def test_slugify_on_save(self):
item = NewsItem(title="My News Item",
publish_date=now(),
create_user_id=1,
feed_id=1)
item.save()
self.assertEquals(item.slug, 'my-news-item')
def test_slugify_on_save(self):
item = NewsItem(title="My News Item",
publish_date=now(),
create_user_id=1,
feed_id=1)
item.save()
self.assertEquals(item.slug, 'my-news-item')
def add(request):
if request.method == "POST":
feed_id = request.POST.get("feed")
title = request.POST.get("title")
link = request.POST.get("link")
summary = request.POST.get("summary")
date = request.POST.get("date")
if not feed_id or not title or not summary:
return HttpResponseBadRequest(
"Bad Request: must include feed, title, and summary\n")
if date:
try:
date = dateutil.parser.parse(date)
except:
return HttpResponseBadRequest(
"Bad Request: invalid date format\n")
else:
date = datetime.datetime.now()
if not link:
link = ""
try:
feed = NewsFeed.objects.get(id=feed_id)
except NewsFeed.DoesNotExist:
return HttpResponseNotFound("Not Found: no such feed id\n")
item = NewsItem(feed=feed,
title=title,
link=link,
date=date,
summary=summary)
item.save()
# get previous item id
prev_id = None
try:
prev_items = NewsItem.objects.filter(
feed=feed, id__lt=item.id).order_by("-id")[:1]
if len(prev_items) > 0:
prev_id = prev_items[0].id
except:
pass
out = dict()
if prev_id:
out["prev_id"] = str(prev_id)
out["items"] = [item.to_json()]
hr = dict()
hr["body"] = json.dumps(out) + "\n"
fanout_publish("feed-" + str(feed.id), str(item.id), str(prev_id), hr)
return HttpResponse("Posted\n")
else:
return HttpResponseNotAllowed(["POST"])
def create_news_items(is_sticky=False, amount=1):
for i in range(0, amount):
item = NewsItem(title='silly news item name',
publish_date=now(),
published=True,
create_user_id=1,
feed_id=randint(1, 2),
sticky=is_sticky)
item.save()
def get_items():
page = int(request.query.page or 1)
limit = int(request.query.limit or 100)
items = NewsItem.select().paginate(page, limit)
result = []
for item in items:
result.append(model_to_dict(item))
pagination = get_pagination_object(NewsItem.select(), page, limit)
response.set_header('content-type', 'application/json')
response_obj = {"data": result, "pagination": pagination}
return json.dumps(response_obj, cls=MyEncoder)
def add(request):
if request.method == "POST":
feed_id = request.POST.get("feed")
title = request.POST.get("title")
link = request.POST.get("link")
summary = request.POST.get("summary")
date = request.POST.get("date")
if not feed_id or not title or not summary:
return HttpResponseBadRequest("Bad Request: must include feed, title, and summary\n")
if date:
try:
date = dateutil.parser.parse(date)
except:
return HttpResponseBadRequest("Bad Request: invalid date format\n")
else:
date = datetime.datetime.now()
if not link:
link = ""
try:
feed = NewsFeed.objects.get(id=feed_id)
except NewsFeed.DoesNotExist:
return HttpResponseNotFound("Not Found: no such feed id\n")
item = NewsItem(feed=feed, title=title, link=link, date=date, summary=summary)
item.save()
# get previous item id
prev_id = None
try:
prev_items = NewsItem.objects.filter(feed=feed, id__lt=item.id).order_by("-id")[:1]
if len(prev_items) > 0:
prev_id = prev_items[0].id
except:
pass
out = dict()
if prev_id:
out["prev_id"] = str(prev_id)
out["items"] = [item.to_json()]
hr = dict()
hr["body"] = json.dumps(out) + "\n"
fanout_publish("feed-" + str(feed.id), str(item.id), str(prev_id), hr)
return HttpResponse("Posted\n")
else:
return HttpResponseNotAllowed(["POST"])
def get_item(news_id):
response.set_header("content-type", "application/json")
try:
item = NewsItem.select().where(NewsItem.id == news_id).get()
obj = model_to_dict(item)
response_obj = {"data": obj}
return json.dumps(response_obj, cls=MyEncoder)
except DoesNotExist:
abort(404)
def post(self):
