRIP) == "Hello"
+    assert md("<p>Hello</p>", strip_document=None) == "\n\nHello\n\n"
diff --git a/tests/test_basic.py b/tests/test_basic.py
index 66f8b6c..b91d8ee 100644
--- a/tests/test_basic.py
+++ b/tests/test_basic.py
@@ -1,4 +1,11 @@
-from markdownify import markdownify as md
+from markdownify import MarkdownConverter
+
+
+def md(html, **options):
+    # disable document-level stripping so separation newlines are included in testing
+    options = {**options, "strip_document": None}
+
+    return MarkdownConverter(**options).convert(html)
 
 
 def test_single_tag():
diff --git a/tests/test_conversions.py b/tests/test_conversions.py
index 4df024e..ed61643 100644
--- a/tests/test_conversions.py
+++ b/tests/test_conversions.py
@@ -1,4 +1,11 @@
-from markdownify import markdownify as md, ATX, ATX_CLOSED, BACKSLASH, SPACES, UNDERSCORE
+from markdownify import MarkdownConverter, ATX, ATX_CLOSED, BACKSLASH, SPACES, UNDERSCORE
+
+
+def md(html, **options):
+    # disable document-level stripping so separation newlines are included in testing
+    options = {**options, "strip_document": None}
+
+    return MarkdownConverter(**options).convert(html)
 
 
 def inline_tests(tag, markup):
@@ -41,7 +48,7 @@ def test_a_no_autolinks():
 
 def test_a_in_code():
     assert md('<code><a href="https://google.com">Google</a></code>') == '`Google`'
-    assert md('<pre><a href="https://google.com">Google</a></pre>') == '```\nGoogle\n```\n\n'
+    assert md('<pre><a href="https://google.com">Google</a></pre>') == '\n\n```\nGoogle\n```\n\n'
 
 
 def test_b():
@@ -56,22 +63,22 @@ def test_b_spaces():
 
 
 def test_blockquote():
-    assert md('<blockquote>Hello</blockquote>') == '> Hello\n\n'
-    assert md('<blockquote>\nHello\n</blockquote>') == '> Hello\n\n'
+    assert md('<blockquote>Hello</blockquote>') == '\n> Hello\n\n'
+    assert md('<blockquote>\nHello\n</blockquote>') == '\n> Hello\n\n'
 
 
 def test_blockquote_with_nested_paragraph():
-    assert md('<blockquote><p>Hello</p></blockquote>') == '> Hello\n\n'
-    assert md('<blockquote><p>Hello</p><p>Hello again</p></blockquote>') == '> Hello\n>\n> Hello again\n\n'
+    assert md('<blockquote><p>Hello</p></blockquote>') == '\n> Hello\n\n'
+    assert md('<blockquote><p>Hello</p><p>Hello again</p></blockquote>') == '\n> Hello\n>\n> Hello again\n\n'
 
 
 def test_blockquote_with_paragraph():
-    assert md('<blockquote>Hello</blockquote><p>handsome</p>') == '> Hello\n\nhandsome\n\n'
+    assert md('<blockquote>Hello</blockquote><p>handsome</p>') == '\n> Hello\n\nhandsome\n\n'
 
 
 def test_blockquote_nested():
     text = md('<blockquote>And she was like <blockquote>Hello</blockquote></blockquote>')
-    assert text == '> And she was like\n> > Hello\n\n'
+    assert text == '\n> And she was like\n> > Hello\n\n'
 
 
 def test_br():
@@ -100,13 +107,13 @@ def test_code():
 
 
 def test_dl():
-    assert md('<dl><dt>term</dt><dd>definition</dd></dl>') == 'term\n:   definition\n'
-    assert md('<dl><dt><p>te</p><p>rm</p></dt><dd>definition</dd></dl>') == 'te rm\n:   definition\n'
-    assert md('<dl><dt>term</dt><dd><p>definition-p1</p><p>definition-p2</p></dd></dl>') == 'term\n:   definition-p1\n\n    definition-p2\n'
-    assert md('<dl><dt>term</dt><dd><p>definition 1</p></dd><dd><p>definition 2</p></dd></dl>') == 'term\n:   definition 1\n:   definition 2\n'
-    assert md('<dl><dt>term 1</dt><dd>definition 1</dd><dt>term 2</dt><dd>definition 2</dd></dl>') == 'term 1\n:   definition 1\nterm 2\n:   definition 2\n'
-    assert md('<dl><dt>term</dt><dd><blockquote><p>line 1</p><p>line 2</p></blockquote></dd></dl>') == 'term\n:   > line 1\n    >\n    > line 2\n'
-    assert md('<dl><dt>term</dt><dd><ol><li><p>1</p><ul><li>2a</li><li>2b</li></ul></li><li><p>3</p></li></ol></dd></dl>') == 'term\n:   1. 1\n\n       * 2a\n       * 2b\n    2. 3\n'
+    assert md('<dl><dt>term</dt><dd>definition</dd></dl>') == '\nterm\n:   definition\n'
+    assert md('<dl><dt><p>te</p><p>rm</p></dt><dd>definition</dd></dl>') == '\nte rm\n:   definition\n'
+    assert md('<dl><dt>term</dt><dd><p>definition-p1</p><p>definition-p2</p></dd></dl>') == '\nterm\n:   definition-p1\n\n    definition-p2\n'
+    assert md('<dl><dt>term</dt><dd><p>definition 1</p></dd><dd><p>definition 2</p></dd></dl>') == '\nterm\n:   definition 1\n:   definition 2\n'
+    assert md('<dl><dt>term 1</dt><dd>definition 1</dd><dt>term 2</dt><dd>definition 2</dd></dl>') == '\nterm 1\n:   definition 1\nterm 2\n:   definition 2\n'
+    assert md('<dl><dt>term</dt><dd><blockquote><p>line 1</p><p>line 2</p></blockquote></dd></dl>') == '\nterm\n:   > line 1\n    >\n    > line 2\n'
+    assert md('<dl><dt>term</dt><dd><ol><li><p>1</p><ul><li>2a</li><li>2b</li></ul></li><li><p>3</p></li></ol></dd></dl>') == '\nterm\n:   1. 1\n\n       * 2a\n       * 2b\n    2. 3\n'
 
