def _parse_action(self):
self.action = None
for match in self._tag_re.finditer(self.text):
end = match.group(1) == '/'
tag = match.group(2).lower()
if tag != 'form':
continue
if end:
break
attrs = utils.parse_attrs(match.group(3))
self.action = attrs.get('action', '')
self.method = attrs.get('method', 'GET')
self.id = attrs.get('id')
self.enctype = attrs.get('enctype',
'application/x-www-form-urlencoded')
else:
assert 0, "No </form> tag found"
assert self.action is not None, (
"No <form> tag found")
def _find_element(self, tag, href_attr, href_extract,
content, id,
href_pattern,
html_pattern,
index, verbose):
content_pat = utils.make_pattern(content)
id_pat = utils.make_pattern(id)
href_pat = utils.make_pattern(href_pattern)
html_pat = utils.make_pattern(html_pattern)
body = self.testbody
_tag_re = re.compile(r'<%s\s+(.*?)>(.*?)</%s>' % (tag, tag),
re.I + re.S)
_script_re = re.compile(r'<script.*?>.*?</script>', re.I | re.S)
bad_spans = []
for match in _script_re.finditer(body):
bad_spans.append((match.start(), match.end()))
def printlog(s):
if verbose:
print(s)
found_links = []
total_links = 0
for match in _tag_re.finditer(body):
found_bad = False
for bad_start, bad_end in bad_spans:
if (match.start() > bad_start
and match.end() < bad_end):
found_bad = True
break
if found_bad:
continue
el_html = match.group(0)
el_attr = match.group(1)
el_content = match.group(2)
attrs = utils.parse_attrs(el_attr)
if verbose:
printlog('Element: %r' % el_html)
if not attrs.get(href_attr):
printlog(' Skipped: no %s attribute' % href_attr)
continue
el_href = attrs[href_attr]
if href_extract:
m = href_extract.search(el_href)
if not m:
printlog(" Skipped: doesn't match extract pattern")
continue
el_href = m.group(1)
attrs['uri'] = el_href
if el_href.startswith('#'):
printlog(' Skipped: only internal fragment href')
continue
if el_href.startswith('javascript:'):
printlog(' Skipped: cannot follow javascript:')
continue
total_links += 1
if content_pat and not content_pat(el_content):
printlog(" Skipped: doesn't match description")
continue
if id_pat and not id_pat(attrs.get('id', '')):
printlog(" Skipped: doesn't match id")
continue
if href_pat and not href_pat(el_href):
printlog(" Skipped: doesn't match href")
continue
if html_pat and not html_pat(el_html):
printlog(" Skipped: doesn't match html")
continue
printlog(" Accepted")
found_links.append((el_html, el_content, attrs))
if not found_links:
raise IndexError(
"No matching elements found (from %s possible)"
% total_links)
if index is None:
if len(found_links) > 1:
raise IndexError(
"Multiple links match: %s"
% ', '.join([repr(anc) for anc, d, attr in found_links]))
found_link = found_links[0]
else:
try:
found_link = found_links[index]
except IndexError:
raise IndexError(
"Only %s (out of %s) links match; index %s out of range"
% (len(found_links), total_links, index))
return found_link
def testparse_attrs(self):
self.assertEqual(parse_attrs("href='foo'"), {'href': 'foo'})
self.assertEqual(parse_attrs('href="foo"'), {'href': 'foo'})
self.assertEqual(parse_attrs('href=""'), {'href': ''})
self.assertEqual(parse_attrs('href="foo" id="bar"'),
{'href': 'foo', 'id': 'bar'})
self.assertEqual(parse_attrs('href="foo" id="bar"'),
{'href': 'foo', 'id': 'bar'})
self.assertEqual(parse_attrs("href='foo' id=\"bar\" "),
{'href': 'foo', 'id': 'bar'})
self.assertEqual(parse_attrs("href='foo' id='bar' "),
{'href': 'foo', 'id': 'bar'})
self.assertEqual(parse_attrs("tag='foo\"'"), {'tag': 'foo"'})
self.assertEqual(
parse_attrs('value="<>&"{"'),
{'value': '<>&"{'})
self.assertEqual(parse_attrs('value="∑"'), {'value': '∑'})
self.assertEqual(parse_attrs('value="€"'), {'value': '€'})
def call_FUT(self, obj):
from webtest.utils import parse_attrs
return parse_attrs(obj)
def _parse_fields(self):
in_select = None
in_textarea = None
fields = OrderedDict()
field_order = []
for match in self._tag_re.finditer(self.text):
end = match.group(1) == '/'
tag = match.group(2).lower()
if tag not in ('input', 'select', 'option', 'textarea',
'button'):
continue
if tag == 'select' and end:
assert in_select, (
'%r without starting select' % match.group(0))
in_select = None
continue
if tag == 'textarea' and end:
assert in_textarea, (
"</textarea> with no <textarea> at %s" % match.start())
in_textarea[0].value = utils.html_unquote(
self.text[in_textarea[1]:match.start()])
in_textarea = None
continue
if end:
continue
attrs = utils.parse_attrs(match.group(3))
if 'name' in attrs:
name = attrs.pop('name')
else:
name = None
if tag == 'option':
in_select.options.append((attrs.get('value'),
'selected' in attrs))
continue
if tag == 'input' and attrs.get('type') == 'radio':
field = fields.get(name)
if not field:
field = self.FieldClass.classes['radio'](
self, tag, name, match.start(), **attrs)
fields.setdefault(name, []).append(field)
field_order.append((name, field))
else:
field = field[0]
assert isinstance(field, self.FieldClass.classes['radio'])
field.options.append((attrs.get('value'),
'checked' in attrs))
continue
tag_type = tag
if tag == 'input':
tag_type = attrs.get('type', 'text').lower()
if tag_type == "select" and attrs.get("multiple"):
FieldClass = self.FieldClass.classes.get("multiple_select",
self.FieldClass)
else:
FieldClass = self.FieldClass.classes.get(tag_type,
self.FieldClass)
field = FieldClass(self, tag, name, match.start(), **attrs)
if tag == 'textarea':
assert not in_textarea, (
"Nested textareas: %r and %r"
% (in_textarea, match.group(0)))
in_textarea = field, match.end()
elif tag == 'select':
assert not in_select, (
"Nested selects: %r and %r"
% (in_select, match.group(0)))
in_select = field
fields.setdefault(name, []).append(field)
field_order.append((name, field))
self.field_order = field_order
self.fields = fields
