I'm writing a script that extracts the content out of an article and removes any unnecessary stuff eg. scripts and styling. Beautiful Soup keeps raising the following exception:
'' object has no attribute 'contents'
Here's the code of the trim function (element is the HTML element that contains the content of the webpage):
def trim(element):
elements_to_remove = ('script', 'style', 'link', 'form', 'object', 'iframe')
for i in elements_to_remove:
remove_all_elements(element, i)
attributes_to_remove = ('class', 'id', 'style')
for i in attributes_to_remove:
remove_all_attributes(element, i)
remove_all_comments(element)
# Remove divs that have more non-p elements than p elements
for div in element.find_all('div'):
p = len(div.find_all('p'))
img = len(div.find_all('img'))
li = len(div.find_all('li'))
a = len(div.find_all('a'))
if p == 0 or img > p or li > p or a > p:
div.decompose()
Looking at the stack trace, the problem seems to be coming from this method right after the for statement:
# Remove divs that have more non-p elements than p elements
for div in element.find_all('div'):
p = len(div.find_all('p')) #
I don't get why this instance of bs4.element.Tag doesn't have the attribute 'contents'? I tried it out on an actual webpage and the element was full of p's and img's...
Here's the traceback (This is part of a Django project I'm working on):
Environment:
Request Method: POST
Request URL: http://localhost:8000/read/add/
Django Version: 1.4.1
Python Version: 2.7.3
Installed Applications:
('django.contrib.auth',
'django.contrib.contenttypes',
'django.contrib.sessions',
'django.contrib.sites',
'django.contrib.messages',
'django.contrib.staticfiles',
'home',
'account',
'read',
'review')
Installed Middleware:
('django.middleware.common.CommonMiddleware',
'django.contrib.sessions.middleware.SessionMiddleware',
'django.middleware.csrf.CsrfViewMiddleware',
'django.contrib.auth.middleware.AuthenticationMiddleware',
'django.contrib.messages.middleware.MessageMiddleware')
Traceback:
File "/home/marco/.virtualenvs/sandra/local/lib/python2.7/site-packages/django/core/handlers/base.py" in get_response
111. response = callback(request, *callback_args, **callback_kwargs)
File "/home/marco/sandra/read/views.py" in add
24. Article.objects.create_article(request.user, url)
File "/home/marco/sandra/read/models.py" in create_article
11. title, content = logic.process_html(web_page.read())
File "/home/marco/sandra/read/logic.py" in process_html
7. soup = htmlbarber.give_haircut(BeautifulSoup(html_code, 'html5lib'))
File "/home/marco/sandra/read/htmlbarber/__init__.py" in give_haircut
45. scissor.trim(element)
File "/home/marco/sandra/read/htmlbarber/scissor.py" in trim
35. p = len(div.find_all('p'))
File "/home/marco/.virtualenvs/sandra/local/lib/python2.7/site-packages/bs4/element.py" in find_all
1128. return self._find_all(name, attrs, text, limit, generator, **kwargs)
File "/home/marco/.virtualenvs/sandra/local/lib/python2.7/site-packages/bs4/element.py" in _find_all
413. return [element for element in generator
File "/home/marco/.virtualenvs/sandra/local/lib/python2.7/site-packages/bs4/element.py" in descendants
1140. if not len(self.contents):
File "/home/marco/.virtualenvs/sandra/local/lib/python2.7/site-packages/bs4/element.py" in __getattr__
924. "'%s' object has no attribute '%s'" % (self.__class__, tag))
Exception Type: AttributeError at /read/add/
Exception Value: '' object has no attribute 'contents'
Here's the source code of remove_all_* functions:
def remove_all_elements(element_to_clean, unwanted_element_name):
for to_remove in element_to_clean.find_all(unwanted_element_name):
to_remove.decompose()
def remove_all_attributes(element_to_clean, unwanted_attribute_name):
for to_inspect in [element_to_clean] + element_to_clean.find_all():
try:
del to_inspect[unwanted_attribute_name]
except KeyError:
pass
def remove_all_comments(element_to_clean):
for comment in element_to_clean.find_all(text=lambda text:isinstance(text, Comment)):
comment.extract()
解决方案
I think the problem is that in remove_all_elements or somewhere else in your code you are deleting the contents attribute of some of your tags.
It looks like this is happening when you call to_remove.decompose(). Here is the source for that method:
def decompose(self):
"""Recursively destroys the contents of this tree."""
self.extract()
i = self
while i is not None:
next = i.next_element
i.__dict__.clear()
i = next
Here is what happens if you call this function manually:
>> soup = BeautifulSoup('
hi
>>> d0 = soup.find_all('div')[0]
>>> d0
hi
>>> d0.decompose()
>>> d0
Traceback (most recent call last):
...
Traceback (most recent call last):
AttributeError: '' object has no attribute 'contents'
It appears that once you have called decompose on a tag you must never attempt to use that tag again. I'm not quite sure where this is happening though.
One thing I would try checking is that len(element.__dict__) > 0 at all times in your trim() function.