 
 def test_del():
@@ -127,41 +134,41 @@ def test_figcaption():
 
 
 def test_header_with_space():
-    assert md('<h3>\n\nHello</h3>') == '### Hello\n\n'
-    assert md('<h3>Hello\n\n\nWorld</h3>') == '### Hello World\n\n'
-    assert md('<h4>\n\nHello</h4>') == '#### Hello\n\n'
-    assert md('<h5>\n\nHello</h5>') == '##### Hello\n\n'
-    assert md('<h5>\n\nHello\n\n</h5>') == '##### Hello\n\n'
-    assert md('<h5>\n\nHello   \n\n</h5>') == '##### Hello\n\n'
+    assert md('<h3>\n\nHello</h3>') == '\n\n### Hello\n\n'
+    assert md('<h3>Hello\n\n\nWorld</h3>') == '\n\n### Hello World\n\n'
+    assert md('<h4>\n\nHello</h4>') == '\n\n#### Hello\n\n'
+    assert md('<h5>\n\nHello</h5>') == '\n\n##### Hello\n\n'
+    assert md('<h5>\n\nHello\n\n</h5>') == '\n\n##### Hello\n\n'
+    assert md('<h5>\n\nHello   \n\n</h5>') == '\n\n##### Hello\n\n'
 
 
 def test_h1():
-    assert md('<h1>Hello</h1>') == 'Hello\n=====\n\n'
+    assert md('<h1>Hello</h1>') == '\n\nHello\n=====\n\n'
 
 
 def test_h2():
-    assert md('<h2>Hello</h2>') == 'Hello\n-----\n\n'
+    assert md('<h2>Hello</h2>') == '\n\nHello\n-----\n\n'
 
 
 def test_hn():
-    assert md('<h3>Hello</h3>') == '### Hello\n\n'
-    assert md('<h4>Hello</h4>') == '#### Hello\n\n'
-    assert md('<h5>Hello</h5>') == '##### Hello\n\n'
-    assert md('<h6>Hello</h6>') == '###### Hello\n\n'
+    assert md('<h3>Hello</h3>') == '\n\n### Hello\n\n'
+    assert md('<h4>Hello</h4>') == '\n\n#### Hello\n\n'
+    assert md('<h5>Hello</h5>') == '\n\n##### Hello\n\n'
+    assert md('<h6>Hello</h6>') == '\n\n###### Hello\n\n'
     assert md('<h10>Hello</h10>') == md('<h6>Hello</h6>')
     assert md('<hn>Hello</hn>') == md('Hello')
 
 
 def test_hn_chained():
-    assert md('<h1>First</h1>\n<h2>Second</h2>\n<h3>Third</h3>', heading_style=ATX) == '# First\n\n## Second\n\n### Third\n\n'
+    assert md('<h1>First</h1>\n<h2>Second</h2>\n<h3>Third</h3>', heading_style=ATX) == '\n\n# First\n\n## Second\n\n### Third\n\n'
     assert md('X<h1>First</h1>', heading_style=ATX) == 'X\n\n# First\n\n'
     assert md('X<h1>First</h1>', heading_style=ATX_CLOSED) == 'X\n\n# First #\n\n'
     assert md('X<h1>First</h1>') == 'X\n\nFirst\n=====\n\n'
 
 
 def test_hn_nested_tag_heading_style():
-    assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX_CLOSED) == '# A P C #\n\n'
-    assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX) == '# A P C\n\n'
+    assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX_CLOSED) == '\n\n# A P C #\n\n'
+    assert md('<h1>A <p>P</p> C </h1>', heading_style=ATX) == '\n\n# A P C\n\n'
 
 
 def test_hn_nested_simple_tag():
@@ -177,9 +184,9 @@ def test_hn_nested_simple_tag():
     ]
 
     for tag, markdown in tag_to_markdown:
-        assert md('<h3>A <' + tag + '>' + tag + '</' + tag + '> B</h3>') == '### A ' + markdown + ' B\n\n'
+        assert md('<h3>A <' + tag + '>' + tag + '</' + tag + '> B</h3>') == '\n\n### A ' + markdown + ' B\n\n'
 
-    assert md('<h3>A <br>B</h3>', heading_style=ATX) == '### A B\n\n'
+    assert md('<h3>A <br>B</h3>', heading_style=ATX) == '\n\n### A B\n\n'
 
