def _get_text_values_for_schema_node(self, node, value):
result = []
if not value: return result
if type(node.typ) == colander.Mapping:
for cnode in node.children:
name = cnode.name
val = value.get(name, None)
if val:
result += self._get_text_values_for_schema_node(cnode, val)
elif type(node.typ) == colander.Sequence:
if node.children:
cnode = node.children[0]
for val in value:
result += self._get_text_values_for_schema_node(cnode, val)
elif type(node.typ) == colander.Tuple:
for (idx, cnode) in enumerate(node.children):
result += self._get_text_values_for_schema_node(cnode, value[idx])
elif type(node.typ) == colander.String:
if getattr(node, 'include_in_text', True):
if getattr(node, 'is_html', False):
value = html_to_text(value, 0)
if value: result.append(value)
#elif type(node.typ) == deform.FileData:
# pass # FIXME: handle PDF, Word, etc?
return result
def test_htmlutil_html_to_text(self):
from audrey import htmlutil
self.assertEqual(htmlutil.html_to_text(u'''<html><head><title>Title</title></head><body><h1>Header1</h1><p>Hello <unknown foo="bar">world!</unknown> Perhaps some other Æon...<br/><a href="http://python.org">Ooh, a link!</a><ul><li>animal</li><li>vegetable</li><li>mineral</li></ul></p><p>© 2012</p></body></html>'''), u'''Header1\n\n\n\nHello world! Perhaps some other \xc6on...\nOoh, a link!\n\n- animal\n\n- vegetable\n\n- mineral\n\n\xa9 2012''')
self.assertEqual(htmlutil.html_to_text('''<a href="http://python.org">Ooh, a link!</a>''', show_link_urls=True), '''Ooh, a link! [http://python.org]''')
self.assertEqual(htmlutil.html_to_text('''Foo &meh; Bar''', unknown_entity_replacement='?'), 'Foo ? Bar')