# takes a NewsItem key and adds the loggeg-in user to votes
# returns the news object with the updated vote count ?
user = users.get_current_user()
if user:
key = cgi.escape(self.request.get('key'))
if key:
item = NewsItem.get_by_key_name(key)
if user not in item.bumps:
item.bumps.append(user)
item.put()
self.response.out.write(item.key().name())
else:
self.error(401)
def post(self):
key = self.request.get('key')
feed = NewsFeed.get_by_key_name(key)
# FIXME check if feed was retrieved
result = urlfetch.fetch(feed.url)
if result.status_code == 200:
rssfeed = feedparser.parse(result.content)
for i in rssfeed.entries:
item = NewsItem(key_name=i.guid)
item.url = i.link
item.title = i.title
item.text = i.summary
item.date = datetime.datetime(*i.date_parsed[:6])
item.orderdate = datetime.datetime(*i.date_parsed[:6])
item.source = feed
item.put()
feed.last_fetch = datetime.datetime.now()
feed.put()
taskqueue.add(queue_name='fetch-news-queue', url='/admin/feeds/fetch/', params={'key':feed.key})
self.response.out.write('feed pulled')
else:
self.error(500)
def get(self):
url = settings.YAHOO_PIPE % 'rss'
result = urlfetch.fetch(url)
if result.status_code == 200:
feed = feedparser.parse(result.content)
for i in feed.entries:
item = NewsItem(key_name=i.guid)
item.url = i.link
item.title = i.title
item.text = i.summary
item.date = datetime.datetime(*i.date_parsed[:6])
item.orderdate = datetime.datetime(*i.date_parsed[:6])
item.put()
items = db.GqlQuery("SELECT * FROM NewsItem ORDER BY orderdate DESC LIMIT 100")
context = {'news':items }
#context = add_user_to_context(context)
self.response.out.write(
template.render(tmpl('templates/news2.html'),
context ))
else:
self.response.out.write('err')
CUR_DIR = os.path.dirname(os.path.realpath(__file__))
from models import NewsItem
# Set up the template engine to look in the current directory
template_loader = jinja2.FileSystemLoader('templates')
template_env = jinja2.Environment(loader=template_loader)
# Load the template file
template_file = "report.tpl.txt"
template = template_env.get_template(template_file)
# Load all the news items from the past week
seven_days_ago_ts = arrow.utcnow().ceil('hour').replace(days=-2).timestamp
news_items = NewsItem.select().where(NewsItem.published_ts>seven_days_ago_ts)
# Create a dictionary with all the stories grouped by source
sources = {}
for news_item in news_items:
if not sources.has_key(news_item.source):
sources[news_item.source] = {
'items': [],
'name': news_item.source,
}
# Add the news item
sources[news_item.source]['items'].append(news_item)
# Sort the news items for each key
h1_el = soup.find('h1', text='Weather Articles')
tr_els = h1_el.findAllNext('tr')
for tr_el in tr_els:
# Make sure the URL is absolute
link = tr_el.a['href'].strip()
if not link.startswith('http'):
link = 'http://www.wunderground.com' + link
# Create a hash from the URL to make a unique identifier
url_hash = hashlib.md5(link).hexdigest()
# See if the item already exists
try:
item = NewsItem.get(NewsItem.url_hash==url_hash)
print 'Item Exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating new item.'
