use static version instead of dynamic git tag info

Co-authored-by: Mico <mico_wu@trendmicro.com>

.github/workflows/python-app.yml

@@ -23,11 +23,10 @@ jobs:
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
-        pip install flake8==3.8.4 pytest
-        if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
-    - name: Lint with flake8
+        pip install --upgrade setuptools setuptools_scm wheel build tox
+    - name: Lint and test
       run: |
-        python setup.py lint
-    - name: Test with pytest
+        tox
+    - name: Build
       run: |
-        python setup.py test
+        python -m build -nwsx .

.github/workflows/python-publish.yml

@@ -21,11 +21,11 @@ jobs:
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
-        pip install setuptools wheel twine
+        pip install --upgrade setuptools setuptools_scm wheel build twine
     - name: Build and publish
       env:
         TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
         TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
       run: |
-        python setup.py sdist bdist_wheel
+        python -m build -nwsx .
         twine upload dist/*

.gitignore

@@ -8,3 +8,5 @@
 /MANIFEST
 /venv
 build/
+.vscode/settings.json
+.tox/

MANIFEST.in

@@ -1 +1,2 @@
 include README.rst
+prune tests

README.rst

@@ -1,8 +1,8 @@
 |build| |version| |license| |downloads|
 
-.. |build| image:: https://img.shields.io/github/workflow/status/matthewwithanm/python-markdownify/Python%20application/develop
+.. |build| image:: https://img.shields.io/github/actions/workflow/status/matthewwithanm/python-markdownify/python-app.yml?branch=develop
     :alt: GitHub Workflow Status
-    :target: https://github.com/matthewwithanm/python-markdownify/actions?query=workflow%3A%22Python+application%22
+    :target: https://github.com/matthewwithanm/python-markdownify/actions/workflows/python-app.yml?query=workflow%3A%22Python+application%22
 
 .. |version| image:: https://img.shields.io/pypi/v/markdownify
     :alt: Pypi version
@@ -32,14 +32,14 @@ Convert some HTML to Markdown:
     from markdownify import markdownify as md
     md('<b>Yay</b> <a href="http://github.com">GitHub</a>')  # > '**Yay** [GitHub](http://github.com)'
 
-Specify tags to exclude (blacklist):
+Specify tags to exclude:
 
 .. code:: python
 
     from markdownify import markdownify as md
     md('<b>Yay</b> <a href="http://github.com">GitHub</a>', strip=['a'])  # > '**Yay** GitHub'
 
-\...or specify the tags you want to include (whitelist):
+\...or specify the tags you want to include:
 
 .. code:: python
 
@@ -53,16 +53,20 @@ Options
 Markdownify supports the following options:
 
 strip
-  A list of tags to strip (blacklist). This option can't be used with the
+  A list of tags to strip. This option can't be used with the
   ``convert`` option.
 
 convert
-  A list of tags to convert (whitelist). This option can't be used with the
+  A list of tags to convert. This option can't be used with the
   ``strip`` option.
 
 autolinks
   A boolean indicating whether the "automatic link" style should be used when
-  a ``a`` tag's contents match its href. Defaults to ``True``
+  a ``a`` tag's contents match its href. Defaults to ``True``.
+
+default_title
+  A boolean to enable setting the title of a link to its href, if no title is
+  given. Defaults to ``False``.
 
 heading_style
   Defines how headings should be converted. Accepted values are ``ATX``,
@@ -80,24 +84,135 @@ strong_em_symbol
   *emphasized* texts. Either of these symbols can be chosen by the options
   ``ASTERISK`` (default) or ``UNDERSCORE`` respectively.
 
+sub_symbol, sup_symbol
+  Define the chars that surround ``<sub>`` and ``<sup>`` text. Defaults to an
+  empty string, because this is non-standard behavior. Could be something like
+  ``~`` and ``^`` to result in ``~sub~`` and ``^sup^``.  If the value starts
+  with ``<`` and ends with ``>``, it is treated as an HTML tag and a ``/`` is
+  inserted after the ``<`` in the string used after the text; this allows
+  specifying ``<sub>`` to use raw HTML in the output for subscripts, for
+  example.
+
 newline_style
   Defines the style of marking linebreaks (``<br>``) in markdown. The default
   value ``SPACES`` of this option will adopt the usual two spaces and a newline,
-  while ``BACKSLASH`` will convert a linebreak to ``\\n`` (a backslash an a
+  while ``BACKSLASH`` will convert a linebreak to ``\\n`` (a backslash and a
   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.
+
+code_language_callback
+  When the HTML code contains ``pre`` tags that in some way provide the code
+  language, for example as class, this callback can be used to extract the
+  language from the tag and prefix it to the converted ``pre`` tag.
+  The callback gets one single argument, an BeautifylSoup object, and returns
+  a string containing the code language, or ``None``.
+  An example to use the class name as code language could be::
+
+    def callback(el):
+        return el['class'][0] if el.has_attr('class') else None
+
+  Defaults to ``None``.
+
+escape_asterisks
+  If set to ``False``, do not escape ``*`` to ``\*`` in text.
+  Defaults to ``True``.
+
+escape_underscores
+  If set to ``False``, do not escape ``_`` to ``\_`` in text.
+  Defaults to ``True``.
+
+escape_misc
+  If set to ``False``, do not escape miscellaneous punctuation characters
+  that sometimes have Markdown significance in text.
+  Defaults to ``True``.
+
+keep_inline_images_in
+  Images are converted to their alt-text when the images are located inside
+  headlines or table cells. If some inline images should be converted to
+  markdown images instead, this option can be set to a list of parent tags
+  that should be allowed to contain inline images, for example ``['td']``.
+  Defaults to an empty list.
+
+wrap, wrap_width
+  If ``wrap`` is set to ``True``, all text paragraphs are wrapped at
+  ``wrap_width`` characters. Defaults to ``False`` and ``80``.
+  Use with ``newline_style=BACKSLASH`` to keep line breaks in paragraphs.
+
 Options may be specified as kwargs to the ``markdownify`` function, or as a
 nested ``Options`` class in ``MarkdownConverter`` subclasses.
 
