def select_ids(selector, html_only):
xml_ids = [element.etree_element.get('id', 'nil') for element in
ElementWrapper.from_xml_root(root).query_all(selector)]
html_ids = [element.etree_element.get('id', 'nil') for element in
ElementWrapper.from_html_root(root).query_all(selector)]
if html_only:
assert xml_ids == []
else:
assert xml_ids == html_ids
return html_ids
def test_select_shakespeare():
document = etree.fromstring(HTML_SHAKESPEARE)
body = document.find('.//{http://www.w3.org/1999/xhtml}body')
body = ElementWrapper.from_xml_root(body)
def count(selector):
return sum(1 for _ in body.query_all(selector))
# Data borrowed from http://mootools.net/slickspeed/
# # Changed from original; probably because I'm only
# # searching the body.
# assert count('*') == 252
assert count('*') == 246
# assert count('div:contains(CELIA)') == 26
assert count('div:only-child') == 22 # ?
assert count('div:nth-child(even)') == 106
assert count('div:nth-child(2n)') == 106
assert count('div:nth-child(odd)') == 137
assert count('div:nth-child(2n+1)') == 137
assert count('div:nth-child(n)') == 243
assert count('div:last-child') == 53
assert count('div:first-child') == 51
assert count('div > div') == 242
assert count('div + div') == 190
assert count('div ~ div') == 190
assert count('body') == 1
assert count('body div') == 243
assert count('div') == 243
assert count('div div') == 242
assert count('div div div') == 241
assert count('div, div, div') == 243
assert count('div, a, span') == 243
assert count('.dialog') == 51
assert count('div.dialog') == 51
assert count('div .dialog') == 51
assert count('div.character, div.dialog') == 99
assert count('div.direction.dialog') == 0
assert count('div.dialog.direction') == 0
assert count('div.dialog.scene') == 1
assert count('div.scene.scene') == 1
assert count('div.scene .scene') == 0
assert count('div.direction .dialog ') == 0
assert count('div .dialog .direction') == 4
assert count('div.dialog .dialog .direction') == 4
assert count('#speech5') == 1
assert count('div#speech5') == 1
assert count('div #speech5') == 1
assert count('div.scene div.dialog') == 49
assert count('div#scene1 div.dialog div') == 142
assert count('#scene1 #speech1') == 1
assert count('div[class]') == 103
assert count('div[class=dialog]') == 50
assert count('div[class^=dia]') == 51
assert count('div[class$=log]') == 50
assert count('div[class*=sce]') == 1
assert count('div[class|=dialog]') == 50 # ? Seems right
# assert count('div[class!=madeup]') == 243 # ? Seems right
assert count('div[class~=dialog]') == 51 # ? Seems right
def test_valid_selectors(test):
if test.get('xfail'):
pytest.xfail()
exclude = test.get('exclude', ())
if 'document' in exclude or 'xhtml' in exclude:
return
root = ElementWrapper.from_xml_root(TEST_DOCUMENT)
result = [e.id for e in root.query_all(test['selector'])]
if result != test['expect']:
print(test['selector'])
print(result)
print('!=')
print(test['expect'])
raise AssertionError(test['name'])
def test_lang():
doc = etree.fromstring('''
<html xmlns="http://www.w3.org/1999/xhtml"></html>
''')
assert not ElementWrapper.from_xml_root(doc).matches(':lang(fr)')
doc = etree.fromstring('''
<html xmlns="http://www.w3.org/1999/xhtml">
<meta http-equiv="Content-Language" content=" fr \t"/>
</html>
''')
root = ElementWrapper.from_xml_root(doc, content_language='en')
assert root.matches(':lang(fr)')
doc = etree.fromstring('''
<html>
<meta http-equiv="Content-Language" content=" fr \t"/>
</html>
''')
root = ElementWrapper.from_xml_root(doc, content_language='en')
assert root.matches(':lang(en)')
doc = etree.fromstring('<html></html>')
root = ElementWrapper.from_xml_root(doc, content_language='en')
assert root.matches(':lang(en)')
root = ElementWrapper.from_xml_root(doc, content_language='en, es')
assert not root.matches(':lang(en)')
root = ElementWrapper.from_xml_root(doc)
