def post(self, request, format=None):
url = request.data.get('url')
source = request.data.get('source')
s = SummarizeUrl(url)
foo = []
for x in s:
foo.append(remove_non_ascii(x))
summary = ' '.join(foo)
hdr = {
'User-Agent':
'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.64 Safari/537.11',
'Accept':
'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.3',
'Accept-Encoding': 'none',
'Accept-Language': 'en-US,en;q=0.8',
'Connection': 'keep-alive'
}
req = urllib2.Request(url, headers=hdr)
page = urllib2.urlopen(req)
p = parse(page)
title = p.find(".//title").text
data = {
'title': title,
'url': url,
'summarize_url': summary,
'source': source
}
serializer = SummarySerializer(data=data)
if serializer.is_valid():
# serializer.save()
return Response(serializer.data, status=status.HTTP_201_CREATED)
return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)
def get_summary(url):
summaries = SummarizeUrl(url)
summary = []
for elem in summaries:
summary.append(elem)
print ' '.join(summary)
def catchContent():
number_bb = t.count(
'(//div[contains(@data-vr-zone, "Top Stories")]//span[contains(@class, "story-headline")])'
)
df_bb = pd.DataFrame(index=range(0, number_bb - 2),
columns=['Sno', 'Title', 'URL', 'Summary', 'Img_URL'])
for n in range(0, number_bb - 2):
title = t.read(
'//div[contains(@data-vr-zone, "Top Stories {}")]//span[contains(@class, "story-headline")]'
.format(n))
URL_b = t.read(
'//div[contains(@data-vr-zone, "Top Stories {}")]//span[contains(@class, "story-headline")]//@href'
.format(n))
URL = "https://www.straitstimes.com/" + str(URL_b)
Img_URL = t.read(
'//div[contains(@data-vr-zone, "Top Stories {}")]//span[contains(@class, "story-headline")]/ancestor::div[contains(@class, "body")]/..//img/@src'
.format(n))
summaries = SummarizeUrl(URL)
df_bb.iloc[n, 0] = n
df_bb.iloc[n, 1] = title
df_bb.iloc[n, 2] = URL
df_bb.iloc[n, 3] = summaries
df_bb.iloc[n, 4] = Img_URL
return df_bb
def fact_extract(user_input):
#user_input = str(raw_input('Enter input: ')) this will come from the front end instead of cli
url = wikipedia.page(user_input).url
# s = Summarize(user_input, wikipedia.page(user_input).content) is to summarize if the url is not available
# this one is for extracting the content directly from the keyword entered by the user
s = SummarizeUrl(url)
summ = ' '.join(word for word in s)
#summ = summ.replace("]","")
#non reg ex way of dealing with expressions
def removeNestedParentheses(s):
ret = ''
skip = 0
for i in s:
if i == '[':
skip += 1
elif i == ']' and skip > 0:
skip -= 1
elif skip == 0:
ret += i
return ret
summ = removeNestedParentheses(summ)
summ = summ.replace(". ", ".\n\n")
return summ
def get_data(rss, num):
pathToCSV = '../fileStore/file.csv'
data = []
with open(pathToCSV, 'w') as csvfile:
wr = csv.writer(csvfile, delimiter='@', quotechar='#')
index = 0
for e in rss['entries']:
if (index == int(num)):
break
wr.writerow([(e['title']).encode('utf-8')])
wr.writerow([(e['link']).encode('utf-8')])
summary = []
try:
for elem in SummarizeUrl(e['link'].encode('utf-8')):
summary.append(elem)
wr.writerow([
' '.join(summary).encode('utf-8').strip().replace(
'\n', '')
])
except TypeError:
wr.writerow(['Summary Unavailable'])
index = index + 1
def main(url):
# open db connection
conn = sqlite3.connect('../db/summaries.db')
c = conn.cursor()
# check if exists
u = (url, )
rows = c.execute('SELECT COUNT(*) FROM summaries WHERE url=?',
u).fetchone()
if (rows[0] != 0):
#print url + " was found in db"
conn.close()
return
# add the new article
summaries = SummarizeUrl(url)
if summaries is None:
conn.close()
return
sums = " ".join(summaries).replace('\n', '. ')
stage = (url, sums, date.today())
c.execute('INSERT INTO summaries VALUES (?,?,?)', stage)
# save & close
conn.commit()
conn.close()
# delay so we dont overflow the site
print url + " added"
time.sleep(5)
def testURLs(self):
urls = (
u'http://www.huffingtonpost.com/2013/11/22/twitter-forward-secrecy_n_4326599.html',
u'http://www.bbc.co.uk/news/world-europe-30035666',
u'http://www.bbc.co.uk/news/magazine-29631332')
#just make sure it doesn't crash
for url in urls:
summaries = SummarizeUrl(url)
def fact_extract(user_input):
url = wikipedia.page(user_input).url
error = ''
# s = Summarize(user_input, wikipedia.page(user_input).content) is to summarize if the url is not available
# this one is for extracting the content directly from the keyword entered by the user
s = SummarizeUrl(url)
try:
summ = ' '.join(word for word in s)
except TypeError:
return error
summ = removeNestedParentheses(summ)
summ = summ.replace(". ", ".\n\n")
return summ
def summary(request):
title = 'Summary'
form = SummaryForm(request.POST or None)
response_data = {}
context = {
"title": title,
"form": form,
}
if request.method == 'POST':
instance = form.save(commit=False)
url = instance.url
if form.is_valid():
summary = SummarizeUrl(url)
instance.summarize_url = summary
hdr = {
'User-Agent':
