def clean_html(html_text,
javascript=True,
scripts=True,
style=True,
embedded=True,
links=True,
forms=True,
frames=True,
comments=True,
annoying_tags=True,
meta=True,
safe_attrs_only=True,
remove_unknown_tags=True,
processing_instructions=True):
"""Clean all the javascript and styles from the HTML returning the string with only the html content"""
# True = Remove | False = Keep
cleaner = Cleaner()
cleaner.javascript = javascript # This is True because we want to activate the javascript filter
cleaner.scripts = scripts # This is True because we want to activate the scripts filter
cleaner.style = style
cleaner.embedded = embedded
cleaner.links = links
cleaner.forms = forms
cleaner.frames = frames
cleaner.comments = comments
cleaner.page_structure = False # Keep page structure
cleaner.annoying_tags = annoying_tags
cleaner.meta = meta
cleaner.safe_attrs_only = safe_attrs_only
cleaner.remove_unknown_tags = remove_unknown_tags
cleaner.processing_instructions = processing_instructions
clean_content = cleaner.clean_html(lxml.html.fromstring(html_text))
return lxml.html.tostring(clean_content)
def html2text(html):
cleaner = Cleaner()
cleaner.javascript = True # This is True because we want to activate the javascript filter
cleaner.style = True
cleaner.scripts = True
cleaner.comments = True
cleaner.links = True
cleaner.meta = True
cleaner.page_structure = True
cleaner.processing_instructions = True
cleaner.forms = True
cleaner.add_nofollow = True
#html = unicodedata.normalize('NFKD', html).encode('ascii','ignore')
try:
document = lxml.html.document_fromstring(html)
c = cleaner.clean_html(document)
html = lxml.html.tostring(c)
soup = BeautifulSoup(html, 'lxml')
parsed_text = soup.get_text()
if (len(parsed_text) > MINSIZE_CHARSDOC):
return parsed_text.lower()
else:
return None
except:
return None
def lxml_extractor(html, url):
'''LXML PARSER'''
cleaner = Cleaner()
cleaner.javascript = True # This is True because we want to activate the javascript filter
cleaner.style = True # This is True because we want to activate the styles & stylesheet filter
cleaner.comments = True
cleaner.embedded = True
cleaner.forms= True
cleaner.frames = True
cleaner.annoying_tags = True
cleaner.kill_tags = NEGATIVE_K
cleaner.allow_tag = POSITIVE_K
cleaner.safe_attrs_only = True
#~ oc = document_fromstring(html, parser=parser, base_url=base_url, **kw)
#~ File "/usr/local/lib/python2.7/dist-packages/lxml/html/__init__.py", line 752, in document_fromstring
#~ value = etree.fromstring(html, parser, **kw)
try:
html = lxml.html.fromstring(html, base_url="url")
tree = cleaner.clean_html(html)
#tree.make_links_absolute(url)
doc = lxml.html.tostring(tree)
doc = soup_extractor(doc, url)
except ValueError:
doc = soup_extractor(html, url)
#~ (title, doc, article, text) = read_extractor(html, url)
#~ print title
#~ doc = (self.doc).replace(unichr(160), " ")
#~ doc = re.sub(spaces,"",self.doc)
return doc
def clearTag_old(self, text: str) -> str:
import lxml
from lxml.html.clean import Cleaner
cleaner = Cleaner()
cleaner.javascript = True
cleaner.style = True
cleaner.links = True
cleaner.meta = True
cleaner.forms = True
cleaner.embedded = True
cleaner.frames = True
cleaner.remove_unknown_tags = True
cleaner.kill_tags = ["img"]
cleaner.remove_tags = [
"strong",
"div",
"body",
"br",
"a",
"p",
"blockquote",
"h3",
"ol",
"li",
"font",
]
return cleaner.clean_html(
lxml.html.document_fromstring(text)).decode("utf-8")
def cleaner_parameters():
