Merge branch 'develop'

instead of manually walking down the dom tree
in a table, we now rely on the main descent loop
and just implement conversion for rows and cells
correctly. this enables the use of html inside a
table cell.

markdownify/__init__.py

@@ -84,10 +84,26 @@ class MarkdownConverter(object):
         if not children_only and isHeading:
             convert_children_as_inline = True
 
-        # Remove whitespace-only textnodes in lists
-        if node.name in ['ol', 'ul', 'li']:
+        # Remove whitespace-only textnodes in purely nested nodes
+        def is_nested_node(el):
+            return el and el.name in ['ol', 'ul', 'li',
+                                      'table', 'thead', 'tbody', 'tfoot',
+                                      'tr', 'td', 'th']
+
+        if is_nested_node(node):
             for el in node.children:
-                if isinstance(el, NavigableString) and six.text_type(el).strip() == '':
+                # Only extract (remove) whitespace-only text node if any of the
+                # conditions is true:
+                # - el is the first element in its parent
+                # - el is the last element in its parent
+                # - el is adjacent to an nested node
+                can_extract = (not el.previous_sibling
+                               or not el.next_sibling
+                               or is_nested_node(el.previous_sibling)
+                               or is_nested_node(el.next_sibling))
+                if (isinstance(el, NavigableString)
+                        and six.text_type(el).strip() == ''
+                        and can_extract):
                     el.extract()
 
         # Convert the children first
@@ -108,7 +124,10 @@ class MarkdownConverter(object):
 
     def process_text(self, el):
         text = six.text_type(el)
-        if el.parent.name == 'li':
+        # remove trailing whitespaces if any of the following condition is true:
+        # - current text node is the last node in li
+        # - current text node is followed by an embedded list
+        if el.parent.name == 'li' and (not el.next_sibling or el.next_sibling.name in ['ul', 'ol']):
             return escape(all_whitespace_re.sub(' ', text or '')).rstrip()
         return escape(whitespace_re.sub(' ', text or ''))
 
@@ -266,21 +285,28 @@ class MarkdownConverter(object):
         return '![%s](%s%s)' % (alt, src, title_part)
 
     def convert_table(self, el, text, convert_as_inline):
-        rows = el.find_all('tr')
-        text_data = []
-        for row in rows:
-            headers = row.find_all('th')
-            columns = row.find_all('td')
-            if len(headers) > 0:
-                headers = [head.text.strip() for head in headers]
-                text_data.append('| ' + ' | '.join(headers) + ' |')
-                text_data.append('| ' + ' | '.join(['---'] * len(headers)) + ' |')
-            elif len(columns) > 0:
-                columns = [colm.text.strip() for colm in columns]
-                text_data.append('| ' + ' | '.join(columns) + ' |')
-            else:
-                continue
-        return '\n'.join(text_data)
+        return '\n\n' + text + '\n'
+
+    def convert_tr(self, el, text, convert_as_inline):
+        cells = el.find_all(['td', 'th'])
+        is_headrow = all([cell.name == 'th' for cell in cells])
+        overline = ''
+        underline = ''
+        if is_headrow and not el.previous_sibling:
+            # first row and is headline: print headline underline
+            underline += '| ' + ' | '.join(['---'] * len(cells)) + ' |' + '\n'
+        elif not el.previous_sibling and not el.parent.name != 'table':
+            # first row, not headline, and the parent is sth. like tbody:
+            # print empty headline above this row
+            overline += '| ' + ' | '.join([''] * len(cells)) + ' |' + '\n'
+            overline += '| ' + ' | '.join(['---'] * len(cells)) + ' |' + '\n'
+        return overline + '|' + text + '\n' + underline
+
+    def convert_th(self, el, text, convert_as_inline):
+        return ' ' + text + ' |'
+
+    def convert_td(self, el, text, convert_as_inline):
+        return ' ' + text + ' |'
 
     def convert_hr(self, el, text, convert_as_inline):
         return '\n\n---\n\n'

setup.cfg

@@ -1,2 +1,2 @@
 [flake8]
-ignore = E501
+ignore = E501 W503

setup.py

@@ -10,7 +10,7 @@ read = lambda filepath: codecs.open(filepath, 'r', 'utf-8').read()
 pkgmeta = {
     '__title__': 'markdownify',
     '__author__': 'Matthew Tretter',
-    '__version__': '0.7.2',
+    '__version__': '0.7.4',
 }
 