     # Nested lists not supported
     # assert md('<h3>A <ul><li>li1</i><li>l2</li></ul></h3>', heading_style=ATX) == '\n### A li1 li2 B\n\n'
@@ -192,23 +199,23 @@ def test_hn_nested_img():
         ("alt='Alt Text' title='Optional title'", "Alt Text", " \"Optional title\""),
     ]
     for image_attributes, markdown, title in image_attributes_to_markdown:
-        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>') == '### A' + (' ' + markdown + ' ' if markdown else ' ') + 'B\n\n'
-        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>', keep_inline_images_in=['h3']) == '### A ![' + markdown + '](/path/to/img.jpg' + title + ') B\n\n'
+        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>') == '\n\n### A' + (' ' + markdown + ' ' if markdown else ' ') + 'B\n\n'
+        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>', keep_inline_images_in=['h3']) == '\n\n### A ![' + markdown + '](/path/to/img.jpg' + title + ') B\n\n'
 
 
 def test_hn_atx_headings():
-    assert md('<h1>Hello</h1>', heading_style=ATX) == '# Hello\n\n'
-    assert md('<h2>Hello</h2>', heading_style=ATX) == '## Hello\n\n'
+    assert md('<h1>Hello</h1>', heading_style=ATX) == '\n\n# Hello\n\n'
+    assert md('<h2>Hello</h2>', heading_style=ATX) == '\n\n## Hello\n\n'
 
 
 def test_hn_atx_closed_headings():
-    assert md('<h1>Hello</h1>', heading_style=ATX_CLOSED) == '# Hello #\n\n'
-    assert md('<h2>Hello</h2>', heading_style=ATX_CLOSED) == '## Hello ##\n\n'
+    assert md('<h1>Hello</h1>', heading_style=ATX_CLOSED) == '\n\n# Hello #\n\n'
+    assert md('<h2>Hello</h2>', heading_style=ATX_CLOSED) == '\n\n## Hello ##\n\n'
 
 
 def test_hn_newlines():
-    assert md("<h1>H1-1</h1>TEXT<h2>H2-2</h2>TEXT<h1>H1-2</h1>TEXT", heading_style=ATX) == '# H1-1\n\nTEXT\n\n## H2-2\n\nTEXT\n\n# H1-2\n\nTEXT'
-    assert md('<h1>H1-1</h1>\n<p>TEXT</p>\n<h2>H2-2</h2>\n<p>TEXT</p>\n<h1>H1-2</h1>\n<p>TEXT</p>', heading_style=ATX) == '# H1-1\n\nTEXT\n\n## H2-2\n\nTEXT\n\n# H1-2\n\nTEXT\n\n'
+    assert md("<h1>H1-1</h1>TEXT<h2>H2-2</h2>TEXT<h1>H1-2</h1>TEXT", heading_style=ATX) == '\n\n# H1-1\n\nTEXT\n\n## H2-2\n\nTEXT\n\n# H1-2\n\nTEXT'
+    assert md('<h1>H1-1</h1>\n<p>TEXT</p>\n<h2>H2-2</h2>\n<p>TEXT</p>\n<h1>H1-2</h1>\n<p>TEXT</p>', heading_style=ATX) == '\n\n# H1-1\n\nTEXT\n\n## H2-2\n\nTEXT\n\n# H1-2\n\nTEXT\n\n'
 
 
 def test_head():
@@ -218,7 +225,7 @@ def test_head():
 def test_hr():
     assert md('Hello<hr>World') == 'Hello\n\n---\n\nWorld'
     assert md('Hello<hr />World') == 'Hello\n\n---\n\nWorld'
-    assert md('<p>Hello</p>\n<hr>\n<p>World</p>') == 'Hello\n\n---\n\nWorld\n\n'
+    assert md('<p>Hello</p>\n<hr>\n<p>World</p>') == '\n\nHello\n\n---\n\nWorld\n\n'
 
 
 def test_i():
@@ -235,49 +242,49 @@ def test_kbd():
 
