Merge branch 'develop'

closes #59

.gitignore

@@ -8,3 +8,4 @@
 /MANIFEST
 /venv
 build/
+.vscode/settings.json

README.rst

@@ -96,6 +96,16 @@ newline_style
   newline). While the latter convention is non-standard, it is commonly
   preferred and supported by a lot of interpreters.
 
+code_language
+  Defines the language that should be assumed for all ``<pre>`` sections.
+  Useful, if all code on a page is in the same programming language and
+  should be annotated with `````python`` or similar.
+  Defaults to ``''`` (empty string) and can be any string.
+
+escape_underscores
+  If set to ``False``, do not escape ``_`` to ``\_`` in text.
+  Defaults to ``True``.
+
 Options may be specified as kwargs to the ``markdownify`` function, or as a
 nested ``Options`` class in ``MarkdownConverter`` subclasses.
 

markdownify/__init__.py

@@ -25,10 +25,12 @@ ASTERISK = '*'
 UNDERSCORE = '_'
 
 
-def escape(text):
+def escape(text, escape_underscores):
     if not text:
         return ''
-    return text.replace('_', r'\_')
+    if escape_underscores:
+        return text.replace('_', r'\_')
+    return text
 
 
 def chomp(text):
@@ -68,8 +70,10 @@ class MarkdownConverter(object):
     class DefaultOptions:
         autolinks = True
         bullets = '*+-'  # An iterable of bullet types.
+        code_language = ''
         convert = None
         default_title = False
+        escape_underscores = True
         heading_style = UNDERLINED
         newline_style = SPACES
         strip = None
@@ -154,7 +158,7 @@ class MarkdownConverter(object):
             text = whitespace_re.sub(' ', text)
 
         if el.parent.name != 'code':
-            text = escape(text)
+            text = escape(text, self.options['escape_underscores'])
 
         # remove trailing whitespaces if any of the following condition is true:
         # - current text node is the last node in li
@@ -314,7 +318,7 @@ class MarkdownConverter(object):
                 el = el.parent
             bullets = self.options['bullets']
             bullet = bullets[depth % len(bullets)]
-        return '%s %s\n' % (bullet, text or '')
+        return '%s %s\n' % (bullet, (text or '').strip())
 
     def convert_p(self, el, text, convert_as_inline):
         if convert_as_inline:
@@ -324,7 +328,7 @@ class MarkdownConverter(object):
     def convert_pre(self, el, text, convert_as_inline):
         if not text:
             return ''
-        return '\n```\n%s\n```\n' % text
+        return '\n```%s\n%s\n```\n' % (self.options['code_language'], text)
 
     convert_s = convert_del
 

setup.py

@@ -10,7 +10,7 @@ read = lambda filepath: codecs.open(filepath, 'r', 'utf-8').read()
 pkgmeta = {
     '__title__': 'markdownify',
     '__author__': 'Matthew Tretter',
-    '__version__': '0.9.3',
+    '__version__': '0.10.2',
 }
 
 
@@ -70,7 +70,7 @@ setup(
     zip_safe=False,
     include_package_data=True,
     setup_requires=[
-        'flake8>=3.8,<4',
+        'flake8>=3.8,<5',
     ],
     tests_require=[
         'pytest>=6.2,<7',

tests/test_conversions.py

@@ -210,3 +210,8 @@ def test_sub():
 def test_sup():
     assert md('<sup>foo</sup>') == 'foo'
     assert md('<sup>foo</sup>', sup_symbol='^') == '^foo^'
+
+
+def test_lang():
+    assert md('<pre>test\n    foo\nbar</pre>', code_language='python') == '\n```python\ntest\n    foo\nbar\n```\n'
+    assert md('<pre><code>test\n    foo\nbar</code></pre>', code_language='javascript') == '\n```javascript\ntest\n    foo\nbar\n```\n'

tests/test_escaping.py

@@ -3,6 +3,7 @@ from markdownify import markdownify as md
 
 def test_underscore():
     assert md('_hey_dude_') == r'\_hey\_dude\_'
+    assert md('_hey_dude_', escape_underscores=False) == r'_hey_dude_'
 
 
 def test_xml_entities():

tests/test_lists.py

@@ -51,6 +51,14 @@ def test_nested_ols():
 
 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>
+     <li>   c
+     </li>
+ </ul>""") == '* a\n* b\n* c\n'
 
 
 def test_inline_ul():