assert not root.matches(':lang(en)')
doc = etree.fromstring('<html lang="eN"></html>')
root = ElementWrapper.from_html_root(doc)
assert root.matches(':lang(en)')
doc = etree.fromstring('<html lang="eN"></html>')
root = ElementWrapper.from_xml_root(doc)
assert not root.matches(':lang(en)')
def __init__(self, bytestring_svg, url):
tree = ElementTree.fromstring(bytestring_svg)
wrapper = ElementWrapper.from_xml_root(tree)
style = parse_stylesheets(wrapper, url)
self.tree = Node(wrapper, style)
self.url = url
self.filters = {}
self.gradients = {}
self.images = {}
self.markers = {}
self.masks = {}
self.patterns = {}
self.paths = {}
self.cursor_position = [0, 0]
self.cursor_d_position = [0, 0]
self.text_path_width = 0
self.parse_defs(self.tree)
def test_lang():
doc = etree.fromstring('''
<html xmlns="http://www.w3.org/1999/xhtml"></html>
''')
assert not ElementWrapper.from_xml_root(doc).matches(':lang(fr)')
doc = etree.fromstring('''
<html xmlns="http://www.w3.org/1999/xhtml">
<meta http-equiv="Content-Language" content=" fr \t"/>
</html>
''')
root = ElementWrapper.from_xml_root(doc, content_language='en')
assert root.matches(':lang(fr)')
doc = etree.fromstring('''
<html>
<meta http-equiv="Content-Language" content=" fr \t"/>
</html>
''')
root = ElementWrapper.from_xml_root(doc, content_language='en')
assert root.matches(':lang(en)')
doc = etree.fromstring('<html></html>')
root = ElementWrapper.from_xml_root(doc, content_language='en')
assert root.matches(':lang(en)')
root = ElementWrapper.from_xml_root(doc, content_language='en, es')
assert not root.matches(':lang(en)')
root = ElementWrapper.from_xml_root(doc)
assert not root.matches(':lang(en)')
doc = etree.fromstring('<html lang="eN"></html>')
root = ElementWrapper.from_html_root(doc)
assert root.matches(':lang(en)')
doc = etree.fromstring('<html lang="eN"></html>')
root = ElementWrapper.from_xml_root(doc)
assert not root.matches(':lang(en)')
def get_main(root: cssselect2.ElementWrapper) -> cssselect2.ElementWrapper:
""" Get the ARIA main content node from the given root. """
selector = '[role="main"]'
# todo: what if there are multiple nodes?
return root.query(selector)
"""
import xml.etree.ElementTree as etree
from pathlib import Path
import pytest
from cssselect2 import ElementWrapper, SelectorError, compile_selector_list
from .w3_selectors import invalid_selectors, valid_selectors
CURRENT_FOLDER = Path(__file__).parent
IDS_ROOT = etree.parse(CURRENT_FOLDER / 'ids.html')
ALL_IDS = [
element.etree_element.get('id', 'nil')
for element in ElementWrapper.from_html_root(IDS_ROOT).query_all('*')
]
SHAKESPEARE_BODY = (ElementWrapper.from_xml_root(
etree.parse(
CURRENT_FOLDER /
'shakespeare.html').find('.//{http://www.w3.org/1999/xhtml}body')))
def get_test_document():
document = etree.parse(CURRENT_FOLDER / 'content.xhtml')
parent = document.find(".//*[@id='root']")
# Setup namespace tests
for id in ('any-namespace', 'no-namespace'):
div = etree.SubElement(parent, '{http://www.w3.org/1999/xhtml}div')
div.set('id', id)
if len(val) == 1:
return val[0]
elif len(val) > 1:
return val
else:
if required:
raise RequirementMissing(item)
return None
items = []
src = sys.stdin.read()
t = html5lib.parse(src, namespaceHTMLElements=False)
doc = ElementWrapper.from_html_root(t)
for scraper in scrapers:
item_selector = scraper.get("item")
if args.verbose: print ("Running scraper {0}".format(item_selector), file=sys.stderr)
for item_elt in doc.query_all(item_selector):
if args.verbose: print ("ITEM: {0}".format(tag_open(item_elt.etree_element)), file=sys.stderr)
item_elt.etree_element.set("itemscope", "itemscope")
# item_elt = match.etree_element
try:
item = {}
for key, selector in scraper['keys'].items():
if args.verbose:
print ("key, selector: {0}, {1}".format(key, selector), file=sys.stderr)
#continue
item[key] = item_selection(item_elt, selector, key, verbose=args.verbose)