 
+Converting BeautifulSoup objects
+================================
+
+.. code:: python
+
+    from markdownify import MarkdownConverter
+
+    # Create shorthand method for conversion
+    def md(soup, **options):
+        return MarkdownConverter(**options).convert_soup(soup)
+
+
+Creating Custom Converters
+==========================
+
+If you have a special usecase that calls for a special conversion, you can
+always inherit from ``MarkdownConverter`` and override the method you want to
+change.
+The function that handles a HTML tag named ``abc`` is called
+``convert_abc(self, el, text, convert_as_inline)`` and returns a string
+containing the converted HTML tag.
+The ``MarkdownConverter`` object will handle the conversion based on the
+function names:
+
+.. code:: python
+
+    from markdownify import MarkdownConverter
+
+    class ImageBlockConverter(MarkdownConverter):
+        """
+        Create a custom MarkdownConverter that adds two newlines after an image
+        """
+        def convert_img(self, el, text, convert_as_inline):
+            return super().convert_img(el, text, convert_as_inline) + '\n\n'
+
+    # Create shorthand method for conversion
+    def md(html, **options):
+        return ImageBlockConverter(**options).convert(html)
+
+.. code:: python
+
+    from markdownify import MarkdownConverter
+
+    class IgnoreParagraphsConverter(MarkdownConverter):
+        """
+        Create a custom MarkdownConverter that ignores paragraphs
+        """
+        def convert_p(self, el, text, convert_as_inline):
+            return ''
+
+    # Create shorthand method for conversion
+    def md(html, **options):
+        return IgnoreParagraphsConverter(**options).convert(html)
+
+
+Command Line Interface
+======================
+
+Use ``markdownify example.html > example.md`` or pipe input from stdin
+(``cat example.html | markdownify > example.md``).
+Call ``markdownify -h`` to see all available options.
+They are the same as listed above and take the same arguments.
+
+
 Development
 ===========
 
-To run tests:
-
-``python setup.py test``
-
-To lint:
-
-``python setup.py lint``
+To run tests and the linter run ``pip install tox`` once, then ``tox``.

markdownify/__init__.py

@@ -1,4 +1,5 @@
-from bs4 import BeautifulSoup, NavigableString, Comment
+from bs4 import BeautifulSoup, NavigableString, Comment, Doctype
+from textwrap import fill
 import re
 import six
 
@@ -25,12 +26,6 @@ ASTERISK = '*'
 UNDERSCORE = '_'
 
 
-def escape(text):
-    if not 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
@@ -44,19 +39,53 @@ 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, with '/' inserted in the string used after
+    the text if it looks like an HTML tag. markup_fn is necessary to allow for
+    references to self.strong_em_symbol etc.
+    """
+    def implementation(self, el, text, convert_as_inline):
+        markup_prefix = markup_fn(self)
+        if markup_prefix.startswith('<') and markup_prefix.endswith('>'):
+            markup_suffix = '</' + markup_prefix[1:]
+        else:
+            markup_suffix = markup_prefix
+        if el.find_parent(['pre', 'code', 'kbd', 'samp']):
+            return text
+        prefix, suffix, text = chomp(text)
+        if not text:
+            return ''
+        return '%s%s%s%s%s' % (prefix, markup_prefix, text, markup_suffix, suffix)
+    return implementation
+
+
 def _todict(obj):
     return dict((k, getattr(obj, k)) for k in dir(obj) if not k.startswith('_'))
 
 
 class MarkdownConverter(object):
     class DefaultOptions:
-        strip = None
-        convert = None
         autolinks = True
-        heading_style = UNDERLINED
         bullets = '*+-'  # An iterable of bullet types.
-        strong_em_symbol = ASTERISK
+        code_language = ''
+        code_language_callback = None
+        convert = None
+        default_title = False
+        escape_asterisks = True
+        escape_underscores = True
+        escape_misc = True
+        heading_style = UNDERLINED
+        keep_inline_images_in = []
         newline_style = SPACES
+        strip = None
+        strong_em_symbol = ASTERISK
+        sub_symbol = ''
+        sup_symbol = ''
+        wrap = False
+        wrap_width = 80
 
     class Options(DefaultOptions):
         pass
@@ -73,26 +102,48 @@ class MarkdownConverter(object):
 
     def convert(self, html):
         soup = BeautifulSoup(html, 'html.parser')
+        return self.convert_soup(soup)
+
+    def convert_soup(self, soup):
         return self.process_tag(soup, convert_as_inline=False, children_only=True)
 
     def process_tag(self, node, convert_as_inline, children_only=False):
         text = ''
-        # markdown headings can't include block elements (elements w/newlines)
+
+        # markdown headings or cells can't include
+        # block elements (elements w/newlines)
         isHeading = html_heading_re.match(node.name) is not None
+        isCell = node.name in ['td', 'th']
         convert_children_as_inline = convert_as_inline
 
-        if not children_only and isHeading:
+        if not children_only and (isHeading or isCell):
             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
         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)
@@ -107,10 +158,25 @@ class MarkdownConverter(object):
         return text
 
     def process_text(self, el):
-        text = six.text_type(el)
-        if el.parent.name == 'li':
-            return escape(all_whitespace_re.sub(' ', text or '')).rstrip()
-        return escape(whitespace_re.sub(' ', text or ''))
+        text = six.text_type(el) or ''
+
+        # normalize whitespace if we're not inside a preformatted element
+        if not el.find_parent('pre'):
+            text = whitespace_re.sub(' ', text)
+
+        # escape special characters if we're not inside a preformatted or code element
+        if not el.find_parent(['pre', 'code', 'kbd', 'samp']):
+            text = self.escape(text)
+
+        # 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'])):
+            text = text.rstrip()
+
+        return text
 
     def __getattr__(self, attr):
         # Handle headings
@@ -138,6 +204,18 @@ class MarkdownConverter(object):
         else:
             return True
 