item = NewsItem()
summary = tr_el.p.text.strip()
headline = tr_el.h3.text.strip()
# Try to get the opengraph data
try:
link_request = requests.get(link)
links_soup = bs4.BeautifulSoup(link_request.text, 'html.parser')
meta_og_title_el = links_soup.find('meta', {'property': 'og:title'})
meta_og_desc_el = links_soup.find('meta', {'property': 'og:description'})
# Set up the template engine to look in the current directory
template_loader = jinja2.FileSystemLoader('templates')
template_env = jinja2.Environment(loader=template_loader)
# Adding filters to enviroment to make them visible in the template
template_env.filters['format_date'] = reformat_date
template_env.filters['get_source_url'] = get_source_url
# Load the template file
template_file = "index.tpl.html"
template = template_env.get_template(template_file)
# Load all the news items
three_days_ago = arrow.utcnow().to('US/Eastern').replace(
hours=-72).format('YYYY-MM-DD')
news_items = NewsItem.select().where(NewsItem.published_date > three_days_ago,
NewsItem.hidden == 0)
news_items.order_by(NewsItem.published_ts)
# Render the template
context = {
'news_items': news_items,
'updated_eastern': arrow.utcnow().to('US/Eastern')
}
output = template.render(context)
# Save the output
filepath = os.path.join(CUR_DIR, 'output/sources.html')
with codecs.open(filepath, 'w', 'utf-8') as f:
f.write(output)
import peewee
import sys
from models import NewsItem
if len(sys.argv) < 2:
sys.exit('Usage: $ python hide_story.py <story id>')
else:
id = sys.argv[1]
try:
id = int(id)
except ValueError:
sys.exit("Invalid Story ID")
# Try to fetch the item
try:
item = NewsItem.get(NewsItem.id==id)
except peewee.DoesNotExist:
sys.exit("Error! News Item with ID %d not found" % id)
print 'You are attempting to hide story id %d' % id
print 'Headline: %s' % item.title
confirm = raw_input("Are you sure? Y/n: ")
if confirm == 'Y':
item.hidden = True
item.save()
'travel',
)
# See if any of the skippable ids are in the story ids
if pcollid in skippable_collection_ids:
print 'Skipping %s story' % pcollid
continue
# If it's also published on weather underground, skip it
if 'wunderground' in tags:
print 'Skipping Weather Underground Story'
continue
# See if the story already exists
try:
item = NewsItem.get(NewsItem.url_hash == url_hash)
print 'Item Exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating new item.'
item = NewsItem()
item.url_hash = url_hash
item.title = unidecode.unidecode(entry['title'].strip())
item.summary = unidecode.unidecode(entry['description'].strip())
item.source = "Weather Channel"
item.link = link
item.published_date = published_date
item.published_ts = published_ts
item.inserted_ts = arrow.utcnow().timestamp
request = requests.get(source_url)
soup = bs4.BeautifulSoup(request.text, 'html.parser')
article_els = soup.findAll('article', class_='story_list span3 col')
for article_el in article_els:
div_el = article_el.find('div', class_='content')
link = 'http://www.sandiegouniontribune.com' + div_el.a['href']
# Create a hash from the URL to make a unique identifier
url_hash = hashlib.md5(link).hexdigest()
# See if the item already exists
try:
item = NewsItem.get(NewsItem.url_hash==url_hash)
print 'Item Exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating new item.'
item = NewsItem()
date = div_el.find('p', class_='date').text.replace('Updated', '').strip()
dt = dateutil.parser.parse(date)
dt = dt.replace(tzinfo=pytz.timezone('US/Pacific'))
published_date = arrow.get(dt).date().strftime('%Y-%m-%d')
headline = div_el.a.text.strip()
published_ts = arrow.get(dt).to('UTC').timestamp
summary = ''
#! /usr/bin/env python
## To be Run every hour
from scrap import UnilagScrap as Scrap
from models import NewsItem
from datetime import datetime
scrapper = Scrap()
news_items = scrapper.get_news_items()
for item in news_items:
## let's insert stuff
# first check that it does not already exist!
to_db = NewsItem.select().where(NewsItem.slug == item['slug'])
if len(to_db) == 0:
# item has not been in the db before
to_db = NewsItem(news_hash=item['news_hash'],
slug=item['slug'],
news_title=item['title'],
news_link=item['link'],
date_updated=item['date_updated'],
intro_text=item['intro_text'],
scrapped_at=datetime.now())
to_db.save()
else:
to_db = to_db.get()
# check if the hash has changed
if to_db.news_hash != item['news_hash']:
to_db.news_hash = item['news_hash']
source_url = 'http://blogs.seattletimes.com/today/category/weather-beat/feed/'
feed = feedparser.parse(source_url)
entries = feed.entries
for entry in entries:
link = entry.link
url_hash = hashlib.md5(link).hexdigest()
date = entry.published_parsed
published_date = arrow.get(date).to('US/Pacific').date().strftime('%Y-%m-%d')
published_ts = arrow.get(date).to('US/Pacific').to('UTC').timestamp
# See if we already have this story
try:
NewsItem.get(NewsItem.url_hash==url_hash)
print 'Item exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating item.'
item = NewsItem()
headline = entry.title
summary = entry.summary
item.url_hash = url_hash
item.link = link
item.source = 'Seattle Times'
item.title = headline
item.summary = summary
item.published_date = published_date
'travel',
)
# See if any of the skippable ids are in the story ids
if pcollid in skippable_collection_ids:
print 'Skipping %s story' % pcollid
continue
# If it's also published on weather underground, skip it
if 'wunderground' in tags:
print 'Skipping Weather Underground Story'
continue
# See if the story already exists
try:
item = NewsItem.get(NewsItem.url_hash==url_hash)
print 'Item Exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating new item.'
item = NewsItem()
item.url_hash = url_hash
item.title = unidecode.unidecode(entry['title'].strip())
item.summary = unidecode.unidecode(entry['description'].strip())
item.source = "Weather Channel"
item.link = link
item.published_date = published_date
item.published_ts = published_ts
item.inserted_ts = arrow.utcnow().timestamp
import peewee
import sys
from models import NewsItem
if len(sys.argv) < 2:
sys.exit('Usage: $ python hide_story.py <story id>')
else:
id = sys.argv[1]
try:
id = int(id)
except ValueError:
sys.exit("Invalid Story ID")
# Try to fetch the item
try:
item = NewsItem.get(NewsItem.id == id)
except peewee.DoesNotExist:
sys.exit("Error! News Item with ID %d not found" % id)
print 'You are attempting to hide story id %d' % id
print 'Headline: %s' % item.title
confirm = raw_input("Are you sure? Y/n: ")
if confirm == 'Y':
item.hidden = True
item.save()
h1_el = soup.find('h1', text='Weather Articles')
tr_els = h1_el.findAllNext('tr')
for tr_el in tr_els:
# Make sure the URL is absolute
link = tr_el.a['href'].strip()
if not link.startswith('http'):
link = 'http://www.wunderground.com' + link
# Create a hash from the URL to make a unique identifier
url_hash = hashlib.md5(link).hexdigest()
# See if the item already exists
try:
item = NewsItem.get(NewsItem.url_hash == url_hash)
print 'Item Exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating new item.'
item = NewsItem()
summary = tr_el.p.text.strip()
headline = tr_el.h3.text.strip()
# Try to get the opengraph data
try:
link_request = requests.get(link)
links_soup = bs4.BeautifulSoup(link_request.text, 'html.parser')
meta_og_title_el = links_soup.find('meta', {'property': 'og:title'})
meta_og_desc_el = links_soup.find('meta',
'PM Update',
)
# Skip the story if it starts with "D.C. area forecast"
prefix_match = False
for prefix in skippable_headline_prefixes:
if entry.title.startswith(prefix):
prefix_match = True
if prefix_match:
print 'Skipping story'
continue
# See if we already have this story
try:
NewsItem.get(NewsItem.url_hash == url_hash)
print 'Item exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating item.'