reject_list = [
'script', 'noscript', 'style', 'meta', 'semantics', 'img', 'label',
'table', 'li', 'ul', 'ol', 'nav', 'dl', 'dd', 'sub', 'sup', 'math'
]
accept_list = [
'div', 'p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6'
'span', 'b', 'a', 'u', 'i', 'body'
]
html_cleaner = Cleaner()
html_cleaner.remove_unknown_tags = True
html_cleaner.processing_instructions = True
html_cleaner.style = True
html_cleaner.comments = True
html_cleaner.scripts = True
html_cleaner.javascript = True
html_cleaner.meta = True
html_cleaner.links = True
html_cleaner.embedded = True
html_cleaner.annoying_tags = True
html_cleaner.frames = True
html_cleaner.forms = True
html_cleaner.remove_tags = accept_list
html_cleaner.kill_tags = reject_list
return html_cleaner
def sanitize_html(rendered):
if rendered:
cleaner = Cleaner(safe_attrs_only=False)
cleaner.host_whitelist = ('www.youtube.com', 'player.vimeo.com',
'embed.ted.com', 'prezi.com',
'www.google.com', 'www.slideshare.net',
'maps.google.com', 'docs.google.com')
cleaner.forms = False
rendered = cleaner.clean_html(rendered)
# remove div wrapper if there is one
if rendered.startswith('<div>'):
rendered = rendered[5:-6]
return rendered
def sanitize_html(rendered):
if rendered:
cleaner = Cleaner(safe_attrs_only=False)
cleaner.host_whitelist = (
'www.youtube.com',
'player.vimeo.com',
'embed.ted.com',
)
cleaner.forms = False
rendered = cleaner.clean_html(rendered)
# remove div wrapper if there is one
if rendered.startswith('<div>'):
rendered = rendered[5:-6]
return rendered
def sanitize_html(rendered):
if rendered:
cleaner = Cleaner(safe_attrs_only=False)
cleaner.host_whitelist = (
"www.youtube.com",
"player.vimeo.com",
"embed.ted.com",
"prezi.com",
"www.google.com",
"www.slideshare.net",
"maps.google.com",
"docs.google.com",
)
cleaner.forms = False
rendered = cleaner.clean_html(rendered)
# remove div wrapper if there is one
if rendered.startswith("<div>"):
rendered = rendered[5:-6]
return rendered
def f_parse(args):
def isAlphabet(word):
alphabet = [
'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'j', 'k', 'i', 'l', 'm',
'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'x', 'y', 'w', 'z',
'à', 'è', 'é', 'ì', 'í', 'ò', 'ó', 'ù', 'ú'
]
guard = True
for t in word:
if t not in alphabet:
guard = False
return guard
loc = args[0]
corpuses = args[1]
MINSIZE_WORD = 4
MAXSIZE_WORD = 15
MINSIZE_CHARSDOC = 100
MINSIZE_WORDSDOC = 50
cleaner = Cleaner()
cleaner.javascript = True # This is True because we want to activate the javascript filter
cleaner.style = True
cleaner.scripts = True
cleaner.comments = True
cleaner.links = True
cleaner.meta = True
cleaner.page_structure = True
cleaner.processing_instructions = True
cleaner.forms = True
cleaner.add_nofollow = True
ret = []
for document in corpuses:
#html = unicodedata.normalize('NFKD', html).encode('ascii','ignore')
if len(document) > 0:
try:
document = lxml.html.document_fromstring(document)
c = cleaner.clean_html(document)
html = lxml.html.tostring(c)
soup = BeautifulSoup(html, 'lxml')
parsed_text = soup.get_text()
if (len(parsed_text) > MINSIZE_CHARSDOC):
parsed_text = parsed_text.lower()
tokenizer = RegexpTokenizer(r'\w+')
# create English stop words list
en_stop = get_stop_words('en')
it_stop = get_stop_words('it')
sp_stop = get_stop_words('es')
ge_stop = get_stop_words('de')
fr_stop = get_stop_words('fr')
# Create p_stemmer of class PorterStemmer
#p_stemmer = PorterStemmer()