+    def escape(self, text):
+        if not text:
+            return ''
+        if self.options['escape_misc']:
+            text = re.sub(r'([\\&<`[>~#=+|-])', r'\\\1', text)
+            text = re.sub(r'([0-9])([.)])', r'\1\\\2', text)
+        if self.options['escape_asterisks']:
+            text = text.replace('*', r'\*')
+        if self.options['escape_underscores']:
+            text = text.replace('_', r'\_')
+        return text
+
     def indent(self, text, level):
         return line_beginning_re.sub('\t' * level, text) if text else ''
 
@@ -149,26 +227,28 @@ class MarkdownConverter(object):
         prefix, suffix, text = chomp(text)
         if not text:
             return ''
-        if convert_as_inline:
-            return text
         href = el.get('href')
         title = el.get('title')
         # For the replacement see #29: text nodes underscores are escaped
-        if self.options['autolinks'] and text.replace(r'\_', '_') == href and not title:
+        if (self.options['autolinks']
+                and text.replace(r'\_', '_') == href
+                and not title
+                and not self.options['default_title']):
             # Shortcut syntax
             return '<%s>' % href
+        if self.options['default_title'] and not title:
+            title = href
         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):
 
         if convert_as_inline:
             return text
 
-        return '\n' + (line_beginning_re.sub('> ', text) + '\n\n') if text else ''
+        return '\n' + (line_beginning_re.sub('> ', text.strip()) + '\n\n') if text else ''
 
     def convert_br(self, el, text, convert_as_inline):
         if convert_as_inline:
@@ -179,19 +259,24 @@ 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:
             return text
 
         style = self.options['heading_style'].lower()
-        text = text.rstrip()
+        text = text.strip()
         if style == UNDERLINED and n <= 2:
             line = '=' if n == 1 else '-'
             return self.underline(text, line)
@@ -200,8 +285,21 @@ 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
+                and el.parent.name not in self.options['keep_inline_images_in']):
+            return alt
+
+        return '![%s](%s%s)' % (alt, src, title_part)
 
     def convert_list(self, el, text, convert_as_inline):
 
@@ -228,7 +326,7 @@ class MarkdownConverter(object):
     def convert_li(self, el, text, convert_as_inline):
         parent = el.parent
         if parent is not None and parent.name == 'ol':
-            if parent.get("start"):
+            if parent.get("start") and str(parent.get("start")).isnumeric():
                 start = int(parent.get("start"))
             else:
                 start = 1
@@ -241,49 +339,94 @@ 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:
             return text
+        if self.options['wrap']:
+            text = fill(text,
+                        width=self.options['wrap_width'],
+                        break_long_words=False,
+                        break_on_hyphens=False)
         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)
+        code_language = self.options['code_language']
 
-    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
+        if self.options['code_language_callback']:
+            code_language = self.options['code_language_callback'](el) or code_language
 
-        return '![%s](%s%s)' % (alt, src, title_part)
+        return '\n```%s\n%s\n```\n' % (code_language, text)
+
+    def convert_script(self, el, text, convert_as_inline):
+        return ''
+
+    def convert_style(self, el, text, convert_as_inline):
+        return ''
+
+    convert_s = convert_del
+
+    convert_strong = convert_b
+
+    convert_samp = convert_code
+
+    convert_sub = abstract_inline_conversion(lambda self: self.options['sub_symbol'])
+
+    convert_sup = abstract_inline_conversion(lambda self: self.options['sup_symbol'])
 
     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_caption(self, el, text, convert_as_inline):
+        return text + '\n'
+
+    def convert_figcaption(self, el, text, convert_as_inline):
+        return '\n\n' + text + '\n\n'
+
+    def convert_td(self, el, text, convert_as_inline):
+        colspan = 1
+        if 'colspan' in el.attrs and el['colspan'].isdigit():
+            colspan = int(el['colspan'])
+        return ' ' + text.strip().replace("\n", " ") + ' |' * colspan
+
+    def convert_th(self, el, text, convert_as_inline):
+        colspan = 1
+        if 'colspan' in el.attrs and el['colspan'].isdigit():
+            colspan = int(el['colspan'])
+        return ' ' + text.strip().replace("\n", " ") + ' |' * colspan
+
+    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])
+            or (not el.previous_sibling and not el.parent.name == 'tbody')
+            or (not el.previous_sibling and el.parent.name == 'tbody' and len(el.parent.parent.find_all(['thead'])) < 1)
+        )
+        overline = ''
+        underline = ''
+        if is_headrow and not el.previous_sibling:
+            # first row and is headline: print headline underline
+            full_colspan = 0
+            for cell in cells:
+                if 'colspan' in cell.attrs and cell['colspan'].isdigit():
+                    full_colspan += int(cell["colspan"])
+                else:
+                    full_colspan += 1
+            underline += '| ' + ' | '.join(['---'] * full_colspan) + ' |' + '\n'
+        elif (not el.previous_sibling
+              and (el.parent.name == 'table'
+                   or (el.parent.name == 'tbody'
+                       and not el.parent.previous_sibling))):
+            # first row, not headline, and:
+            # - the parent is table or
+            # - the parent is tbody at the beginning of a table.
+            # 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):

