Bump to 0.3.0

This causes a lot more tests to fail. But it'll be worth it in the end.

README.rst

@@ -27,3 +27,15 @@ Specify tags to exclude (blacklist):
 
     from markdownify import markdownify as md
     md('<b>Yay</b> <a href="http://github.com">GitHub</a>', convert=['b'])  # > '**Yay** GitHub'
+
+
+Development
+===========
+
+To run tests:
+
+``python setup.py test``
+
+To lint:
+
+``python setup.py lint``

markdownify/__init__.py

@@ -1,10 +1,12 @@
-from lxml.html.soupparser import fromstring
+from bs4 import BeautifulSoup, NavigableString
 import re
 
 
 convert_heading_re = re.compile(r'convert_h(\d+)')
 line_beginning_re = re.compile(r'^', re.MULTILINE)
 whitespace_re = re.compile(r'[\r\n\s\t ]+')
+FRAGMENT_ID = '__MARKDOWNIFY_WRAPPER__'
+wrapped = '<div id="%s">%%s</div>' % FRAGMENT_ID
 
 
 def escape(text):
@@ -17,26 +19,32 @@ class MarkdownConverter(object):
     def __init__(self, tags_to_strip=None, tags_to_convert=None):
         if tags_to_strip is not None and tags_to_convert is not None:
             raise ValueError('You may specify either tags to strip or tags to'
-                    ' convert, but not both.')
+                             ' convert, but not both.')
         self.tags_to_strip = tags_to_strip
         self.tags_to_convert = tags_to_convert
 
     def convert(self, html):
-        soup = fromstring(html)
-        return self.process_tag(soup)
+        # We want to take advantage of the html5 parsing, but we don't actually
+        # want a full document. Therefore, we'll mark our fragment with an id,
+        # create the document, and extract the element with the id.
+        html = wrapped % html
+        soup = BeautifulSoup(html)
+        return self.process_tag(soup.find(id=FRAGMENT_ID), children_only=True)
 
-    def process_tag(self, node):
-        text = self.process_text(node.text)
+    def process_tag(self, node, children_only=False):
+        text = ''
 
         # Convert the children first
-        for el in node.findall('*'):
-            text += self.process_tag(el)
+        for el in node.children:
+            if isinstance(el, NavigableString):
+                text += self.process_text(unicode(el))
+            else:
+                text += self.process_tag(el)
 
-        convert_fn = getattr(self, 'convert_%s' % node.tag, None)
-        if convert_fn and self.should_convert_tag(node.tag):
-            text = convert_fn(node, text)
-
-        text += self.process_text(node.tail)
+        if not children_only:
+            convert_fn = getattr(self, 'convert_%s' % node.name, None)
+            if convert_fn and self.should_convert_tag(node.name):
+                text = convert_fn(node, text)
 
         return text
 
@@ -102,8 +110,8 @@ class MarkdownConverter(object):
         return self.convert_em(el, text)
 
     def convert_li(self, el, text):
-        parent = el.getparent()
-        if parent is not None and parent.tag == 'ol':
+        parent = el.parent
+        if parent is not None and parent.name == 'ol':
             bullet = '%s.' % (parent.index(el) + 1)
         else:
             bullet = '*'

markdownify/pkgmeta.py

@@ -1,7 +1,7 @@
 pkgmeta = dict(
     __title__='markdownify',
     __author__='Matthew Tretter',
-    __version__='0.2.0',
+    __version__='0.3.0',
 )
 
 globals().update(pkgmeta)

setup.cfg

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

setup.py

@@ -2,7 +2,7 @@
 import codecs
 import os
 from setuptools import setup, find_packages
-from setuptools.command.test import test as TestCommand
+from setuptools.command.test import test as TestCommand, Command
 
 
 read = lambda filepath: codecs.open(filepath, 'r', 'utf-8').read()
@@ -25,6 +25,37 @@ class PyTest(TestCommand):
         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.engine 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.',
@@ -37,12 +68,14 @@ setup(
     packages=find_packages(),
     zip_safe=False,
     include_package_data=True,
+    setup_requires=[
+        'flake8',
+    ],
     tests_require=[
         'pytest',
     ],
     install_requires=[
-        'lxml',
-        'BeautifulSoup',
+        'beautifulsoup4',
     ],
     classifiers=[
         'Environment :: Web Environment',
@@ -55,8 +88,8 @@ setup(
         'Programming Language :: Python :: 2.7',
         'Topic :: Utilities'
     ],
-    setup_requires=[],
     cmdclass={
         'test': PyTest,
+        'lint': LintCommand,
     },
 )