def parse_news_text(self, page_html: str, url: str) -> dict:
if self._extractor is None:
config = Configuration()
config.stopwords_class = StopWords
config.strict = False
extractor = Goose(config)
self._extractor = extractor
article = self._extractor.extract(raw_html=page_html)
news_text = re.sub(r'\s+', r' ', article.cleaned_text)
return {'url': url, 'text': news_text}
def __init__(self, config=None):
self.config = Configuration()
if isinstance(config, dict):
for k, v in list(config.items()):
if hasattr(self.config, k):
setattr(self.config, k, v)
# we don't need to go further if image extractor or local_storage is not set
if not self.config.local_storage_path or \
not self.config.enable_image_fetching:
return
# test if config.local_storage_path is a directory
if not os.path.isdir(self.config.local_storage_path):
os.makedirs(self.config.local_storage_path)
if not os.path.isdir(self.config.local_storage_path):
raise Exception(self.config.local_storage_path +
" directory does not seem to exist, "
"you need to set this for image processing downloads"
)
# test to write a dummy file to the directory to check is directory is writable
level, path = mkstemp(dir=self.config.local_storage_path)
try:
f = os.fdopen(level, "w")
f.close()
os.remove(path)
except IOError:
raise Exception(self.config.local_storage_path +
" directory is not writeble, "
"you need to set this for image processing downloads"
)
def __process_goose(self):
goose_config = Configuration()
goose_config.browser_user_agent = 'Mozilla 5.0'
goose_config.enable_image_fetching = True
g = Goose(config=goose_config)
try:
article = g.extract(self.url)
if article.top_image.src:
self.images = self.get_all_images_from_example_src(
article.top_image.src)
except (requests.exceptions.ConnectionError,
requests.exceptions.ReadTimeout):
return None
return article
def __init__(self, config=None):
# Use the passed in configuration if it is of the right type, otherwise
# use the default as a base
if isinstance(config, Configuration):
self.config = config
else:
self.config = Configuration()
# if config was a passed in dict, parse it into the stored configuration
if isinstance(config, dict):
for k, v in list(config.items()):
if hasattr(self.config, k):
setattr(self.config, k, v)
# we don't need to go further if image extractor or local_storage is not set
if not self.config.local_storage_path or not self.config.enable_image_fetching:
return
# test if config.local_storage_path is a directory
if not os.path.isdir(self.config.local_storage_path):
os.makedirs(self.config.local_storage_path)
if not os.path.isdir(self.config.local_storage_path):
raise Exception(
self.config.local_storage_path +
" directory does not seem to exist, "
"you need to set this for image processing downloads")
def __init__(self, config=None):
# Use the passed in configuration if it is of the right type, otherwise
# use the default as a base
if isinstance(config, Configuration):
self.config = config
else:
self.config = Configuration()
# if config was a passed in dict, parse it into the stored configuration
if isinstance(config, dict):
for k, v in list(config.items()):
if hasattr(self.config, k):
setattr(self.config, k, v)
# setup a single network connection
self.fetcher = NetworkFetcher(self.config)
self.finalizer = weakref.finalize(self, self.close)
# we don't need to go further if image extractor or local_storage is not set
if not self.config.local_storage_path or not self.config.enable_image_fetching:
return
# test if config.local_storage_path is a directory
if not os.path.isdir(self.config.local_storage_path):
os.makedirs(self.config.local_storage_path)
if not os.path.isdir(self.config.local_storage_path):
msg = (
'{} directory does not seem to exist, you need to set this for '
'image processing downloads').format(
self.config.local_storage_path)
raise Exception(msg)
# test to write a dummy file to the directory to check is directory is writable
level, path = mkstemp(dir=self.config.local_storage_path)
try:
with os.fdopen(level, "w"):
pass
os.remove(path)
except IOError:
msg = (
'{} directory is not writeble, you need to set this for image '
'processing downloads').format(self.config.local_storage_path)
raise Exception(msg)
import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import PassiveAggressiveClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score, confusion_matrix
from sklearn.model_selection import cross_val_score
from deep_translator import GoogleTranslator
import requests
from goose3 import Goose
from goose3.configuration import Configuration, ArticleContextPattern
import pickle
import joblib
import numpy as np
#funcao para pegar o text no link e limpar
config = Configuration()
config.known_context_patterns = [
ArticleContextPattern(attr="class", value="n_text")
]
classes = np.arange(100)
df = pd.read_csv('files/newDataset.csv')
conversion_dict = {0: 'Real', 1: 'Fake', 2: 'Neutral'}
df['label'] = df['label'].replace(conversion_dict)
x_train, x_test, y_train, y_test = train_test_split(df['text'],
df['label'],
test_size=0.25,
random_state=7,
shuffle=True)
__version__ = '1.0'
import pickle
from sklearn.model_selection import train_test_split
from algorithm.data_preprocess_main import FETCH_DATA, FEATURE_EXTRACTION, PREPROCESS
from Utility import Utility_func
import sys
import os
from algorithm.predictive_model import Classify
import json
import pandas as pd
from goose3 import Goose
from goose3.configuration import Configuration
config = Configuration()
config.strict = False # turn of strict exception handling
config.browser_user_agent = 'Mozilla 5.0' # set the browser agent string
config.http_timeout = 5.05 # set http timeout in seconds
with Goose(config) as g:
pass
def getContent(url):
try:
content = g.extract(url)
rs = {
"url": url,
"type": None,
"title": content.title,
"description": content.meta_description
def getConfig(self):
config = Configuration()
config.enable_image_fetching = False
return config
import requests
from CQUPT_Spider.utils.common import get_main_content
#url = "http://xwzx.cqupt.edu.cn/cqupt_xwzx/news.jsp?id=5U5933J7L60QL982"
#url = "http://yjs.cqupt.edu.cn/info/1006/4248.htm"
url = "http://cxy.cqupt.edu.cn/info/1105/1282.htm"
#url = "http://xylyh.cqupt.edu.cn/info/1009/1786.htm"
g = Goose({'stopwords_class': StopWordsChinese})
resp = requests.get(url)
content = g.extract(
raw_html=resp.text.encode(resp.encoding).decode(resp.apparent_encoding))
# print(content.cleaned_text)
# 清理标签
config = Configuration()
doc_clean = DocumentCleaner(config, None)
crawler = Crawler(config)
elemetn_html = doc_clean.clean(
crawler.get_document(
resp.text.encode(resp.encoding).decode(resp.apparent_encoding)))
no_tags_html = elemetn_html.text_content()
# 绘图参数
html_content_list = no_tags_html.splitlines()
line_length = [len(line.strip()) for line in html_content_list]
lines = [line.strip() for line in no_tags_html.splitlines()]
main_content, content_lines = get_main_content(lines)
print(main_content)
print(content_lines)
main_line_x = [i for i in range(content_lines[0], content_lines[1] + 1)]
def __init__(self, url):
self.url = url
self.cleaned_text = None
self.config = Configuration()
self.config.browser_user_agent = 'Mozilla 5.0'
self.config.stopwords_class = StopWordsChinese
def extractArticle(url):
from goose3.configuration import Configuration
config = Configuration()
# config.local_storage_path = tmp_dir
return Goose(config).extract(url=url)