# clean and tokenize document string
tokens = tokenizer.tokenize(parsed_text)
# remove stop words from tokens
stopped_tokens1 = [i for i in tokens if not i in en_stop]
stopped_tokens2 = [
i for i in stopped_tokens1 if not i in it_stop
]
stopped_tokens3 = [
i for i in stopped_tokens2 if not i in sp_stop
]
stopped_tokens4 = [
i for i in stopped_tokens3 if not i in ge_stop
]
stopped_tokens5 = [
i for i in stopped_tokens4 if not i in fr_stop
]
for word in stopped_tokens5:
if not any(char.isdigit() for char in word):
if len(word) > 1:
#check if the word has the alphabet character
if isAlphabet(word):
ret.append(word)
except:
print('Exception : Document empty')
return [loc, ret]
import lxml
from lxml import etree
from lxml.html.clean import autolink_html
from lxml.html.clean import Cleaner
#LXML PARSER
cleaner = Cleaner()
cleaner.javascript = True # This is True because we want to activate the javascript filter
cleaner.style = True # This is True because we want to activate the styles & stylesheet filter
cleaner.comments = True
cleaner.embedded = True
cleaner.forms= True
cleaner.frames = True
#cleaner.safe_attrs_only = True
import re
notalpha = re.compile('[^a-zA-Z]')
#BS PARSER
from bs4 import BeautifulSoup as bs
from bs4 import Comment
def make_links_absolute(soup, url):
return [urlparse.urljoin(url, tag['href']) for tag in soup.findAll('a', href=True)]
def clean_html(soup):
soup = bs(" ".join([s.extract() for s in soup('script')]))
soup = bs(" ".join([s.extract() for s in soup('iframe')]))
soup = bs(" ".join([s.extract() for s in soup('form')]))
soup = bs(" ".join([s.extract() for s in soup('embed')]))
soup = bs(" ".join([s.extract() for s in soup('style')]))
from lxml.html.clean import Cleaner from .filters import duplicate_test, textfilter from .settings import CUT_EMPTY_ELEMS, DEFAULT_CONFIG, MANUALLY_CLEANED, MANUALLY_STRIPPED from .utils import trim LOGGER = logging.getLogger(__name__) # HTML_CLEANER config # http://lxml.de/api/lxml.html.clean.Cleaner-class.html # https://lxml.de/apidoc/lxml.html.clean.html HTML_CLEANER = Cleaner() HTML_CLEANER.annoying_tags = False # True HTML_CLEANER.comments = True HTML_CLEANER.embedded = False # True HTML_CLEANER.forms = False # True HTML_CLEANER.frames = False # True HTML_CLEANER.javascript = False HTML_CLEANER.links = False HTML_CLEANER.meta = False HTML_CLEANER.page_structure = False HTML_CLEANER.processing_instructions = True HTML_CLEANER.remove_unknown_tags = False HTML_CLEANER.safe_attrs_only = False HTML_CLEANER.scripts = False HTML_CLEANER.style = False #HTML_CLEANER.remove_tags = MANUALLY_STRIPPED #HTML_CLEANER.kill_tags = MANUALLY_CLEANED def tree_cleaning(tree, include_tables, include_images=False):
"//*[contains(@id, 'date') or contains(@id, 'Date') or contains(@id, 'datum') or contains(@id, 'Datum') or contains(@id, 'time') or contains(@class, 'post-meta-time')]",
"//*[contains(@class, 'date') or contains(@class, 'Date') or contains(@class, 'datum') or contains(@class, 'Datum')]",
"//*[contains(@class, 'postmeta') or contains(@class, 'post-meta') or contains(@class, 'entry-meta') or contains(@class, 'postMeta') or contains(@class, 'post_meta') or contains(@class, 'post__meta')]",
"//*[@class='meta' or @class='meta-before' or @class='asset-meta' or contains(@id, 'article-metadata') or contains(@class, 'article-metadata') or contains(@class, 'byline') or contains(@class, 'subline')]",