item = NewsItem()
soup = bs4.BeautifulSoup(entry.description, 'html.parser')
item.summary = unidecode.unidecode(soup.text.strip())
item.title = unidecode.unidecode(entry.title)
item.url_hash = url_hash
item.link = link
item.authors = ''
item.source = 'Capital WX Gang'
item.published_date = published_date
'PM Update',
)
# Skip the story if it starts with "D.C. area forecast"
prefix_match = False
for prefix in skippable_headline_prefixes:
if entry.title.startswith(prefix):
prefix_match = True
if prefix_match:
print 'Skipping story'
continue
# See if we already have this story
try:
NewsItem.get(NewsItem.url_hash==url_hash)
print 'Item exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating item.'
item = NewsItem()
soup = bs4.BeautifulSoup(entry.description, 'html.parser')
item.summary = unidecode.unidecode(soup.text.strip())
item.title = unidecode.unidecode(entry.title)
item.url_hash = url_hash
item.link = link
item.authors = ''
item.source = 'Capital WX Gang'
item.published_date = published_date
headline = li_el.find('h4').text
date = li_el.find('h5').text
description = li_el.find('p').text
# Parse the date
dt = dateutil.parser.parse(date)
dt = dt.replace(tzinfo=pytz.timezone('US/Eastern'))
utc_dt = arrow.get(dt).to('UTC')
published_date = arrow.get(dt).date().strftime('%Y-%m-%d')
# Create a unique identifier from the hash of the URL
url_hash = hashlib.md5(link).hexdigest()
# See if the story already exists
try:
item = NewsItem.get(NewsItem.url_hash==url_hash)
print 'Item Exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating new item.'
item = NewsItem()
# Try to get the opengraph data
try:
link_request = requests.get(link)
links_soup = bs4.BeautifulSoup(link_request.text, 'html.parser')
meta_og_title_el = links_soup.find('meta', {'property': 'og:title'})
meta_og_desc_el = links_soup.find('meta', {'property': 'og:description'})
meta_og_url_el = links_soup.find('meta', {'property': 'og:url'})
except Exception, e:
meta_og_title_el = None
feed = feedparser.parse(source_url)
entries = feed.entries
for entry in entries:
link = entry.link
url_hash = hashlib.md5(link).hexdigest()
date = entry.published_parsed
published_date = arrow.get(date).to('US/Pacific').date().strftime(
'%Y-%m-%d')
published_ts = arrow.get(date).to('US/Pacific').to('UTC').timestamp
# See if we already have this story
try:
NewsItem.get(NewsItem.url_hash == url_hash)
print 'Item exists. Skipping.'
continue
except peewee.DoesNotExist:
print 'Creating item.'
item = NewsItem()
headline = entry.title
summary = entry.summary
item.url_hash = url_hash
item.link = link
item.source = 'Seattle Times'
item.title = headline
item.summary = summary
item.published_date = published_date
return '#'
# Set up the template engine to look in the current directory
template_loader = jinja2.FileSystemLoader('templates')
template_env = jinja2.Environment(loader=template_loader)
# Adding filters to enviroment to make them visible in the template
template_env.filters['format_date'] = reformat_date
template_env.filters['get_source_url'] = get_source_url
# Load the template file
template_file = "index.tpl.html"
template = template_env.get_template(template_file)
# Load all the news items
three_days_ago = arrow.utcnow().to('US/Eastern').replace(hours=-72).format('YYYY-MM-DD')
news_items = NewsItem.select().where(
NewsItem.published_date > three_days_ago,
NewsItem.hidden == 0
)
news_items.order_by(NewsItem.published_ts)
# Render the template
context = {'news_items': news_items, 'updated_eastern': arrow.utcnow().to('US/Eastern') }
output = template.render(context)
# Save the output
filepath = os.path.join(CUR_DIR, 'output/sources.html')
with codecs.open(filepath, 'w', 'utf-8') as f:
f.write(output)