markdownify/main.py

@@ -0,0 +1,73 @@
+#!/usr/bin/env python
+
+import argparse
+import sys
+
+from markdownify import markdownify, ATX, ATX_CLOSED, UNDERLINED, \
+    SPACES, BACKSLASH, ASTERISK, UNDERSCORE
+
+
+def main(argv=sys.argv[1:]):
+    parser = argparse.ArgumentParser(
+        prog='markdownify',
+        description='Converts html to markdown.',
+    )
+
+    parser.add_argument('html', nargs='?', type=argparse.FileType('r'),
+                        default=sys.stdin,
+                        help="The html file to convert. Defaults to STDIN if not "
+                        "provided.")
+    parser.add_argument('-s', '--strip', nargs='*',
+                        help="A list of tags to strip. This option can't be used with "
+                        "the --convert option.")
+    parser.add_argument('-c', '--convert', nargs='*',
+                        help="A list of tags to convert. This option can't be used with "
+                        "the --strip option.")
+    parser.add_argument('-a', '--autolinks', action='store_true',
+                        help="A boolean indicating whether the 'automatic link' style "
+                        "should be used when a 'a' tag's contents match its href.")
+    parser.add_argument('--default-title', action='store_false',
+                        help="A boolean to enable setting the title of a link to its "
+                        "href, if no title is given.")
+    parser.add_argument('--heading-style', default=UNDERLINED,
+                        choices=(ATX, ATX_CLOSED, UNDERLINED),
+                        help="Defines how headings should be converted.")
+    parser.add_argument('-b', '--bullets', default='*+-',
+                        help="A string of bullet styles to use; the bullet will "
+                        "alternate based on nesting level.")
+    parser.add_argument('--strong-em-symbol', default=ASTERISK,
+                        choices=(ASTERISK, UNDERSCORE),
+                        help="Use * or _ to convert strong and italics text"),
+    parser.add_argument('--sub-symbol', default='',
+                        help="Define the chars that surround '<sub>'.")
+    parser.add_argument('--sup-symbol', default='',
+                        help="Define the chars that surround '<sup>'.")
+    parser.add_argument('--newline-style', default=SPACES,
+                        choices=(SPACES, BACKSLASH),
+                        help="Defines the style of <br> conversions: two spaces "
+                        "or backslash at the and of the line thet should break.")
+    parser.add_argument('--code-language', default='',
+                        help="Defines the language that should be assumed for all "
+                        "'<pre>' sections.")
+    parser.add_argument('--no-escape-asterisks', dest='escape_asterisks',
+                        action='store_false',
+                        help="Do not escape '*' to '\\*' in text.")
+    parser.add_argument('--no-escape-underscores', dest='escape_underscores',
+                        action='store_false',
+                        help="Do not escape '_' to '\\_' in text.")
+    parser.add_argument('-i', '--keep-inline-images-in', nargs='*',
+                        help="Images are converted to their alt-text when the images are "
+                        "located inside headlines or table cells. If some inline images "
+                        "should be converted to markdown images instead, this option can "
+                        "be set to a list of parent tags that should be allowed to "
+                        "contain inline images.")
+    parser.add_argument('-w', '--wrap', action='store_true',
+                        help="Wrap all text paragraphs at --wrap-width characters.")
+    parser.add_argument('--wrap-width', type=int, default=80)
+
+    args = parser.parse_args(argv)
+    print(markdownify(**vars(args)))
+
+
+if __name__ == '__main__':
+    main()

pyproject.toml

@@ -0,0 +1,45 @@
+[build-system]
+requires = ["setuptools>=61.2", "setuptools_scm[toml]>=3.4.3"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "markdownify"
+version = "0.13.0"
+authors = [{name = "Matthew Tretter", email = "m@tthewwithanm.com"}]
+description = "Convert HTML to markdown."
+readme = "README.rst"
+classifiers = [
+    "Environment :: Web Environment",
+    "Framework :: Django",
+    "Intended Audience :: Developers",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 2.5",
+    "Programming Language :: Python :: 2.6",
+    "Programming Language :: Python :: 2.7",
+    "Programming Language :: Python :: 3.6",
+    "Programming Language :: Python :: 3.7",
+    "Programming Language :: Python :: 3.8",
+    "Topic :: Utilities",
+]
+dependencies = [
+    "beautifulsoup4>=4.9,<5",
+    "six>=1.15,<2"
+]
+
+[project.urls]
+Homepage = "http://github.com/matthewwithanm/python-markdownify"
+Download = "http://github.com/matthewwithanm/python-markdownify/tarball/master"
+
+[project.scripts]
+markdownify = "markdownify.main:main"
+
+[tool.setuptools]
+zip-safe = false
+include-package-data = true
+
+[tool.setuptools.packages.find]
+include = ["markdownify", "markdownify.*"]
+namespaces = false
+
+[tool.setuptools_scm]