 
 def test_p():
-    assert md('<p>hello</p>') == 'hello\n\n'
-    assert md("<p><p>hello</p></p>") == "hello\n\n"
-    assert md('<p>123456789 123456789</p>') == '123456789 123456789\n\n'
-    assert md('<p>123456789\n\n\n123456789</p>') == '123456789\n123456789\n\n'
-    assert md('<p>123456789\n\n\n123456789</p>', wrap=True, wrap_width=80) == '123456789 123456789\n\n'
-    assert md('<p>123456789\n\n\n123456789</p>', wrap=True, wrap_width=None) == '123456789 123456789\n\n'
-    assert md('<p>123456789 123456789</p>', wrap=True, wrap_width=10) == '123456789\n123456789\n\n'
-    assert md('<p><a href="https://example.com">Some long link</a></p>', wrap=True, wrap_width=10) == '[Some long\nlink](https://example.com)\n\n'
-    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '12345\\\n67890\n\n'
-    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=50, newline_style=BACKSLASH) == '12345\\\n67890\n\n'
-    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '12345  \n67890\n\n'
-    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=50, newline_style=SPACES) == '12345  \n67890\n\n'
-    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '12345678901\\\n12345\n\n'
-    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=50, newline_style=BACKSLASH) == '12345678901\\\n12345\n\n'
-    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '12345678901  \n12345\n\n'
-    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=50, newline_style=SPACES) == '12345678901  \n12345\n\n'
-    assert md('<p>1234 5678 9012<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '1234 5678\n9012\\\n67890\n\n'
-    assert md('<p>1234 5678 9012<br />67890</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '1234 5678\n9012  \n67890\n\n'
+    assert md('<p>hello</p>') == '\n\nhello\n\n'
+    assert md("<p><p>hello</p></p>") == "\n\nhello\n\n"
+    assert md('<p>123456789 123456789</p>') == '\n\n123456789 123456789\n\n'
+    assert md('<p>123456789\n\n\n123456789</p>') == '\n\n123456789\n123456789\n\n'
+    assert md('<p>123456789\n\n\n123456789</p>', wrap=True, wrap_width=80) == '\n\n123456789 123456789\n\n'
+    assert md('<p>123456789\n\n\n123456789</p>', wrap=True, wrap_width=None) == '\n\n123456789 123456789\n\n'
+    assert md('<p>123456789 123456789</p>', wrap=True, wrap_width=10) == '\n\n123456789\n123456789\n\n'
+    assert md('<p><a href="https://example.com">Some long link</a></p>', wrap=True, wrap_width=10) == '\n\n[Some long\nlink](https://example.com)\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '\n\n12345\\\n67890\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=50, newline_style=BACKSLASH) == '\n\n12345\\\n67890\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '\n\n12345  \n67890\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=50, newline_style=SPACES) == '\n\n12345  \n67890\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '\n\n12345678901\\\n12345\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=50, newline_style=BACKSLASH) == '\n\n12345678901\\\n12345\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '\n\n12345678901  \n12345\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=50, newline_style=SPACES) == '\n\n12345678901  \n12345\n\n'
+    assert md('<p>1234 5678 9012<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '\n\n1234 5678\n9012\\\n67890\n\n'
+    assert md('<p>1234 5678 9012<br />67890</p>', wrap=True, wrap_width=10, newline_style=SPACES) == '\n\n1234 5678\n9012  \n67890\n\n'
     assert md('First<p>Second</p><p>Third</p>Fourth') == 'First\n\nSecond\n\nThird\n\nFourth'
 
 
 def test_pre():
-    assert md('<pre>test\n    foo\nbar</pre>') == '```\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre><code>test\n    foo\nbar</code></pre>') == '```\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre>*this_should_not_escape*</pre>') == '```\n*this_should_not_escape*\n```\n\n'
-    assert md('<pre><span>*this_should_not_escape*</span></pre>') == '```\n*this_should_not_escape*\n```\n\n'
-    assert md('<pre>\t\tthis  should\t\tnot  normalize</pre>') == '```\n\t\tthis  should\t\tnot  normalize\n```\n\n'
-    assert md('<pre><span>\t\tthis  should\t\tnot  normalize</span></pre>') == '```\n\t\tthis  should\t\tnot  normalize\n```\n\n'
-    assert md('<pre>foo<b>\nbar\n</b>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<i>\nbar\n</i>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo\n<i>bar</i>\nbaz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<i>\n</i>baz</pre>') == '```\nfoo\nbaz\n```\n\n'
-    assert md('<pre>foo<del>\nbar\n</del>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<em>\nbar\n</em>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<code>\nbar\n</code>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<strong>\nbar\n</strong>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<s>\nbar\n</s>baz</pre>') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<sup>\nbar\n</sup>baz</pre>', sup_symbol='^') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '```\nfoo\nbar\nbaz\n```\n\n'
-    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>test\n    foo\nbar</pre>') == '\n\n```\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre><code>test\n    foo\nbar</code></pre>') == '\n\n```\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre>*this_should_not_escape*</pre>') == '\n\n```\n*this_should_not_escape*\n```\n\n'
+    assert md('<pre><span>*this_should_not_escape*</span></pre>') == '\n\n```\n*this_should_not_escape*\n```\n\n'
+    assert md('<pre>\t\tthis  should\t\tnot  normalize</pre>') == '\n\n```\n\t\tthis  should\t\tnot  normalize\n```\n\n'
+    assert md('<pre><span>\t\tthis  should\t\tnot  normalize</span></pre>') == '\n\n```\n\t\tthis  should\t\tnot  normalize\n```\n\n'
+    assert md('<pre>foo<b>\nbar\n</b>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<i>\nbar\n</i>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo\n<i>bar</i>\nbaz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<i>\n</i>baz</pre>') == '\n\n```\nfoo\nbaz\n```\n\n'
+    assert md('<pre>foo<del>\nbar\n</del>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<em>\nbar\n</em>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<code>\nbar\n</code>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<strong>\nbar\n</strong>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<s>\nbar\n</s>baz</pre>') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<sup>\nbar\n</sup>baz</pre>', sup_symbol='^') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
+    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '\n\n```\nfoo\nbar\nbaz\n```\n\n'
 
     assert md('foo<pre>bar</pre>baz', sub_symbol='^') == 'foo\n\n```\nbar\n```\n\nbaz'
-    assert md("<p>foo</p>\n<pre>bar</pre>\n</p>baz</p>", sub_symbol="^") == 'foo\n\n```\nbar\n```\n\nbaz'
+    assert md("<p>foo</p>\n<pre>bar</pre>\n</p>baz</p>", sub_symbol="^") == "\n\nfoo\n\n```\nbar\n```\n\nbaz"
 
 
 def test_script():
@@ -320,24 +327,24 @@ def test_sup():
 
 
 def test_lang():
-    assert md('<pre>test\n    foo\nbar</pre>', code_language='python') == '```python\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre><code>test\n    foo\nbar</code></pre>', code_language='javascript') == '```javascript\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre>test\n    foo\nbar</pre>', code_language='python') == '\n\n```python\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre><code>test\n    foo\nbar</code></pre>', code_language='javascript') == '\n\n```javascript\ntest\n    foo\nbar\n```\n\n'
 
 
 def test_lang_callback():
     def callback(el):
         return el['class'][0] if el.has_attr('class') else None
 
