Merge branch 'develop'

closes #15

markdownify/__init__.py

@@ -6,6 +6,7 @@ import six
 convert_heading_re = re.compile(r'convert_h(\d+)')
 line_beginning_re = re.compile(r'^', re.MULTILINE)
 whitespace_re = re.compile(r'[\t ]+')
+all_whitespace_re = re.compile(r'[\s]+')
 html_heading_re = re.compile(r'h[1-6]')
 
 
@@ -43,6 +44,22 @@ def chomp(text):
     return (prefix, suffix, text)
 
 
+def abstract_inline_conversion(markup_fn):
+    """
+    This abstracts all simple inline tags like b, em, del, ...
+    Returns a function that wraps the chomped text in a pair of the string
+    that is returned by markup_fn. markup_fn is necessary to allow for
+    references to self.strong_em_symbol etc.
+    """
+    def implementation(self, el, text, convert_as_inline):
+        markup = markup_fn(self)
+        prefix, suffix, text = chomp(text)
+        if not text:
+            return ''
+        return '%s%s%s%s%s' % (prefix, markup, text, markup, suffix)
+    return implementation
+
+
 def _todict(obj):
     return dict((k, getattr(obj, k)) for k in dir(obj) if not k.startswith('_'))
 
@@ -83,12 +100,34 @@ class MarkdownConverter(object):
         if not children_only and isHeading:
             convert_children_as_inline = True
 
+        # 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:
+                # 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
         for el in node.children:
             if isinstance(el, Comment):
                 continue
             elif isinstance(el, NavigableString):
-                text += self.process_text(six.text_type(el))
+                text += self.process_text(el)
             else:
                 text += self.process_tag(el, convert_children_as_inline)
 
@@ -99,8 +138,23 @@ class MarkdownConverter(object):
 
         return text
 
-    def process_text(self, text):
-        return escape(whitespace_re.sub(' ', text or ''))
+    def process_text(self, el):
+        text = six.text_type(el)
+
+        # dont remove any whitespace when handling pre or code in pre
+        if (el.parent.name == 'pre'
+                or (el.parent.name == 'code' and el.parent.parent.name == 'pre')):
+            return escape(text or '')
+
+        cleaned_text = escape(whitespace_re.sub(' ', text or ''))
+
+        # 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 cleaned_text.rstrip()
+
+        return cleaned_text
 
     def __getattr__(self, attr):
         # Handle headings
@@ -150,8 +204,7 @@ class MarkdownConverter(object):
         title_part = ' "%s"' % title.replace('"', r'\"') if title else ''
         return '%s[%s](%s%s)%s' % (prefix, text, href, title_part, suffix) if href else text
 
-    def convert_b(self, el, text, convert_as_inline):
-        return self.convert_strong(el, text, convert_as_inline)
+    convert_b = abstract_inline_conversion(lambda self: 2 * self.options['strong_em_symbol'])
 
     def convert_blockquote(self, el, text, convert_as_inline):
 
@@ -169,12 +222,17 @@ class MarkdownConverter(object):
         else:
             return '  \n'
 
-    def convert_em(self, el, text, convert_as_inline):
-        em_tag = self.options['strong_em_symbol']
-        prefix, suffix, text = chomp(text)
-        if not text:
-            return ''
-        return '%s%s%s%s%s' % (prefix, em_tag, text, em_tag, suffix)
+    def convert_code(self, el, text, convert_as_inline):
+        if el.parent.name == 'pre':
+            return text
+        converter = abstract_inline_conversion(lambda self: '`')
+        return converter(self, el, text, convert_as_inline)
+
+    convert_del = abstract_inline_conversion(lambda self: '~~')
+
+    convert_em = abstract_inline_conversion(lambda self: self.options['strong_em_symbol'])
+
+    convert_kbd = convert_code
 
     def convert_hn(self, n, el, text, convert_as_inline):
         if convert_as_inline:
@@ -190,8 +248,20 @@ class MarkdownConverter(object):
             return '%s %s %s\n\n' % (hashes, text, hashes)
         return '%s %s\n\n' % (hashes, text)
 
