def post(self):
import os, static, hashlib
from models import StoryAuthor, Story, StoryDocument
from django.template.defaultfilters import slugify
from google.appengine.runtime.apiproxy_errors import RequestTooLargeError
title = self.request.get('title')
content = self.request.get('content')
full_name = self.request.get('full_name')
mobile_number = self.request.get('mobile_number')
email = self.request.get('email')
author = StoryAuthor(full_name=full_name, email=email, mobile_number=mobile_number)
author.put()
story = Story(title=title)
if content:
story.content = content
story.author = author
story.put()
try:
request_document = self.request.get('document')
if request_document:
document_file = self.request.POST['document']
document_body = document_file.value
document_digest = hashlib.sha1(document_body).hexdigest()
split_name = os.path.splitext(os.path.basename(document_file.filename))
filename = slugify(split_name[0]) or document_digest
document_name = filename + split_name[1]
document_path = '/story/%d/document/%s/%s' % (story.key().id(), document_digest, document_name)
logging.info(document_path)
story_document = StoryDocument(story=story, path=document_path, name=document_name)
story_document.put()
document = static.set(document_path, document_body, document_file.type, name=document_name)
self.render_to_response("thanks/story.html", document=story_document, story=story)
else:
self.render_to_response("thanks/story.html", story=story)
except RequestTooLargeError, message:
from api_preferences import facebook as fb_prefs, google_friend_connect as gfc
self.render_to_response("start.html",
FACEBOOK_API_KEY=fb_prefs.get('api_key'),
FACEBOOK_CROSS_DOMAIN_RECEIVER_URL=fb_prefs.get('cross_domain_receiver_url'),
GOOGLE_FRIEND_CONNECT_SITE_ID=gfc.get('site_id'),
request_too_large_error=True,
title=title,
content=content,
full_name=full_name,
email=email,
mobile_number=mobile_number)
def create_story(request): token = getToken(request.META) user = getUser(token) if(not user): return Response(status=status.HTTP_403_FORBIDDEN) serialized = StorySerializer(data=request.DATA) if(serialized.is_valid): story = Story() story.title = serialized.initial_data['title'] story.content = serialized.initial_data['content'] story.save() return Response(status=status.HTTP_201_CREATED) else: return Response(serialized._errors, status=status.HTTP_400_BAD_REQUEST)
def newspaperize(article_url):
"""Takes a string url that contains an article. Returns a Story object from
models.py containing information scraped from the article located at the url."""
article = Article(article_url) # create Article object
print("Downloading:", article_url)
try: # returns None if url fails to download
article.download()
except:
print("Failed to download url:", article_url)
return None
try: # returns None if url cannot be parsed
article.parse()
except:
print("Failed to parse url:", article_url)
return None
article.nlp()
# variables to hold values for Story attributes
headline = article.title
imageurl = article.top_image
timestamp = article.publish_date
content = article.text
keywords = article.keywords
summary = article.summary
description = article.meta_description
clickbait = -1 # placeholder for clickbait label
# populates keyword object with article.keywords
list_of_keyword_obj = []
for word in keywords:
if word not in stopword: # prevents stopwords from being keywords
k = Keyword()
k.keyword = word
list_of_keyword_obj.append(k)
s = Story() # create Story object
# set attributes
s.name = headline
s.imageurl = imageurl
s.url = article_url
current_time = datetime.datetime.now()
if timestamp is not None:
s.timestamp = timestamp.isoformat()
else: # generate timestamp if none found
s.timestamp = current_time
s.description = description
s.keywords = list_of_keyword_obj
s.summary = summary
s.content = content
s.clickbait = clickbait
s.createtime = current_time
return s
