def fetch(self) -> None:
try:
r = requests.get(self.url, headers=self.headers)
except ConnectionError:
self.is_invalid = True
return
r.encoding = "utf-8"
page = MetadataParser(html=r.text)
self.title = page.get_metadata("title")
self.description = page.get_metadata("description")
self.image = page.get_metadata("image")
def get_or_create_ressource(self, url):
try:
ressource = self.get(url=url)
raise Exception
except Ressource.DoesNotExist:
ressource = Ressource(url=url)
md_strategy = ['og', 'dc', 'page', 'meta', ]
md = MetadataParser(url=url)
ressource.title = md.get_metadata('title', )
ressource.excerpt = md.get_metadata('description', )
ressource.image = md.get_metadata('image', )
ressource.save()
return ressource
def url_matcher(self, msg, match):
url = match.group(0)
r = requests.head(url)
max_size = self.config['DOC_MAX_SIZE']
max_len = self.config['DOC_MAX_LEN']
# files that are too big cause trouble. Let's just ignore them.
if 'content-length' in r.headers and \
int(r.headers['content-length']) > max_size:
return
# ignore anything that is not allowed in configuration
allowed_content_types = self.config['ALLOWED_CONTENT_TYPES']
content_type = ''
if 'content-type' in r.headers:
content_type = re.sub(r'\s*\;.*$', '', r.headers['content-type'])
content_type = content_type.strip()
if content_type not in allowed_content_types:
return
html = requests.get(url).text
readable_article = Document(html).summary()
readable_article = self.text_cleanup(readable_article)
if len(readable_article) > max_len:
readable_article = readable_article[:max_len] + '...'
readable_title = Document(html).title()
page = MetadataParser(html=html)
readable_description = page.get_metadata('description')
if readable_description is None:
readable_description = ''
readable_description = self.text_cleanup(readable_description)
description = ''
if len(readable_description) > len(readable_article):
description = readable_description
else:
description = readable_article
if description:
return "~> {}\n~> {}\n~> {}".format(url,
readable_title,
description)
else:
return "~> {}\n~> {}".format(url,
readable_title)
def url_matcher(self, msg, match):
url = match.group(0)
r = requests.head(url)
max_size = self.config['DOC_MAX_SIZE']
max_len = self.config['DOC_MAX_LEN']
# files that are too big cause trouble. Let's just ignore them.
if 'content-length' in r.headers and \
int(r.headers['content-length']) > max_size:
return
# ignore anything that is not allowed in configuration
allowed_content_types = self.config['ALLOWED_CONTENT_TYPES']
content_type = ''
if 'content-type' in r.headers:
content_type = re.sub(r'\s*\;.*$', '', r.headers['content-type'])
content_type = content_type.strip()
if content_type not in allowed_content_types:
return
html = requests.get(url).text
readable_article = Document(html).summary()
readable_article = self.text_cleanup(readable_article)
if len(readable_article) > max_len:
readable_article = readable_article[:max_len] + '...'
readable_title = Document(html).title()
page = MetadataParser(html=html)
readable_description = page.get_metadata('description')
if readable_description is None:
readable_description = ''
readable_description = self.text_cleanup(readable_description)
description = ''
if len(readable_description) > len(readable_article):
description = readable_description
else:
description = readable_article
if description:
return "~> {}\n~> {}\n~> {}".format(url, readable_title,
description)
else:
return "~> {}\n~> {}".format(url, readable_title)