'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.64 Safari/537.11',
'Accept':
'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.3',
'Accept-Encoding': 'none',
'Accept-Language': 'en-US,en;q=0.8',
'Connection': 'keep-alive'
}
req = urllib2.Request(url, headers=hdr)
page = urllib2.urlopen(req)
p = parse(page)
title = p.find(".//title").text
instance.title = title
# instance.save()
response_data['url'] = instance.url
response_data['title'] = title
response_data['summary'] = instance.summarize_url
context = {
"title": title,
"url": url,
"summary": summary,
}
return JsonResponse(response_data)
else:
return JsonResponse({"nothing to see": "this isn't happening"})
else:
form = SummaryForm()
return render(request, "website/index.html", context)
def process(links):
for s in range(1, 6):
spath = "story" + str(s) + ".txt"
delete_file_contents(spath)
s = 1
i = 0
for link in links:
try:
i += 1
spath = "story" + str(s) + ".txt"
summaries = SummarizeUrl(link)
append_file(spath, summaries, 0)
if i == 3:
i = 0
s += 1
retrimmer(spath)
except:
pass
def run():
conn = util.create_connection("./db/news.db")
site = util.getSiteByName(conn, "BBC News")
site_url = site[0][2]
site_id = site[0][0]
results, url, number_BBC, Img_link = NewsFromBBC(site_url)
df = pd.DataFrame(index=range(0,number_BBC-1), columns = ['Sno', 'Title', 'URL','Summary', 'Img_URL'])
for i in range(1, number_BBC):
summaries = SummarizeUrl(url[i])
df.iloc[i-1, 0] = i
df.iloc[i-1, 1] = results[i]
df.iloc[i-1, 2] = url[i]
df.iloc[i-1, 3] = summaries
df.iloc[i-1, 4] = Img_link[i]
df = util.fixImgLink(df, "https://cf-templates-fghyux9ggb7t-ap-southeast-1.s3-ap-southeast-1.amazonaws.com/bbc.png")
df = util.fixSummary(df)
util.updateNews(conn, site_id, df)
def getRSSFeedAttributes(feed, dangerFactors, citiesUSA, c, returnList, Errors, feedErrors, linkErrors):
a = feedparser.parse(feed)
for posts in a.entries:
now = datetime.datetime.now()
Date_Of_Access = str(now)
# if 'published' in posts:
try:
date_published = posts['published']
dtobject = parser.parse(date_published)
date_published_reformated = dtobject.strftime('%Y-%m-%d %H:%M:%S')
except Exception, e:
date_published_reformated = None
title = posts.title
title = title.replace("'","''")
if title:
link = posts.link
if link:
factorList = []
domain = []
locationList = []
wordFreqFactor = []
wordFreqLocation = []
try:
summary = (SummarizeUrl(link))
except Exception, e:
SummaryError = str(e)
Errors.append(SummaryError)
feedErrors.append(feed)
linkErrors.append(link)
if summary:
summary = u' '.join(summary)
summary = summary.replace("'","''").replace("\n"," ")
summaryLower = summary.lower()
getFactorsFromText(summaryLower, summary, factorList, domain, locationList, wordFreqFactor, wordFreqLocation, dangerFactors, citiesUSA, c, title, link, date_published_reformated, Date_Of_Access, returnList, feed)
def catchContent():
number = t.count('(//li[contains(@class, "css-1iski2w")]/a)')
df = pd.DataFrame(index=range(0,number), columns = ['Sno', 'Title', 'URL', 'Summary','Img_URL'])
for n in range(1, number+1):
title=t.read('//li[contains(@class, "css-1iski2w")][{}]/a/div'.format(n))
URL=t.read('//li[contains(@class, "css-1iski2w")][{}]//@href'.format(n))
Img_link=t.read('//li[contains(@class, "css-1iski2w")][{}]//img/@src'.format(n))
summaries = SummarizeUrl(URL)
df.iloc[n-1, 0] = n
df.iloc[n-1, 1] = title.decode('utf-8')
df.iloc[n-1, 2] = URL
df.iloc[n-1, 3] = summaries
df.iloc[n-1, 4] = Img_link
df['Summary'].replace('None', np.nan, inplace=True)
df.dropna(subset=['Summary'], inplace=True, how='any')
df= df.reset_index(drop=True)
df['Sno'] = df.index
return df
url_list.append('https://news.google.com/{}'.format(raw[i]))
except IndexError:
# Comment plez
url_list.append('https://news.google.com/{}'.format(raw[i]))
i += 1
for q in range(12):
url_list.pop(0)
return tuple(url_list)
# def __init__(self, q):
# # self.query = q
# Comment this stuff out
if __name__ == "__main__":
# Use this Unicode format though or it will break Google
query = "massachusetts"
lis = scraper(query)
urls = (
u'https://www.cnet.com/news/from-friday-to-pizza-rat-top-viral-videos-of-2010s/',
u'http://www.bbc.co.uk/news/world-europe-30035666',
u'http://www.bbc.co.uk/news/magazine-29631332')
for url in urls:
summaries = SummarizeUrl(url)
pprint(summaries)
# Add random articles
Urls = {
'Huffington':
'http://www.huffingtonpost.com/2013/11/22/twitter-forward-secrecy_n_4326599.html'
}
Urls.update({'BBC': 'http://www.bbc.co.uk/news/business-32817114'})
Urls.update({'BBC Tech': 'http://www.bbc.co.uk/news/technology-32814309'})
Urls.update({'BBC News': 'http://www.bbc.co.uk/news/uk-england-32822298'})
print('Summarising Websites\n\n')
# Loop through each site in the Dictionary
for Site in Urls:
# Print the name of the Site in Use
print('Website: %s') % (Site)
# Use the SummarizeURL function
summaries = SummarizeUrl(Urls[Site])
# Print Each Sentence on a new line
for sentence in summaries:
print sentence
print('\n')