-    assert md('<pre class="python">test\n    foo\nbar</pre>', code_language_callback=callback) == '```python\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre class="javascript"><code>test\n    foo\nbar</code></pre>', code_language_callback=callback) == '```javascript\ntest\n    foo\nbar\n```\n\n'
-    assert md('<pre class="javascript"><code class="javascript">test\n    foo\nbar</code></pre>', code_language_callback=callback) == '```javascript\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre class="python">test\n    foo\nbar</pre>', code_language_callback=callback) == '\n\n```python\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre class="javascript"><code>test\n    foo\nbar</code></pre>', code_language_callback=callback) == '\n\n```javascript\ntest\n    foo\nbar\n```\n\n'
+    assert md('<pre class="javascript"><code class="javascript">test\n    foo\nbar</code></pre>', code_language_callback=callback) == '\n\n```javascript\ntest\n    foo\nbar\n```\n\n'
 
 
 def test_spaces():
-    assert md('<p> a b </p> <p> c d </p>') == 'a b\n\nc d\n\n'
-    assert md('<p> <i>a</i> </p>') == '*a*\n\n'
+    assert md('<p> a b </p> <p> c d </p>') == '\n\na b\n\nc d\n\n'
+    assert md('<p> <i>a</i> </p>') == '\n\n*a*\n\n'
     assert md('test <p> again </p>') == 'test\n\nagain\n\n'
     assert md('test <blockquote> text </blockquote> after') == 'test\n> text\n\nafter'
-    assert md(' <ol> <li> x </li> <li> y </li> </ol> ') == '1. x\n2. y\n'
-    assert md(' <ul> <li> x </li> <li> y </li> </ol> ') == '* x\n* y\n'
+    assert md(' <ol> <li> x </li> <li> y </li> </ol> ') == '\n\n1. x\n2. y\n'
+    assert md(' <ul> <li> x </li> <li> y </li> </ol> ') == '\n\n* x\n* y\n'
     assert md('test <pre> foo </pre> bar') == 'test\n\n```\n foo \n```\n\nbar'
diff --git a/tests/test_escaping.py b/tests/test_escaping.py
index 878760a..3634b37 100644
--- a/tests/test_escaping.py
+++ b/tests/test_escaping.py
@@ -1,6 +1,13 @@
 import warnings
 from bs4 import MarkupResemblesLocatorWarning
-from markdownify import markdownify as md
+from markdownify import MarkdownConverter
+
+
+def md(html, **options):
+    # disable document-level stripping so separation newlines are included in testing
+    options = {**options, "strip_document": None}
+
+    return MarkdownConverter(**options).convert(html)
 
 
 def test_asterisks():
diff --git a/tests/test_lists.py b/tests/test_lists.py
index 2d7b570..21ab541 100644
--- a/tests/test_lists.py
+++ b/tests/test_lists.py
@@ -1,4 +1,11 @@
-from markdownify import markdownify as md
+from markdownify import MarkdownConverter
+
+
+def md(html, **options):
+    # disable document-level stripping so separation newlines are included in testing
+    options = {**options, "strip_document": None}
+
+    return MarkdownConverter(**options).convert(html)
 
 
 nested_uls = """
@@ -41,21 +48,21 @@
 
 
 def test_ol():
-    assert md('<ol><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
-    assert md('<ol start="3"><li>a</li><li>b</li></ol>') == '3. a\n4. b\n'
+    assert md('<ol><li>a</li><li>b</li></ol>') == '\n\n1. a\n2. b\n'
+    assert md('<ol start="3"><li>a</li><li>b</li></ol>') == '\n\n3. a\n4. b\n'
     assert md('foo<ol start="3"><li>a</li><li>b</li></ol>bar') == 'foo\n\n3. a\n4. b\n\nbar'
-    assert md('<ol start="-1"><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
-    assert md('<ol start="foo"><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
-    assert md('<ol start="1.5"><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
-    assert md('<ol start="1234"><li><p>first para</p><p>second para</p></li><li><p>third para</p><p>fourth para</p></li></ol>') == '1234. first para\n\n      second para\n1235. third para\n\n      fourth para\n'
+    assert md('<ol start="-1"><li>a</li><li>b</li></ol>') == '\n\n1. a\n2. b\n'
+    assert md('<ol start="foo"><li>a</li><li>b</li></ol>') == '\n\n1. a\n2. b\n'
+    assert md('<ol start="1.5"><li>a</li><li>b</li></ol>') == '\n\n1. a\n2. b\n'
+    assert md('<ol start="1234"><li><p>first para</p><p>second para</p></li><li><p>third para</p><p>fourth para</p></li></ol>') == '\n\n1234. first para\n\n      second para\n1235. third para\n\n      fourth para\n'
 