-    def convert_i(self, el, text, convert_as_inline):
-        return self.convert_em(el, text, convert_as_inline)
+    def convert_hr(self, el, text, convert_as_inline):
+        return '\n\n---\n\n'
+
+    convert_i = convert_em
+
+    def convert_img(self, el, text, convert_as_inline):
+        alt = el.attrs.get('alt', None) or ''
+        src = el.attrs.get('src', None) or ''
+        title = el.attrs.get('title', None) or ''
+        title_part = ' "%s"' % title.replace('"', r'\"') if title else ''
+        if convert_as_inline:
+            return alt
+
+        return '![%s](%s%s)' % (alt, src, title_part)
 
     def convert_list(self, el, text, convert_as_inline):
 
@@ -199,6 +269,9 @@ class MarkdownConverter(object):
         # Ignoring convert_to_inline for list.
 
         nested = False
+        before_paragraph = False
+        if el.next_sibling and el.next_sibling.name not in ['ul', 'ol']:
+            before_paragraph = True
         while el:
             if el.name == 'li':
                 nested = True
@@ -207,7 +280,7 @@ class MarkdownConverter(object):
         if nested:
             # remove trailing newline if nested
             return '\n' + self.indent(text, 1).rstrip()
-        return '\n' + text + '\n'
+        return text + ('\n' if before_paragraph else '')
 
     convert_ul = convert_list
     convert_ol = convert_list
@@ -235,22 +308,40 @@ class MarkdownConverter(object):
             return text
         return '%s\n\n' % text if text else ''
 
-    def convert_strong(self, el, text, convert_as_inline):
-        strong_tag = 2 * self.options['strong_em_symbol']
-        prefix, suffix, text = chomp(text)
+    def convert_pre(self, el, text, convert_as_inline):
         if not text:
             return ''
-        return '%s%s%s%s%s' % (prefix, strong_tag, text, strong_tag, suffix)
+        return '\n```\n%s\n```\n' % text
 
-    def convert_img(self, el, text, convert_as_inline):
-        alt = el.attrs.get('alt', None) or ''
-        src = el.attrs.get('src', None) or ''
-        title = el.attrs.get('title', None) or ''
-        title_part = ' "%s"' % title.replace('"', r'\"') if title else ''
-        if convert_as_inline:
-            return alt
+    convert_s = convert_del
 
-        return '![%s](%s%s)' % (alt, src, title_part)
+    convert_strong = convert_b
+
+    convert_samp = convert_code
+
+    def convert_table(self, el, text, convert_as_inline):
+        return '\n\n' + text + '\n'
+
+    def convert_td(self, el, text, convert_as_inline):
+        return ' ' + text + ' |'
+
+    def convert_th(self, el, text, convert_as_inline):
+        return ' ' + text + ' |'
+
+    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 markdownify(html, **options):

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__': '0.6.6',
+    '__version__': '0.8.0',
 }
 
 

tests/test_advanced.py

@@ -14,3 +14,10 @@ def test_ignore_comments():
 def test_ignore_comments_with_other_tags():
     text = md("<!-- This is a comment --><a href='http://example.com/'>example link</a>")
     assert text == "[example link](http://example.com/)"
+
+
+def test_code_with_tricky_content():
+    assert md('<code>></code>') == "`>`"
+    assert md('<code>/home/</code><b>username</b>') == "`/home/`**username**"
+    assert md('First line <code>blah blah<br />blah blah</code> second line') \
+        == "First line `blah blah  \nblah blah` second line"

tests/test_conversions.py

@@ -1,8 +1,7 @@
 from markdownify import markdownify as md, ATX, ATX_CLOSED, BACKSLASH, UNDERSCORE
-import re
 