setup.cfg

@@ -1,2 +0,0 @@
-[flake8]
-ignore = E501

setup.py

@@ -1,99 +0,0 @@
-#/usr/bin/env python
-import codecs
-import os
-from setuptools import setup, find_packages
-from setuptools.command.test import test as TestCommand, Command
-
-
-read = lambda filepath: codecs.open(filepath, 'r', 'utf-8').read()
-
-pkgmeta = {
-    '__title__': 'markdownify',
-    '__author__': 'Matthew Tretter',
-    '__version__': '0.7.2',
-}
-
-
-class PyTest(TestCommand):
-    def finalize_options(self):
-        TestCommand.finalize_options(self)
-        self.test_args = ['tests', '-s']
-        self.test_suite = True
-
-    def run_tests(self):
-        import pytest
-        errno = pytest.main(self.test_args)
-        raise SystemExit(errno)
-
-
-class LintCommand(Command):
-    """
-    A copy of flake8's Flake8Command
-
-    """
-    description = "Run flake8 on modules registered in setuptools"
-    user_options = []
-
-    def initialize_options(self):
-        pass
-
-    def finalize_options(self):
-        pass
-
-    def distribution_files(self):
-        if self.distribution.packages:
-            for package in self.distribution.packages:
-                yield package.replace(".", os.path.sep)
-
-        if self.distribution.py_modules:
-            for filename in self.distribution.py_modules:
-                yield "%s.py" % filename
-
-    def run(self):
-        from flake8.api.legacy import get_style_guide
-        flake8_style = get_style_guide(config_file='setup.cfg')
-        paths = self.distribution_files()
-        report = flake8_style.check_files(paths)
-        raise SystemExit(report.total_errors > 0)
-
-
-setup(
-    name='markdownify',
-    description='Convert HTML to markdown.',
-    long_description=read(os.path.join(os.path.dirname(__file__), 'README.rst')),
-    version=pkgmeta['__version__'],
-    author=pkgmeta['__author__'],
-    author_email='m@tthewwithanm.com',
-    url='http://github.com/matthewwithanm/python-markdownify',
-    download_url='http://github.com/matthewwithanm/python-markdownify/tarball/master',
-    packages=find_packages(),
-    zip_safe=False,
-    include_package_data=True,
-    setup_requires=[
-        'flake8>=3.8,<4',
-    ],
-    tests_require=[
-        'pytest>=6.2,<7',
-    ],
-    install_requires=[
-        'beautifulsoup4>=4.9,<5', 'six>=1.15,<2'
-    ],
-    classifiers=[
-        'Environment :: Web Environment',
-        'Framework :: Django',
-        'Intended Audience :: Developers',
-        'License :: OSI Approved :: MIT License',
-        'Operating System :: OS Independent',
-        'Programming Language :: Python :: 2.5',
-        'Programming Language :: Python :: 2.6',
-        'Programming Language :: Python :: 2.7',
-        'Programming Language :: Python :: 3.6',
-        'Programming Language :: Python :: 3.7',
-        'Programming Language :: Python :: 3.8',
-        'Topic :: Utilities'
-    ],
-    cmdclass={
-        'test': PyTest,
-        'lint': LintCommand,
-    },
-)

shell.nix

@@ -0,0 +1,10 @@
+{ pkgs ? import <nixpkgs> {} }:
+pkgs.mkShell {
+  name = "python-shell";
+  buildInputs = with pkgs; [
+    python38
+    python38Packages.tox
+    python38Packages.setuptools
+    python38Packages.virtualenv
+  ];
+}

tests/test_advanced.py

@@ -1,6 +1,17 @@
 from markdownify import markdownify as md
 
 
+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_nested():
     text = md('<p>This is an <a href="http://example.com/">example link</a>.</p>')
     assert text == 'This is an [example link](http://example.com/).\n\n'
@@ -14,3 +25,15 @@ 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

@@ -1,130 +1,17 @@
 from markdownify import markdownify as md, ATX, ATX_CLOSED, BACKSLASH, UNDERSCORE
-import re
 
 
-nested_uls = """
-    <ul>
-        <li>1
-            <ul>
-                <li>a
-                    <ul>
-                        <li>I</li>
-                        <li>II</li>
-                        <li>III</li>
-                    </ul>
-                </li>
-                <li>b</li>
-                <li>c</li>
-            </ul>
-        </li>
-        <li>2</li>
-        <li>3</li>
-    </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 = re.sub(r'\s+', '', """
-<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_head_body = re.sub(r'\s+', '', """
-<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 = re.sub(r'\s+', '', """
-<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>
-""")
-
-
-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 inline_tests(tag, markup):
+    # test template for different inline tags
+    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_a():
     assert md('<a href="https://google.com">Google</a>') == '[Google](https://google.com)'
-    assert md('<a href="https://google.com">https://google.com</a>', autolinks=False) == '[https://google.com](https://google.com)'
     assert md('<a href="https://google.com">https://google.com</a>') == '<https://google.com>'
     assert md('<a href="https://community.kde.org/Get_Involved">https://community.kde.org/Get_Involved</a>') == '<https://community.kde.org/Get_Involved>'
     assert md('<a href="https://community.kde.org/Get_Involved">https://community.kde.org/Get_Involved</a>', autolinks=False) == '[https://community.kde.org/Get\\_Involved](https://community.kde.org/Get_Involved)'
@@ -140,6 +27,7 @@ def test_a_spaces():
 def test_a_with_title():
     text = md('<a href="http://google.com" title="The &quot;Goog&quot;">Google</a>')
     assert text == r'[Google](http://google.com "The \"Goog\"")'
+    assert md('<a href="https://google.com">https://google.com</a>', default_title=True) == '[https://google.com](https://google.com "https://google.com")'
 
 
 def test_a_shortcut():
@@ -148,8 +36,7 @@ def test_a_shortcut():
 
 
 def test_a_no_autolinks():
-    text = md('<a href="http://google.com">http://google.com</a>', autolinks=False)
-    assert text == '[http://google.com](http://google.com)'
+    assert md('<a href="https://google.com">https://google.com</a>', autolinks=False) == '[https://google.com](https://google.com)'
 
 
 def test_b():
@@ -165,30 +52,71 @@ def test_b_spaces():
 
 def test_blockquote():
     assert md('<blockquote>Hello</blockquote>') == '\n> Hello\n\n'
+    assert md('<blockquote>\nHello\n</blockquote>') == '\n> Hello\n\n'
+
+
+def test_blockquote_with_nested_paragraph():
+    assert md('<blockquote><p>Hello</p></blockquote>') == '\n> Hello\n\n'
+    assert md('<blockquote><p>Hello</p><p>Hello again</p></blockquote>') == '\n> Hello\n> \n> Hello again\n\n'
 
 
 def test_blockquote_with_paragraph():
     assert md('<blockquote>Hello</blockquote><p>handsome</p>') == '\n> Hello\n\nhandsome\n\n'
 