 
 def test_nested_ols():
-    assert md(nested_ols) == '1. 1\n   1. a\n      1. I\n      2. II\n      3. III\n   2. b\n   3. c\n2. 2\n3. 3\n'
+    assert md(nested_ols) == '\n\n1. 1\n   1. a\n      1. I\n      2. II\n      3. III\n   2. b\n   3. c\n2. 2\n3. 3\n'
 
 
 def test_ul():
-    assert md('<ul><li>a</li><li>b</li></ul>') == '* a\n* b\n'
+    assert md('<ul><li>a</li><li>b</li></ul>') == '\n\n* a\n* b\n'
     assert md("""<ul>
      <li>
              a
@@ -63,12 +70,12 @@ def test_ul():
      <li> b </li>
      <li>   c
      </li>
- </ul>""") == '* a\n* b\n* c\n'
-    assert md('<ul><li><p>first para</p><p>second para</p></li><li><p>third para</p><p>fourth para</p></li></ul>') == '* first para\n\n  second para\n* third para\n\n  fourth para\n'
+ </ul>""") == '\n\n* a\n* b\n* c\n'
+    assert md('<ul><li><p>first para</p><p>second para</p></li><li><p>third para</p><p>fourth para</p></li></ul>') == '\n\n* first para\n\n  second para\n* third para\n\n  fourth para\n'
 
 
 def test_inline_ul():
-    assert md('<p>foo</p><ul><li>a</li><li>b</li></ul><p>bar</p>') == 'foo\n\n* a\n* b\n\nbar\n\n'
+    assert md('<p>foo</p><ul><li>a</li><li>b</li></ul><p>bar</p>') == '\n\nfoo\n\n* a\n* b\n\nbar\n\n'
     assert md('foo<ul><li>bar</li></ul>baz') == 'foo\n\n* bar\n\nbaz'
 
 
@@ -77,12 +84,12 @@ def test_nested_uls():
     Nested ULs should alternate bullet characters.
 
     """
-    assert md(nested_uls) == '* 1\n  + a\n    - I\n    - II\n    - III\n  + b\n  + c\n* 2\n* 3\n'
+    assert md(nested_uls) == '\n\n* 1\n  + a\n    - I\n    - II\n    - III\n  + b\n  + c\n* 2\n* 3\n'
 
 
 def test_bullets():
-    assert md(nested_uls, bullets='-') == '- 1\n  - a\n    - I\n    - II\n    - III\n  - b\n  - c\n- 2\n- 3\n'
+    assert md(nested_uls, bullets='-') == '\n\n- 1\n  - a\n    - I\n    - II\n    - III\n  - b\n  - c\n- 2\n- 3\n'
 
 
 def test_li_text():
-    assert md('<ul><li>foo <a href="#">bar</a></li><li>foo bar  </li><li>foo <b>bar</b>   <i>space</i>.</ul>') == '* foo [bar](#)\n* foo bar\n* foo **bar** *space*.\n'
+    assert md('<ul><li>foo <a href="#">bar</a></li><li>foo bar  </li><li>foo <b>bar</b>   <i>space</i>.</ul>') == '\n\n* foo [bar](#)\n* foo bar\n* foo **bar** *space*.\n'
diff --git a/tests/test_tables.py b/tests/test_tables.py
index dee0960..f0aee56 100644
--- a/tests/test_tables.py
+++ b/tests/test_tables.py
@@ -1,4 +1,11 @@
-from markdownify import markdownify as md
+from markdownify import MarkdownConverter
+
+
+def md(html, **options):
+    # disable document-level stripping so separation newlines are included in testing
+    options = {**options, "strip_document": None}
+
+    return MarkdownConverter(**options).convert(html)
 
 
 table = """<table>
@@ -269,34 +276,34 @@
 
 
 def test_table():
-    assert md(table) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_html_content) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_paragraphs) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_linebreaks) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith Jackson | 50 |\n| Eve | Jackson Smith | 94 |\n\n'
-    assert md(table_with_header_column) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_head_body) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_head_body_multiple_head) == '|  |  |  |\n| --- | --- | --- |\n| Creator | Editor | Server |\n| Operator | Manager | Engineer |\n| Bob | Oliver | Tom |\n| Thomas | Lucas | Ethan |\n\n'
-    assert md(table_head_body_missing_head) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_missing_text) == '|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_missing_head) == '|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_body) == '|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_html_content) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_paragraphs) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_linebreaks) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith Jackson | 50 |\n| Eve | Jackson Smith | 94 |\n\n'
+    assert md(table_with_header_column) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_head_body) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_head_body_multiple_head) == '\n\n|  |  |  |\n| --- | --- | --- |\n| Creator | Editor | Server |\n| Operator | Manager | Engineer |\n| Bob | Oliver | Tom |\n| Thomas | Lucas | Ethan |\n\n'
+    assert md(table_head_body_missing_head) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_missing_text) == '\n\n|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_missing_head) == '\n\n|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_body) == '\n\n|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
     assert md(table_with_caption) == 'TEXT\n\nCaption\n\n|  |  |  |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n\n'
-    assert md(table_with_colspan) == '| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_undefined_colspan) == '| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'
+    assert md(table_with_colspan) == '\n\n| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_undefined_colspan) == '\n\n| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'
 