 
-nested_uls = re.sub(r'\s+', '', """
+nested_uls = """
     <ul>
         <li>1
             <ul>
@@ -19,7 +18,146 @@ nested_uls = re.sub(r'\s+', '', """
         </li>
         <li>2</li>
         <li>3</li>
-    </ul>""")
+    </ul>"""
+
+nested_ols = """
+    <ol>
+        <li>1
+            <ol>
+                <li>a
+                    <ol>
+                        <li>I</li>
+                        <li>II</li>
+                        <li>III</li>
+                    </ol>
+                </li>
+                <li>b</li>
+                <li>c</li>
+            </ol>
+        </li>
+        <li>2</li>
+        <li>3</li>
+    </ul>"""
+
+
+table = """<table>
+    <tr>
+        <th>Firstname</th>
+        <th>Lastname</th>
+        <th>Age</th>
+    </tr>
+    <tr>
+        <td>Jill</td>
+        <td>Smith</td>
+        <td>50</td>
+    </tr>
+    <tr>
+        <td>Eve</td>
+        <td>Jackson</td>
+        <td>94</td>
+    </tr>
+</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>
+            <th>Firstname</th>
+            <th>Lastname</th>
+            <th>Age</th>
+        </tr>
+    </thead>
+    <tbody>
+        <tr>
+            <td>Jill</td>
+            <td>Smith</td>
+            <td>50</td>
+        </tr>
+        <tr>
+            <td>Eve</td>
+            <td>Jackson</td>
+            <td>94</td>
+        </tr>
+    </tbody>
+</table>"""
+
+table_missing_text = """<table>
+    <thead>
+        <tr>
+            <th></th>
+            <th>Lastname</th>
+            <th>Age</th>
+        </tr>
+    </thead>
+    <tbody>
+        <tr>
+            <td>Jill</td>
+            <td></td>
+            <td>50</td>
+        </tr>
+        <tr>
+            <td>Eve</td>
+            <td>Jackson</td>
+            <td>94</td>
+        </tr>
+    </tbody>
+</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():
@@ -102,6 +240,40 @@ def test_em_spaces():
     assert md('foo <em></em> bar') == 'foo  bar'
 
 
+def inline_tests(tag, markup):
+    # Basically re-use test_em() and test_em_spaces(),
+    assert md(f'<{tag}>Hello</{tag}>') == f'{markup}Hello{markup}'
+    assert md(f'foo <{tag}>Hello</{tag}> bar') == f'foo {markup}Hello{markup} bar'
+    assert md(f'foo<{tag}> Hello</{tag}> bar') == f'foo {markup}Hello{markup} bar'
+    assert md(f'foo <{tag}>Hello </{tag}>bar') == f'foo {markup}Hello{markup} bar'
+    assert md(f'foo <{tag}></{tag}> bar') in ['foo  bar', 'foo bar']  # Either is OK
+
+
+def test_code():
+    inline_tests('code', '`')
+
+
+def test_samp():
+    inline_tests('samp', '`')
+
+
+def test_kbd():
+    inline_tests('kbd', '`')
+
+
+def test_pre():
+    assert md('<pre>test\n    foo\nbar</pre>') == '\n```\ntest\n    foo\nbar\n```\n'
+    assert md('<pre><code>test\n    foo\nbar</code></pre>') == '\n```\ntest\n    foo\nbar\n```\n'
+
+
+def test_del():
+    inline_tests('del', '~~')
+
+
+def test_s():
+    inline_tests('s', '~~')
+
+
 def test_h1():
     assert md('<h1>Hello</h1>') == 'Hello\n=====\n\n'
 
@@ -159,7 +331,9 @@ def test_hn_nested_img():
 
 
 def test_hr():
-    assert md('<hr>hr</hr>') == '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\n\n---\n\n\nWorld\n\n'
 
 
 def test_head():
@@ -181,8 +355,8 @@ def test_i():
 
 
 def test_ol():
-    assert md('<ol><li>a</li><li>b</li></ol>') == '\n1. a\n2. b\n\n'
-    assert md('<ol start="3"><li>a</li><li>b</li></ol>') == '\n3. a\n4. b\n\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'
 
 
 def test_p():
@@ -194,11 +368,15 @@ def test_strong():
 
 
 def test_ul():
-    assert md('<ul><li>a</li><li>b</li></ul>') == '\n* a\n* b\n\n'
+    assert md('<ul><li>a</li><li>b</li></ul>') == '* a\n* b\n'
+
+
+def test_nested_ols():
+    assert md(nested_ols) == '\n1. 1\n\t1. a\n\t\t1. I\n\t\t2. II\n\t\t3. III\n\t2. b\n\t3. c\n2. 2\n3. 3\n'
 
 
 def test_inline_ul():
-    assert md('<p>foo</p><ul><li>a</li><li>b</li></ul><p>bar</p>') == 'foo\n\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'
 
 
 def test_nested_uls():
@@ -206,11 +384,15 @@ def test_nested_uls():
     Nested ULs should alternate bullet characters.
 
     """
-    assert md(nested_uls) == '\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\n'
+    assert md(nested_uls) == '\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_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\n'
+    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():
@@ -222,6 +404,15 @@ def test_div():
     assert md('Hello</div> World') == 'Hello World'
 
 
+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_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():
     assert md('<strong>Hello</strong>', strong_em_symbol=UNDERSCORE) == '__Hello__'
     assert md('<b>Hello</b>', strong_em_symbol=UNDERSCORE) == '__Hello__'