 
-def test_nested_blockquote():
+def test_blockquote_nested():
     text = md('<blockquote>And she was like <blockquote>Hello</blockquote></blockquote>')
-    assert text == '\n> And she was like \n> > Hello\n> \n> \n\n'
+    assert text == '\n> And she was like \n> > Hello\n\n'
 
 
 def test_br():
     assert md('a<br />b<br />c') == 'a  \nb  \nc'
+    assert md('a<br />b<br />c', newline_style=BACKSLASH) == 'a\\\nb\\\nc'
+
+
+def test_caption():
+    assert md('TEXT<figure><figcaption>Caption</figcaption><span>SPAN</span></figure>') == 'TEXT\n\nCaption\n\nSPAN'
+    assert md('<figure><span>SPAN</span><figcaption>Caption</figcaption></figure>TEXT') == 'SPAN\n\nCaption\n\nTEXT'
+
+
+def test_code():
+    inline_tests('code', '`')
+    assert md('<code>*this_should_not_escape*</code>') == '`*this_should_not_escape*`'
+    assert md('<kbd>*this_should_not_escape*</kbd>') == '`*this_should_not_escape*`'
+    assert md('<samp>*this_should_not_escape*</samp>') == '`*this_should_not_escape*`'
+    assert md('<code><span>*this_should_not_escape*</span></code>') == '`*this_should_not_escape*`'
+    assert md('<code>this  should\t\tnormalize</code>') == '`this should normalize`'
+    assert md('<code><span>this  should\t\tnormalize</span></code>') == '`this should normalize`'
+    assert md('<code>foo<b>bar</b>baz</code>') == '`foobarbaz`'
+    assert md('<kbd>foo<i>bar</i>baz</kbd>') == '`foobarbaz`'
+    assert md('<samp>foo<del> bar </del>baz</samp>') == '`foo bar baz`'
+    assert md('<samp>foo <del>bar</del> baz</samp>') == '`foo bar baz`'
+    assert md('<code>foo<em> bar </em>baz</code>') == '`foo bar baz`'
+    assert md('<code>foo<code> bar </code>baz</code>') == '`foo bar baz`'
+    assert md('<code>foo<strong> bar </strong>baz</code>') == '`foo bar baz`'
+    assert md('<code>foo<s> bar </s>baz</code>') == '`foo bar baz`'
+    assert md('<code>foo<sup>bar</sup>baz</code>', sup_symbol='^') == '`foobarbaz`'
+    assert md('<code>foo<sub>bar</sub>baz</code>', sub_symbol='^') == '`foobarbaz`'
+
+
+def test_del():
+    inline_tests('del', '~~')
+
+
+def test_div():
+    assert md('Hello</div> World') == 'Hello World'
 
 
 def test_em():
-    assert md('<em>Hello</em>') == '*Hello*'
+    inline_tests('em', '*')
 
 
-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_header_with_space():
+    assert md('<h3>\n\nHello</h3>') == '### Hello\n\n'
+    assert md('<h4>\n\nHello</h4>') == '#### Hello\n\n'
+    assert md('<h5>\n\nHello</h5>') == '##### Hello\n\n'
+    assert md('<h5>\n\nHello\n\n</h5>') == '##### Hello\n\n'
+    assert md('<h5>\n\nHello   \n\n</h5>') == '##### Hello\n\n'
 
 
 def test_h1():
@@ -201,6 +129,8 @@ def test_h2():
 
 def test_hn():
     assert md('<h3>Hello</h3>') == '### Hello\n\n'
+    assert md('<h4>Hello</h4>') == '#### Hello\n\n'
+    assert md('<h5>Hello</h5>') == '##### Hello\n\n'
     assert md('<h6>Hello</h6>') == '###### Hello\n\n'
 
 
@@ -236,15 +166,28 @@ def test_hn_nested_simple_tag():
 
 
 def test_hn_nested_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)'
     image_attributes_to_markdown = [
-        ("", ""),
-        ("alt='Alt Text'", "Alt Text"),
-        ("alt='Alt Text' title='Optional title'", "Alt Text"),
+        ("", "", ""),
+        ("alt='Alt Text'", "Alt Text", ""),
+        ("alt='Alt Text' title='Optional title'", "Alt Text", " \"Optional title\""),
     ]
-    for image_attributes, markdown in image_attributes_to_markdown:
-        assert md('<h3>A <img src="/path/to/img.jpg " ' + image_attributes + '/> B</h3>') == '### A ' + markdown + ' B\n\n'
+    for image_attributes, markdown, title in image_attributes_to_markdown:
+        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>') == '### A ' + markdown + ' B\n\n'
+        assert md('<h3>A <img src="/path/to/img.jpg" ' + image_attributes + '/> B</h3>', keep_inline_images_in=['h3']) == '### A ![' + markdown + '](/path/to/img.jpg' + title + ') B\n\n'
+
+
+def test_hn_atx_headings():
+    assert md('<h1>Hello</h1>', heading_style=ATX) == '# Hello\n\n'
+    assert md('<h2>Hello</h2>', heading_style=ATX) == '## Hello\n\n'
+
+
+def test_hn_atx_closed_headings():
+    assert md('<h1>Hello</h1>', heading_style=ATX_CLOSED) == '# Hello #\n\n'
+    assert md('<h2>Hello</h2>', heading_style=ATX_CLOSED) == '## Hello ##\n\n'
+
+
+def test_head():
+    assert md('<head>head</head>') == 'head'
 
 
 def test_hr():
@@ -253,74 +196,66 @@ def test_hr():
     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():
-    assert md('<head>head</head>') == 'head'
-
-
-def test_atx_headings():
-    assert md('<h1>Hello</h1>', heading_style=ATX) == '# Hello\n\n'
-    assert md('<h2>Hello</h2>', heading_style=ATX) == '## Hello\n\n'
-
-
-def test_atx_closed_headings():
-    assert md('<h1>Hello</h1>', heading_style=ATX_CLOSED) == '# Hello #\n\n'
-    assert md('<h2>Hello</h2>', heading_style=ATX_CLOSED) == '## Hello ##\n\n'
-
-
 def test_i():
     assert md('<i>Hello</i>') == '*Hello*'
 