"//*[contains(@class, 'published') or contains(@class, 'posted') or contains(@class, 'submitted') or contains(@class, 'created-post')]",
"//*[contains(@id, 'lastmod') or contains(@itemprop, 'date') or contains(@class, 'time')]",
"//footer",
"//*[@class='post-footer' or @class='footer' or @id='footer']",
"//small",
"//*[contains(@class, 'author') or contains(@class, 'autor') or contains(@class, 'field-content') or @class='meta' or contains(@class, 'info') or contains(@class, 'fa-clock-o')]",
]
CLEANER = Cleaner()
CLEANER.comments = False
CLEANER.embedded = True
CLEANER.forms = False
CLEANER.frames = True
CLEANER.javascript = True
CLEANER.links = False
CLEANER.meta = False
CLEANER.page_structure = True
CLEANER.processing_instructions = True
CLEANER.remove_unknown_tags = False
CLEANER.safe_attrs_only = False
CLEANER.scripts = False
CLEANER.style = True
CLEANER.kill_tags = [
'audio', 'canvas', 'label', 'map', 'math', 'object', 'picture', 'rdf',
'svg', 'video'
] # 'embed', 'figure', 'img', 'table'
from bs4 import BeautifulSoup
import urllib.request
import sys
import testApp.processing as process
import re
from lxml.html.clean import Cleaner
cleaner = Cleaner()
cleaner.javascript = True # This is True because we want to activate the javascript filter
cleaner.style = True
cleaner.scripts = True
cleaner.links = True
cleaner.meta = True
cleaner.page_structure = True
cleaner.frames = True
cleaner.forms = True
cleaner.annoying_tags = True
def get_url_content(url):
try:
with urllib.request.urlopen(url) as page:
text = page.read()
except Exception as e:
return "Couldn't load url"
return text
def index(request):
return HttpResponse("Hello, world. You're at the homepage.")
]
COMMENTS_DISCARD_XPATH = ['.//*[(self::div or self::section)][starts-with(@id, "respond")]', \
'.//cite', \
'.//quote', \
'.//*[starts-with(@id, "reply-") or starts-with(@class, "reply-title")]', \
'.//*[contains(@id, "akismet") or contains(@class, "akismet")]', \
]
# HTML_CLEANER config # http://lxml.de/api/lxml.html.clean.Cleaner-class.html
HTML_CLEANER = Cleaner()
HTML_CLEANER.annoying_tags = True
HTML_CLEANER.comments = True
HTML_CLEANER.embedded = True
HTML_CLEANER.forms = True
HTML_CLEANER.frames = True
HTML_CLEANER.javascript = True
HTML_CLEANER.links = False
HTML_CLEANER.meta = False
HTML_CLEANER.page_structure = False
HTML_CLEANER.processing_instructions = True
HTML_CLEANER.remove_unknown_tags = False
HTML_CLEANER.safe_attrs_only = False
HTML_CLEANER.scripts = True
HTML_CLEANER.style = False
HTML_CLEANER.remove_tags = [
'a', 'abbr', 'acronym', 'address', 'big', 'cite', 'font', 'ins', 'meta',
'small', 'sub', 'sup', 'wbr'
] # 'center', 'table', 'tbody', 'td', 'th', 'tr', 'span',
HTML_CLEANER.kill_tags = [
MAX_YEAR = datetime.date.today().year
## DateDataParser object
PARSERCONFIG = {
'PREFER_DAY_OF_MONTH': 'first',
'PREFER_DATES_FROM': 'past',
'DATE_ORDER': 'DMY'
}
logger.debug('settings: %s %s %s', MIN_YEAR, TODAY, MAX_YEAR)
logger.debug('dateparser configuration: %s', PARSERCONFIG)
cleaner = Cleaner()
cleaner.comments = True
cleaner.embedded = True
cleaner.forms = False
cleaner.frames = True
cleaner.javascript = False
cleaner.links = False
cleaner.meta = False
cleaner.page_structure = True
cleaner.processing_instructions = True
cleaner.remove_unknown_tags = False
cleaner.safe_attrs_only = False
cleaner.scripts = False
cleaner.style = False
cleaner.kill_tags = [
'audio', 'canvas', 'label', 'map', 'math', 'object', 'picture', 'table',
'svg', 'video'
]
# 'embed', 'figure', 'img',