 
 def test_table_infer_header():
-    assert md(table, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_html_content, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_paragraphs, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_linebreaks, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith Jackson | 50 |\n| Eve | Jackson Smith | 94 |\n\n'
-    assert md(table_with_header_column, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_head_body, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_head_body_multiple_head, table_infer_header=True) == '| Creator | Editor | Server |\n| --- | --- | --- |\n| Operator | Manager | Engineer |\n| Bob | Oliver | Tom |\n| Thomas | Lucas | Ethan |\n\n'
-    assert md(table_head_body_missing_head, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_missing_text, table_infer_header=True) == '|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_missing_head, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_body, table_infer_header=True) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_html_content, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_paragraphs, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_linebreaks, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith Jackson | 50 |\n| Eve | Jackson Smith | 94 |\n\n'
+    assert md(table_with_header_column, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_head_body, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_head_body_multiple_head, table_infer_header=True) == '\n\n| Creator | Editor | Server |\n| --- | --- | --- |\n| Operator | Manager | Engineer |\n| Bob | Oliver | Tom |\n| Thomas | Lucas | Ethan |\n\n'
+    assert md(table_head_body_missing_head, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_missing_text, table_infer_header=True) == '\n\n|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_missing_head, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_body, table_infer_header=True) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
     assert md(table_with_caption, table_infer_header=True) == 'TEXT\n\nCaption\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n\n'
-    assert md(table_with_colspan, table_infer_header=True) == '| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
-    assert md(table_with_undefined_colspan, table_infer_header=True) == '| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'
+    assert md(table_with_colspan, table_infer_header=True) == '\n\n| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_undefined_colspan, table_infer_header=True) == '\n\n| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'

From 1586f9834fdb6176c79c436b205c9816757502db Mon Sep 17 00:00:00 2001
From: chrispy <chrispy@synopsys.com>
Date: Mon, 27 Jan 2025 11:35:30 -0500
Subject: [PATCH 3/3] change default to strip_document=STRIP, move unit-testing
 md() to common test utils file

Signed-off-by: chrispy <chrispy@synopsys.com>
---
 README.rst                     |  4 ++--
 markdownify/__init__.py        |  2 +-
 tests/test_advanced.py         |  9 +--------
 tests/test_args.py             | 13 +++++++------
 tests/test_basic.py            |  9 +--------
 tests/test_conversions.py      | 10 ++--------
 tests/test_custom_converter.py |  4 ++--
 tests/test_escaping.py         |  9 +--------
 tests/test_lists.py            |  9 +--------
 tests/test_tables.py           |  9 +--------
 tests/utils.py                 |  9 +++++++++
 11 files changed, 28 insertions(+), 59 deletions(-)
 create mode 100644 tests/utils.py

diff --git a/README.rst b/README.rst
index 5c655d2..b37a503 100644
--- a/README.rst
+++ b/README.rst
@@ -153,9 +153,9 @@ wrap, wrap_width
 strip_document
   Controls whether leading and/or trailing separation newlines are removed from
   the final converted document. Supported values are ``LSTRIP`` (leading),
-  ``RSTRIP`` (trailing), ``STRIP`` (both), and ``None`` (no removal). Newlines
+  ``RSTRIP`` (trailing), ``STRIP`` (both), and ``None`` (neither). Newlines
   within the document are unaffected.
-  Defaults to ``LSTRIP``.
+  Defaults to ``STRIP``.
 
 Options may be specified as kwargs to the ``markdownify`` function, or as a
 nested ``Options`` class in ``MarkdownConverter`` subclasses.
diff --git a/markdownify/__init__.py b/markdownify/__init__.py
index 6f77d59..7d14fe7 100644
--- a/markdownify/__init__.py
+++ b/markdownify/__init__.py
@@ -104,7 +104,7 @@ class DefaultOptions:
         keep_inline_images_in = []
         newline_style = SPACES
         strip = None
-        strip_document = LSTRIP
+        strip_document = STRIP
         strong_em_symbol = ASTERISK
         sub_symbol = ''
         sup_symbol = ''
diff --git a/tests/test_advanced.py b/tests/test_advanced.py
index 286b354..6123d8c 100644
--- a/tests/test_advanced.py
+++ b/tests/test_advanced.py
@@ -1,11 +1,4 @@
-from markdownify import MarkdownConverter
-
-
-def md(html, **options):
-    # disable document-level stripping so separation newlines are included in testing
-    options = {**options, "strip_document": None}
-
-    return MarkdownConverter(**options).convert(html)
+from .utils import md
 
 
 def test_chomp():
diff --git a/tests/test_args.py b/tests/test_args.py
index 30324ad..301c19f 100644
--- a/tests/test_args.py
+++ b/tests/test_args.py
@@ -2,7 +2,8 @@
 Test whitelisting/blacklisting of specific tags.
 