 
-def test_ol():
-    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():
-    assert md('<p>hello</p>') == 'hello\n\n'
-
-
-def test_strong():
-    assert md('<strong>Hello</strong>') == '**Hello**'
-
-
-def test_ul():
-    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* a\n* b\n\nbar\n\n'
-
-
-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'
-
-
-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_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)'
 
 
-def test_div():
-    assert md('Hello</div> World') == 'Hello World'
+def test_kbd():
+    inline_tests('kbd', '`')
 
 
-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 |'
+def test_p():
+    assert md('<p>hello</p>') == 'hello\n\n'
+    assert md('<p>123456789 123456789</p>') == '123456789 123456789\n\n'
+    assert md('<p>123456789 123456789</p>', wrap=True, wrap_width=10) == '123456789\n123456789\n\n'
+    assert md('<p><a href="https://example.com">Some long link</a></p>', wrap=True, wrap_width=10) == '[Some long\nlink](https://example.com)\n\n'
+    assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '12345\\\n67890\n\n'
+    assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '12345678901\\\n12345\n\n'
+
+
+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'
+    assert md('<pre>*this_should_not_escape*</pre>') == '\n```\n*this_should_not_escape*\n```\n'
+    assert md('<pre><span>*this_should_not_escape*</span></pre>') == '\n```\n*this_should_not_escape*\n```\n'
+    assert md('<pre>\t\tthis  should\t\tnot  normalize</pre>') == '\n```\n\t\tthis  should\t\tnot  normalize\n```\n'
+    assert md('<pre><span>\t\tthis  should\t\tnot  normalize</span></pre>') == '\n```\n\t\tthis  should\t\tnot  normalize\n```\n'
+    assert md('<pre>foo<b>\nbar\n</b>baz</pre>') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo<i>\nbar\n</i>baz</pre>') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo\n<i>bar</i>\nbaz</pre>') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo<i>\n</i>baz</pre>') == '\n```\nfoo\nbaz\n```\n'
+    assert md('<pre>foo<del>\nbar\n</del>baz</pre>') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo<em>\nbar\n</em>baz</pre>') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo<code>\nbar\n</code>baz</pre>') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo<strong>\nbar\n</strong>baz</pre>') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo<s>\nbar\n</s>baz</pre>') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo<sup>\nbar\n</sup>baz</pre>', sup_symbol='^') == '\n```\nfoo\nbar\nbaz\n```\n'
+    assert md('<pre>foo<sub>\nbar\n</sub>baz</pre>', sub_symbol='^') == '\n```\nfoo\nbar\nbaz\n```\n'
+
+
+def test_script():
+    assert md('foo <script>var foo=42;</script> bar') == 'foo  bar'
+
+
+def test_style():
+    assert md('foo <style>h1 { font-size: larger }</style> bar') == 'foo  bar'
+
+
+def test_s():
+    inline_tests('s', '~~')
+
+
+def test_samp():
+    inline_tests('samp', '`')
+
+
+def test_strong():
+    assert md('<strong>Hello</strong>') == '**Hello**'
 
 
 def test_strong_em_symbol():
@@ -330,5 +265,27 @@ def test_strong_em_symbol():
     assert md('<i>Hello</i>', strong_em_symbol=UNDERSCORE) == '_Hello_'
 
 
-def test_newline_style():
-    assert md('a<br />b<br />c', newline_style=BACKSLASH) == 'a\\\nb\\\nc'
+def test_sub():
+    assert md('<sub>foo</sub>') == 'foo'
+    assert md('<sub>foo</sub>', sub_symbol='~') == '~foo~'
+    assert md('<sub>foo</sub>', sub_symbol='<sub>') == '<sub>foo</sub>'
+
+
+def test_sup():
+    assert md('<sup>foo</sup>') == 'foo'
+    assert md('<sup>foo</sup>', sup_symbol='^') == '^foo^'
+    assert md('<sup>foo</sup>', sup_symbol='<sup>') == '<sup>foo</sup>'
+
+
+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'
+
+
+def test_lang_callback():
+    def callback(el):
+        return el['class'][0] if el.has_attr('class') else None
+
+    assert md('<pre class="python">test\n    foo\nbar</pre>', code_language_callback=callback) == '\n```python\ntest\n    foo\nbar\n```\n'
+    assert md('<pre class="javascript"><code>test\n    foo\nbar</code></pre>', code_language_callback=callback) == '\n```javascript\ntest\n    foo\nbar\n```\n'
+    assert md('<pre class="javascript"><code class="javascript">test\n    foo\nbar</code></pre>', code_language_callback=callback) == '\n```javascript\ntest\n    foo\nbar\n```\n'

tests/test_custom_converter.py

@@ -0,0 +1,25 @@
+from markdownify import MarkdownConverter
+from bs4 import BeautifulSoup
+
+
+class ImageBlockConverter(MarkdownConverter):
+    """
+    Create a custom MarkdownConverter that adds two newlines after an image
+    """
+    def convert_img(self, el, text, convert_as_inline):
+        return super().convert_img(el, text, convert_as_inline) + '\n\n'
+
+
+def test_img():
+    # Create shorthand method for conversion
+    def md(html, **options):
+        return ImageBlockConverter(**options).convert(html)
+
+    assert md('<img src="/path/to/img.jpg" alt="Alt text" title="Optional title" />') == '![Alt text](/path/to/img.jpg "Optional title")\n\n'
+    assert md('<img src="/path/to/img.jpg" alt="Alt text" />') == '![Alt text](/path/to/img.jpg)\n\n'
+
+
+def test_soup():
+    html = '<b>test</b>'
+    soup = BeautifulSoup(html, 'html.parser')
+    assert MarkdownConverter().convert_soup(soup) == '**test**'

