def htmldiff(old_html, new_html):
try:
old_html_tokens = tokenize(old_html, include_hrefs=False)
new_html_tokens = tokenize(new_html, include_hrefs=False)
except (KeyError, ParserError):
return new_html
result = htmldiff_tokens(old_html_tokens, new_html_tokens)
result = ''.join(result).strip()
return fixup_ins_del_tags(result)
def htmldiff(old_html, new_html):
try:
old_html_tokens = tokenize(old_html, include_hrefs=False)
new_html_tokens = tokenize(new_html, include_hrefs=False)
except (KeyError, ParserError):
return new_html
result = htmldiff_tokens(old_html_tokens, new_html_tokens)
result = ''.join(result).strip()
return fixup_ins_del_tags(result)
def htmldiff(old_html, new_html):
""" Modified lxml.html.diff.htmldiff:
* include_hrefs=False - it's hard to fix this " Link: href " stuff,
and it's not needed (right?)
* do not do fixup_ins_del_tags, as it re-parses everything
and we don't need it here
"""
old_html_tokens = lxml_diff.tokenize(old_html, include_hrefs=False)
new_html_tokens = lxml_diff.tokenize(new_html, include_hrefs=False)
result = lxml_diff.htmldiff_tokens(old_html_tokens, new_html_tokens)
result = ''.join(result).strip()
return result
def compare_html(self, original_text, output_text):
"""Do a diff of two HTML files.
Only the text, <img> tags and <a href=***> attributes in the HTML are diffed.
"""
# We start with a diff of the text tokens alone. This allows us to check that the content is the same (weather or not some non-visible structural elements may have been added/removed/modified)
old_html_tokens = tokenize(output_text)
new_html_tokens = tokenize(original_text)
s = InsensitiveSequenceMatcher(a=old_html_tokens, b=new_html_tokens)
commands = s.get_opcodes()
# If the content is the same it will only have one opcode which states that the objects are equal
self.assertEqual(len(commands), 1)
self.assertEqual('equal', commands[0][0])
# Now we do the real test of equality between the original and the de-encapsulated copy
self.assertEqual(original_text, output_text)
