def test_fetch_document(self):
""" Test 1: Just make sure fetching works """
self.init_webdriver()
# Right now, browser_html should be the raw inner html
browser_html = self.execute_script("""
var data = broadcaster.start_document();
return data['html'];
""")
# Internet explorer values contain windows line endings
browser_html = browser_html.replace('\r\n', '\n')
print "Browser HTML: %s" % (browser_html)
browser_tree = parse_html(browser_html)
browser_xml = lxml.etree.tostring(browser_tree)
# Compare it to the actual HTML file
html_path = util.get_html_path(self.HTML_CONTENT_FILE)
actual_html = open(html_path, 'r').read()
# Semi hack: We expect all browsers to insert tbody, so we'll manually
# insert tbodies into our "expected" html too
#import lxml.html
#actual_tree = lxml.html.fromstring(actual_html)
#util.force_insert_tbody(actual_tree)
#browser_tree = lxml.html.fromstring(browser_html)
assert self.compare_html(actual_html, browser_xml, ignore_script_content=True)
def apply_and_compare(self, html, desired_html=None,
**compare_kwargs):
browser_html = self.to_browser_html(html)
#print "Browser html: %s" % (browser_html)
desired_html = html if desired_html is None else desired_html
parsed_tree = parse_html(browser_html)
parsed_html = lxml.etree.tostring(parsed_tree)
return self.compare_html(desired_html, parsed_html, **compare_kwargs)
def test_apply_document(self):
""" Test 2: Apply document to viewer """
self.init_webdriver()
desired_html = self.TEST_APPLY_DOCUMENT
self.execute_script("""
var de = viewer.get_document_element();
viewer.apply_document(de, arguments[0]);
""", desired_html)
# Parse the viewer html. We need to parse it with our own custom HTML
# parser and then dump it back out as well-formed XML, to allow the
# comparison to proceed.
viewer_html = self.get_viewer_html()
viewer_tree = parse_html(viewer_html)
viewer_xml = lxml.etree.tostring(viewer_tree)
assert self.compare_html(desired_html, viewer_xml, strip_localhost_hrefs_for_ie=True)
def parse_and_compare(self, raw, desired=None, is_fragment=False,
**compare_args):
"""
:param raw: Unsanitised HTML, the type you would expect to see
from a browser's innerHTML. This well be parsed
with mirrordom's parsing routine.
:param desired: What the HTML should look like. MUST be well formed XML.
Will be parsed with a strict XML parser.
"""
global compare_html
desired = raw if desired is None else desired
raw = raw.strip()
desired = desired.strip()
parsed_tree = parse_html(raw)
parsed_html = lxml.etree.tostring(parsed_tree)
return self.compare_html(desired, parsed_html, **compare_args)
def html_to_xml(self, html):
tree = parse_html(html)
return lxml.etree.tostring(tree)
def html_to_xml(self, html):
tree = parse_html(html)
return lxml.etree.tostring(tree)