tests/test_escaping.py

@@ -1,12 +1,18 @@
 from markdownify import markdownify as md
 
 
+def test_asterisks():
+    assert md('*hey*dude*') == r'\*hey\*dude\*'
+    assert md('*hey*dude*', escape_asterisks=False) == r'*hey*dude*'
+
+
 def test_underscore():
     assert md('_hey_dude_') == r'\_hey\_dude\_'
+    assert md('_hey_dude_', escape_underscores=False) == r'_hey_dude_'
 
 
 def test_xml_entities():
-    assert md('&') == '&'
+    assert md('&') == r'\&'
 
 
 def test_named_entities():
@@ -19,4 +25,23 @@ def test_hexadecimal_entities():
 
 
 def test_single_escaping_entities():
-    assert md('&') == '&'
+    assert md('&') == r'\&'
+
+
+def text_misc():
+    assert md('\\*') == r'\\\*'
+    assert md('<foo>') == r'\<foo\>'
+    assert md('# foo') == r'\# foo'
+    assert md('> foo') == r'\> foo'
+    assert md('~~foo~~') == r'\~\~foo\~\~'
+    assert md('foo\n===\n') == 'foo\n\\=\\=\\=\n'
+    assert md('---\n') == '\\-\\-\\-\n'
+    assert md('+ x\n+ y\n') == '\\+ x\n\\+ y\n'
+    assert md('`x`') == r'\`x\`'
+    assert md('[text](link)') == r'\[text](link)'
+    assert md('1. x') == r'1\. x'
+    assert md('not a number. x') == r'not a number. x'
+    assert md('1) x') == r'1\) x'
+    assert md('not a number) x') == r'not a number) x'
+    assert md('|not table|') == r'\|not table\|'
+    assert md(r'\ <foo> &amp;amp; | ` `', escape_misc=False) == r'\ <foo> &amp; | ` `'

tests/test_lists.py

@@ -0,0 +1,84 @@
+from markdownify import markdownify as md
+
+
+nested_uls = """
+    <ul>
+        <li>1
+            <ul>
+                <li>a
+                    <ul>
+                        <li>I</li>
+                        <li>II</li>
+                        <li>III</li>
+                    </ul>
+                </li>
+                <li>b</li>
+                <li>c</li>
+            </ul>
+        </li>
+        <li>2</li>
+        <li>3</li>
+    </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>"""
+
+
+def test_ol():
+    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'
+    assert md('<ol start="-1"><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
+    assert md('<ol start="foo"><li>a</li><li>b</li></ol>') == '1. a\n2. b\n'
+    assert md('<ol start="1.5"><li>a</li><li>b</li></ol>') == '1. a\n2. 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_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():
+    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():
+    """
+    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'
+
+
+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'

tests/test_tables.py

@@ -0,0 +1,254 @@
+from markdownify import markdownify as md
+
+
+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_paragraphs = """<table>
+    <tr>
+        <th>Firstname</th>
+        <th><p>Lastname</p></th>
+        <th>Age</th>
+    </tr>
+    <tr>
+        <td><p>Jill</p></td>
+        <td><p>Smith</p></td>
+        <td><p>50</p></td>
+    </tr>
+    <tr>
+        <td>Eve</td>
+        <td>Jackson</td>
+        <td>94</td>
+    </tr>
+</table>"""
+
+table_with_linebreaks = """<table>
+    <tr>
+        <th>Firstname</th>
+        <th>Lastname</th>
+        <th>Age</th>
+    </tr>
+    <tr>
+        <td>Jill</td>
+        <td>Smith
+        Jackson</td>
+        <td>50</td>
+    </tr>
+    <tr>
+        <td>Eve</td>
+        <td>Jackson
+        Smith</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_head_body_missing_head = """<table>
+    <thead>
+        <tr>
+            <td>Firstname</td>
+            <td>Lastname</td>
+            <td>Age</td>
+        </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>"""
+
+table_body = """<table>
+    <tbody>
+        <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>
+    </tbody>
+</table>"""
+
+table_with_caption = """TEXT<table><caption>Caption</caption>
+    <tbody><tr><td>Firstname</td>
+            <td>Lastname</td>
+            <td>Age</td>
+        </tr>
+    </tbody>
+</table>"""
+
+table_with_colspan = """<table>
+    <tr>
+        <th colspan="2">Name</th>
+        <th>Age</th>
+    </tr>
+    <tr>
+        <td colspan="1">Jill</td>
+        <td>Smith</td>
+        <td>50</td>
+    </tr>
+    <tr>
+        <td>Eve</td>
+        <td>Jackson</td>
+        <td>94</td>
+    </tr>
+</table>"""
+
+table_with_undefined_colspan = """<table>
+    <tr>
+        <th colspan="undefined">Name</th>
+        <th>Age</th>
+    </tr>
+    <tr>
+        <td colspan="-1">Jill</td>
+        <td>Smith</td>
+    </tr>
+</table>"""
+
+
+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_paragraphs) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_linebreaks) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith  Jackson | 50 |\n| Eve | Jackson  Smith | 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_head_body_missing_head) == '\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| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_body) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_caption) == 'TEXT\n\nCaption\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n\n'
+    assert md(table_with_colspan) == '\n\n| Name | | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
+    assert md(table_with_undefined_colspan) == '\n\n| Name | Age |\n| --- | --- |\n| Jill | Smith |\n\n'

tox.ini

@@ -0,0 +1,15 @@
+[tox]
+envlist = py38
+
+[testenv]
+passenv = PYTHONPATH
+deps =
+	pytest==8
+	flake8
+	restructuredtext_lint
+	Pygments
+commands =
+	pytest
+	flake8 --ignore=E501,W503 markdownify tests
+	restructuredtext-lint README.rst
+