from metadata_parser import MetadataParser
import pdb
import pprint
# hey use lxml >= 2.3.5 ; use 3.x though!
# otherwise this site will break ! http://www.nasa.gov/externalflash/discovery/index.html
if 0:
a = MetadataParser(url='http://cnn.com')
print(a.get_metadata('title'))
b = MetadataParser(url='http://nyt.com')
print(b.get_metadata('title'))
c = MetadataParser(url='http://thedailybeast.com')
print(c.get_metadata('title'))
print("\n-------------------------------------------------------\n")
print(a.metadata)
print("\n-------------------------------------------------------\n")
print(b.metadata)
print("\n-------------------------------------------------------\n")
print(c.metadata)
print("\n-------------------------------------------------------\n")
print(c.get_metadata('title'))
print(c.get_metadata('canonical'))
print(c.get_metadata('url'))
print(c.absolute_url(c.get_metadata('canonical')))
print(c.absolute_url(c.get_metadata('url')))
print(c.get_discrete_url())
def parsearticle(article, pathuuid):
mainimage = {}
images = []
req = requests.get(
"http://" + os.getenv("RENDER_HOST") + ":3000/render/" +
urllib.parse.quote_plus(json.loads(article.decode('utf-8'))["link"]))
print("http://" + os.getenv("RENDER_HOST") + ":3000/render/" +
urllib.parse.quote_plus(json.loads(article.decode('utf-8'))["link"]))
articletext = MetadataParser(html=json.loads(req.text)['html'])
imgurl = str(articletext.get_metadata('image'))
if not imgurl.startswith("http"):
imgurl = 'http:' + imgurl
imgurlnopost = imgurl.rsplit('?', 1)[0]
imgname = imgurlnopost.rsplit('/', 1)[-1]
imgpath = pathuuid + '/' + imgname + str(uuid.uuid4())
publication = json.loads(article.decode('utf-8'))["publication"]
category = json.loads(article.decode('utf-8'))["category"]
title = json.loads(article.decode('utf-8'))["title"]
articleurl = json.loads(article.decode('utf-8'))["link"]
geturl = None
os.mkdir(pathuuid)
count = 0
try:
geturl = urllib.request.urlretrieve(imgurl, imgpath)
except:
pass
while not geturl:
req = requests.get("http://" + os.getenv("RENDER_HOST") +
":3000/render/" + urllib.parse.quote_plus(
json.loads(article.decode('utf-8'))["link"]))
articletext = MetadataParser(html=json.loads(req.text)['html'])
imgurl = str(articletext.get_metadata('image'))
imgurlnopost = imgurl.rsplit('?', 1)[0]
imgname = imgurlnopost.rsplit('/', 1)[-1]
try:
geturl = urllib.request.urlretrieve(imgurl, imgpath)
count += 1
except:
if count > 10:
raise ValueError('Article failed too many times')
pass
mainimage['imgurl'] = imgurl
mainimage['imgname'] = imgname
mainimage['imgpath'] = imgpath
mainimage['content_type'] = geturl[1]['Content-Type']
images.append(mainimage)
images1 = getimages(
json.loads(req.text)['html'],
json.loads(req.text)['tree']['frameTree']['resources'], images,
pathuuid)
try:
articletext = fulltext(json.loads(req.text)['html'], language='en')
except:
articletext = ""
thing = {}
thing['title'] = json.loads(article.decode('utf-8'))["title"]
thing['articletext'] = articletext
thing['summary'] = summarize(articletext)
thing['assets'] = images1
thing['publication'] = publication
thing['category'] = category
thing['articleurl'] = articleurl
thing['html'] = json.loads(req.text)['html']
return thing
from metadata_parser import MetadataParser
if 0:
a= MetadataParser(url='http://cnn.com')
print a.get_metadata('title')
b= MetadataParser(url='http://nyt.com')
print b.get_metadata('title')
c= MetadataParser(url='http://thedailybeast.com')
print c.get_metadata('title')
print "\n-------------------------------------------------------\n"
print a.metadata
print "\n-------------------------------------------------------\n"
print b.metadata
print "\n-------------------------------------------------------\n"
print c.metadata
print "\n-------------------------------------------------------\n"
print c.get_metadata('title')
print c.get_metadata('canonical')
print c.get_metadata('url')
print c.absolute_url(c.get_metadata('canonical'))
print c.absolute_url(c.get_metadata('url'))
print c.get_discrete_url()
if 0:
a= MetadataParser(url='http://liqr.co/rsvpnewyork')
print "title:"
print a.get_metadata('title')
"group by url order by count(*) desc;"
cur.execute(sql)
urls = cur.fetchall()
i = 0
for url in urls:
i = i + 1
url = remove_characters(url[0])
try:
headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2228.0 Safari/537.36',
'From': '*****@*****.**' # This is another valid field
}
page = MetadataParser(url=url, requests_timeout=5, url_headers=headers)
title = remove_characters(page.get_metadata('title'))
url_resolved = remove_characters(page.get_metadata('url'))
image = remove_characters(page.get_metadata('image'))
description = remove_characters(page.get_metadata('description'))
sql = "insert into url_meta (title, description, url, url_md5, image) " \
"values ('" + title + "', '" + description + "', '" + url_resolved + "', md5('" + url + "'), '" + image + "');"
except Exception as e:
e = remove_characters(str(e))
sql = "insert into url_meta (title, description, url, url_md5, image) " \
"values ('error', '" + e + "', '" + url + "', md5('" + url + "'), '');"
finally:
cur.execute(sql)
cur.execute("commit;")
if i % 100 == 0: print i
from metadata_parser import MetadataParser
from opengraph import OpenGraph
import webpreview
url = 'https://health.usnews.com/wellness/health-buzz/articles/2018-01-05/smelling-your-partners-shirt-could-decrease-your-stress-levels-study-says'
page = MetadataParser(url=url)
print page.metadata
print page.get_metadata('title')
og = OpenGraph(url=url)
print og
wb = webpreview.OpenGraph(url, ['og:title', 'og:description'])
print wb.title
print wb.description