 

tests/test_conversions.py

@@ -1,5 +1,4 @@
 from markdownify import markdownify as md, ATX, ATX_CLOSED, BACKSLASH, UNDERSCORE
-import re
 
 
 nested_uls = """
@@ -41,8 +40,7 @@ nested_ols = """
     </ul>"""
 
 
-table = re.sub(r'\s+', '', """
-<table>
+table = """<table>
     <tr>
         <th>Firstname</th>
         <th>Lastname</th>
@@ -58,18 +56,54 @@ table = re.sub(r'\s+', '', """
         <td>Jackson</td>
         <td>94</td>
     </tr>
-</table>
-""")
+</table>"""
 
 
-table_head_body = re.sub(r'\s+', '', """
-<table>
+table_with_html_content = """<table>
+    <tr>
+        <th>Firstname</th>
+        <th>Lastname</th>
+        <th>Age</th>
+    </tr>
+    <tr>
+        <td><b>Jill</b></td>
+        <td><i>Smith</i></td>
+        <td><a href="#">50</a></td>
+    </tr>
+    <tr>
+        <td>Eve</td>
+        <td>Jackson</td>
+        <td>94</td>
+    </tr>
+</table>"""
+
+
+table_with_header_column = """<table>
+    <tr>
+        <th>Firstname</th>
+        <th>Lastname</th>
+        <th>Age</th>
+    </tr>
+    <tr>
+        <th>Jill</th>
+        <td>Smith</td>
+        <td>50</td>
+    </tr>
+    <tr>
+        <th>Eve</th>
+        <td>Jackson</td>
+        <td>94</td>
+    </tr>
+</table>"""
+
+
+table_head_body = """<table>
     <thead>
-            <tr>
+        <tr>
             <th>Firstname</th>
             <th>Lastname</th>
             <th>Age</th>
-            </tr>
+        </tr>
     </thead>
     <tbody>
         <tr>
@@ -83,17 +117,15 @@ table_head_body = re.sub(r'\s+', '', """
             <td>94</td>
         </tr>
     </tbody>
-</table>
-""")
+</table>"""
 
-table_missing_text = re.sub(r'\s+', '', """
-<table>
+table_missing_text = """<table>
     <thead>
-            <tr>
+        <tr>
             <th></th>
             <th>Lastname</th>
             <th>Age</th>
-            </tr>
+        </tr>
     </thead>
     <tbody>
         <tr>
@@ -107,8 +139,25 @@ table_missing_text = re.sub(r'\s+', '', """
             <td>94</td>
         </tr>
     </tbody>
-</table>
-""")
+</table>"""
+
+table_missing_head = """<table>
+    <tr>
+        <td>Firstname</td>
+        <td>Lastname</td>
+        <td>Age</td>
+    </tr>
+    <tr>
+        <td>Jill</td>
+        <td>Smith</td>
+        <td>50</td>
+    </tr>
+    <tr>
+        <td>Eve</td>
+        <td>Jackson</td>
+        <td>94</td>
+    </tr>
+</table>"""
 
 
 def test_chomp():
@@ -308,6 +357,10 @@ def test_bullets():
     assert md(nested_uls, bullets='-') == '\n- 1\n\t- a\n\t\t- I\n\t\t- II\n\t\t- III\n\t- b\n\t- 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'
+
+
 def test_img():
     assert md('<img src="/path/to/img.jpg" alt="Alt text" title="Optional title" />') == '![Alt text](/path/to/img.jpg "Optional title")'
     assert md('<img src="/path/to/img.jpg" alt="Alt text" />') == '![Alt text](/path/to/img.jpg)'
@@ -318,9 +371,12 @@ def test_div():
 
 
 def test_table():
-    assert md(table) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |'
-    assert md(table_head_body) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |'
-    assert md(table_missing_text) == '|  | Lastname | Age |\n| --- | --- | --- |\n| Jill |  | 50 |\n| Eve | Jackson | 94 |'
+    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_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_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'
 
 
 def test_strong_em_symbol():