 """
-from markdownify import markdownify as md, LSTRIP, RSTRIP, STRIP
+from markdownify import markdownify, LSTRIP, RSTRIP, STRIP
+from .utils import md
 
 
 def test_strip():
@@ -26,8 +27,8 @@ def test_do_not_convert():
 
 
 def test_strip_document():
-    assert md("<p>Hello</p>") == "Hello\n\n"  # defaults to LSTRIP
-    assert md("<p>Hello</p>", strip_document=LSTRIP) == "Hello\n\n"
-    assert md("<p>Hello</p>", strip_document=RSTRIP) == "\n\nHello"
-    assert md("<p>Hello</p>", strip_document=STRIP) == "Hello"
-    assert md("<p>Hello</p>", strip_document=None) == "\n\nHello\n\n"
+    assert markdownify("<p>Hello</p>") == "Hello"  # test default of STRIP
+    assert markdownify("<p>Hello</p>", strip_document=LSTRIP) == "Hello\n\n"
+    assert markdownify("<p>Hello</p>", strip_document=RSTRIP) == "\n\nHello"
+    assert markdownify("<p>Hello</p>", strip_document=STRIP) == "Hello"
+    assert markdownify("<p>Hello</p>", strip_document=None) == "\n\nHello\n\n"
diff --git a/tests/test_basic.py b/tests/test_basic.py
index b91d8ee..584adb9 100644
--- a/tests/test_basic.py
+++ b/tests/test_basic.py
@@ -1,11 +1,4 @@
-from markdownify import MarkdownConverter
-
-
-def md(html, **options):
-    # disable document-level stripping so separation newlines are included in testing
-    options = {**options, "strip_document": None}
-
-    return MarkdownConverter(**options).convert(html)
+from .utils import md
 
 
 def test_single_tag():
diff --git a/tests/test_conversions.py b/tests/test_conversions.py
index ed61643..1367006 100644
--- a/tests/test_conversions.py
+++ b/tests/test_conversions.py
@@ -1,11 +1,5 @@
-from markdownify import MarkdownConverter, ATX, ATX_CLOSED, BACKSLASH, SPACES, UNDERSCORE
-
-
-def md(html, **options):
-    # disable document-level stripping so separation newlines are included in testing
-    options = {**options, "strip_document": None}
-
-    return MarkdownConverter(**options).convert(html)
+from markdownify import ATX, ATX_CLOSED, BACKSLASH, SPACES, UNDERSCORE
+from .utils import md
 
 
 def inline_tests(tag, markup):
diff --git a/tests/test_custom_converter.py b/tests/test_custom_converter.py
index adc83f7..0d3f6af 100644
--- a/tests/test_custom_converter.py
+++ b/tests/test_custom_converter.py
@@ -20,8 +20,8 @@ def test_custom_conversion_functions():
     def md(html, **options):
         return UnitTestConverter(**options).convert(html)
 
-    assert md('<img src="/path/to/img.jpg" alt="Alt text" title="Optional title" />') == '![Alt text](/path/to/img.jpg "Optional title")\n\n'
-    assert md('<img src="/path/to/img.jpg" alt="Alt text" />') == '![Alt text](/path/to/img.jpg)\n\n'
+    assert md('<img src="/path/to/img.jpg" alt="Alt text" title="Optional title" />text') == '![Alt text](/path/to/img.jpg "Optional title")\n\ntext'
+    assert md('<img src="/path/to/img.jpg" alt="Alt text" />text') == '![Alt text](/path/to/img.jpg)\n\ntext'
 
     assert md("<custom-tag>text</custom-tag>") == "FUNCTION USED: text"
 
diff --git a/tests/test_escaping.py b/tests/test_escaping.py
index 3634b37..d213675 100644
--- a/tests/test_escaping.py
+++ b/tests/test_escaping.py
@@ -1,13 +1,6 @@
 import warnings
 from bs4 import MarkupResemblesLocatorWarning
-from markdownify import MarkdownConverter
-
-
-def md(html, **options):
-    # disable document-level stripping so separation newlines are included in testing
-    options = {**options, "strip_document": None}
-
-    return MarkdownConverter(**options).convert(html)
+from .utils import md
 
 
 def test_asterisks():
diff --git a/tests/test_lists.py b/tests/test_lists.py
index 21ab541..6b320ca 100644
--- a/tests/test_lists.py
+++ b/tests/test_lists.py
@@ -1,11 +1,4 @@
-from markdownify import MarkdownConverter
-
-
-def md(html, **options):
-    # disable document-level stripping so separation newlines are included in testing
-    options = {**options, "strip_document": None}
-
-    return MarkdownConverter(**options).convert(html)
+from .utils import md
 
 
 nested_uls = """
diff --git a/tests/test_tables.py b/tests/test_tables.py
index f0aee56..e41b389 100644
--- a/tests/test_tables.py
+++ b/tests/test_tables.py
@@ -1,11 +1,4 @@
-from markdownify import MarkdownConverter
-
-
-def md(html, **options):
-    # disable document-level stripping so separation newlines are included in testing
-    options = {**options, "strip_document": None}
-
-    return MarkdownConverter(**options).convert(html)
+from .utils import md
 
 
 table = """<table>
diff --git a/tests/utils.py b/tests/utils.py
new file mode 100644
index 0000000..0dac580
--- /dev/null
+++ b/tests/utils.py
@@ -0,0 +1,9 @@
+from markdownify import MarkdownConverter
+
+
+# for unit testing, disable document-level stripping by default so that
+# separation newlines are included in testing
+def md(html, **options):
+    options = {"strip_document": None, **options}
+
+    return MarkdownConverter(**options).convert(html)