Merge pull request #14 from AlexVonB/fix-inline-spaces

remove prefixed and suffixed spaces from inline tags

markdownify/__init__.py

@@ -22,6 +22,17 @@ def escape(text):
         return ''
     return text.replace('_', r'\_')
 
+def chomp(text):
+    """
+    If the text in an inline tag like b, a, or em contains a leading or trailing
+    space, strip the string and return a space as suffix of prefix, if needed.
+    This function is used to prevent conversions like
+        <b> foo</b> => ** foo**
+    """
+    prefix = ' ' if text and text[0] == ' ' else ''
+    suffix = ' ' if text and text[-1] == ' ' else ''
+    text = text.strip()
+    return (prefix, suffix, text)
 
 def _todict(obj):
     return dict((k, getattr(obj, k)) for k in dir(obj) if not k.startswith('_'))
@@ -110,13 +121,16 @@ class MarkdownConverter(object):
         return '%s\n%s\n\n' % (text, pad_char * len(text)) if text else ''
 
     def convert_a(self, el, text):
+        prefix, suffix, text = chomp(text)
+        if not text:
+            return ''
         href = el.get('href')
         title = el.get('title')
         if self.options['autolinks'] and text == href and not title:
             # Shortcut syntax
             return '<%s>' % href
         title_part = ' "%s"' % title.replace('"', r'\"') if title else ''
-        return '[%s](%s%s)' % (text or '', href, title_part) if href else text or ''
+        return '%s[%s](%s%s)%s' % (prefix, text, href, title_part, suffix) if href else text
 
     def convert_b(self, el, text):
         return self.convert_strong(el, text)
@@ -128,7 +142,10 @@ class MarkdownConverter(object):
         return '  \n'
 
     def convert_em(self, el, text):
-        return '*%s*' % text if text else ''
+        prefix, suffix, text = chomp(text)
+        if not text:
+            return ''
+        return '%s*%s*%s' % (prefix, text, suffix)
 
     def convert_hn(self, n, el, text):
         style = self.options['heading_style']
@@ -176,7 +193,10 @@ class MarkdownConverter(object):
         return '%s\n\n' % text if text else ''
 
     def convert_strong(self, el, text):
-        return '**%s**' % text if text else ''
+        prefix, suffix, text = chomp(text)
+        if not text:
+            return ''
+        return '%s**%s**%s' % (prefix, text, suffix)
 
     def convert_img(self, el, text):
         alt = el.attrs.get('alt', None) or ''

tests/test_conversions.py

@@ -21,10 +21,24 @@ nested_uls = re.sub('\s+', '', """
         <li>3</li>
     </ul>""")
 
+def test_chomp():
+    assert md(' <b></b> ') == '  '
+    assert md(' <b> </b> ') == '  '
+    assert md(' <b>  </b> ') == '  '
+    assert md(' <b>   </b> ') == '  '
+    assert md(' <b>s </b> ') == ' **s**  '
+    assert md(' <b> s</b> ') == '  **s** '
+    assert md(' <b> s </b> ') == '  **s**  '
+    assert md(' <b>  s  </b> ') == '  **s**  '
 
 def test_a():
     assert md('<a href="http://google.com">Google</a>') == '[Google](http://google.com)'
 
+def test_a_spaces():
+    assert md('foo <a href="http://google.com">Google</a> bar') == 'foo [Google](http://google.com) bar'
+    assert md('foo<a href="http://google.com"> Google</a> bar') == 'foo [Google](http://google.com) bar'
+    assert md('foo <a href="http://google.com">Google </a>bar') == 'foo [Google](http://google.com) bar'
+    assert md('foo <a href="http://google.com"></a> bar') == 'foo  bar'
 
 def test_a_with_title():
     text = md('<a href="http://google.com" title="The &quot;Goog&quot;">Google</a>')
@@ -44,6 +58,12 @@ def test_a_no_autolinks():
 def test_b():
     assert md('<b>Hello</b>') == '**Hello**'
 
+def test_b_spaces():
+    assert md('foo <b>Hello</b> bar') == 'foo **Hello** bar'
+    assert md('foo<b> Hello</b> bar') == 'foo **Hello** bar'
+    assert md('foo <b>Hello </b>bar') == 'foo **Hello** bar'
+    assert md('foo <b></b> bar') == 'foo  bar'
+
 
 def test_blockquote():
     assert md('<blockquote>Hello</blockquote>').strip() == '> Hello'
@@ -61,6 +81,12 @@ def test_br():
 def test_em():
     assert md('<em>Hello</em>') == '*Hello*'
 
+def test_em_spaces():
+    assert md('foo <em>Hello</em> bar') == 'foo *Hello* bar'
+    assert md('foo<em> Hello</em> bar') == 'foo *Hello* bar'
+    assert md('foo <em>Hello </em>bar') == 'foo *Hello* bar'
+    assert md('foo <em></em> bar') == 'foo  bar'
+
 
 def test_h1():
     assert md('<h1>Hello</h1>') == 'Hello\n=====\n\n'