Merge branch 'develop'

fixes #45

markdownify/__init__.py

@@ -1,4 +1,4 @@
-from bs4 import BeautifulSoup, NavigableString, Comment
+from bs4 import BeautifulSoup, NavigableString, Comment, Doctype
 import re
 import six
 
@@ -44,6 +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('_'))
 
@@ -84,23 +100,31 @@ class MarkdownConverter(object):
         if not children_only and isHeading:
             convert_children_as_inline = True
 
-        # Remove whitespace-only textnodes in lists
-        def is_list_node(el):
-            return el and el.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_list_node(node):
+        if is_nested_node(node):
             for el in node.children:
-                # Only extract (remove) whitespace-only text node if any of the conditions is true:
+                # 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 list node
-                can_extract = not el.previous_sibling or not el.next_sibling or is_list_node(el.previous_sibling) or is_list_node(el.next_sibling)
-                if isinstance(el, NavigableString) and six.text_type(el).strip() == '' and can_extract:
+                # - 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):
+            if isinstance(el, Comment) or isinstance(el, Doctype):
                 continue
             elif isinstance(el, NavigableString):
                 text += self.process_text(el)
@@ -116,12 +140,21 @@ class MarkdownConverter(object):
 
     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 escape(all_whitespace_re.sub(' ', text or '')).rstrip()
-        return escape(whitespace_re.sub(' ', text or ''))
+            return cleaned_text.rstrip()
+
+        return cleaned_text
 
     def __getattr__(self, attr):
         # Handle headings
@@ -171,8 +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):
 
@@ -190,12 +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:
@@ -211,8 +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):
 
@@ -259,42 +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):
-        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_hr(self, el, text, convert_as_inline):
-        return '\n\n---\n\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__': '0.7.3',
+    '__version__': '0.8.1',
 }
 
 

tests/test_advanced.py

@@ -14,3 +14,15 @@ 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"
+
+
+def test_special_tags():
+    assert md('<!DOCTYPE html>') == ''
+    assert md('<![CDATA[foobar]]>') == 'foobar'

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():
@@ -191,6 +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'
 
@@ -322,9 +405,12 @@ 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():