################# Output #################
# Importing Modules
#
# Adding Websites
#
# Summarising Websites
#
# Website: BBC
# Separately, the Federal Reserve fined a sixth bank, Bank of America, $205m over foreign exchange-rigging.
def getNews(rssDict, service, searchedImages):
try:
startTime = time.time()
directory = "./data/" + service + '/'
# create directory for service if doesnt yet exist
if not os.path.exists(directory):
os.makedirs(directory)
print('created directory: %s' % directory)
# iterate each feed in service -> fetch data -> write to temp file -> copy to actual file
for key, value in rssDict.items():
fileName = directory + key + "-write.json"
# delete {category}-write.json file if one already exists
if os.path.exists(fileName):
os.remove(fileName)
print('deleted existing file: %s' % fileName)
feed = feedparser.parse(value) #parse feed to get all the posts
feedDict = {}
feedCounter = 0
# loop through posts in category
for post in feed.entries[:20]: #limit to 20 entries per feed
imgUrl = "none"
# caching enabled. this prevents asking google for images every-time
if post.link in searchedImages:
imgUrl = searchedImages[post.link]
print('found image in cache for %s. done!' % post.link)
else:
query = post.title.split()
query = '+'.join(query)
if (service == 'reuters'):
imgSearch = (
"https://ajax.googleapis.com/ajax/services/search/images?v=1.0&q="
+ query)
else:
imgSearch = (
"https://ajax.googleapis.com/ajax/services/search/images?v=1.0&q="
+ service + "+" + query)
imgSearchRequest = requests.get(imgSearch, verify=False)
if (imgSearchRequest.status_code == 200): #on get success
imgSearchData = imgSearchRequest.json()
try:
getNextImg = 1
imgUrl = imgSearchData['responseData']['results'][
0]['url']
if (service == 'reuters'):
imgUrl = urllib2.unquote(imgUrl)
badBbcUrl = 'http://ichef.bbci.co.uk/news/660/media/images/80201000/png/_80201000_breaking_image_large-3.png'
if (service == 'bbc' and imgUrl == badBbcUrl):
imgUrl = imgSearchData['responseData'][
'results'][1]['url']
getNextImg = 2
# check if select url is actually an image
# if not, choose the next url
if not 'image' in requests.get(
imgUrl,
verify=False).headers['content-type']:
print(
"MISSED FIRST IMG URL = BAD CONTENT. SECOND FETCH!"
)
imgUrl = imgSearchData['responseData'][
'results'][getNextImg]['url']
searchedImages[
post.
link] = imgUrl # add to image cache if img found
print(
'image not in cache but new one fetched for %s. done!'
% post.link)
except (TypeError, IndexError,
requests.exceptions.MissingSchema):
print('DENIAL FROM GOOGLE for %s. failed!' %
post.link)
imgUrl = "200F"
else:
imgUrl = "404"
print(
'image not in cache. also couldnt fetch new one for %s. failed!'
% post.link)
summary = SummarizeUrl(
post.link) # summarize text from article
feedDict[feedCounter] = [
post.title, post.link, summary, imgUrl
]
feedCounter += 1
# write the collected data to {category}-write.json in json format
with open(fileName, 'w') as fp:
json.dump(feedDict, fp)
print('wrote file: %s' % fileName)
# iterate through all cateogries and copy temp files to the actual files
for key, value in rssDict.items():
source = directory + key + "-write.json"
destination = directory + key + ".json"
if os.path.exists(source):
copyfile(source, destination)
print('copied file: %s' % destination)
else:
print('cannot copy file: source %s not found' % source)
print("--- %s seconds ---\n" %
(time.time() - startTime)) #iteration runtime
except Exception:
print("EXCEPTION ERROR EXCEPTION ERROR!!")
pass
len(monarchs) - len(featured) - 1) + ' features queued)'
to = 'test-email'
if production == True:
to = 'Stevie Wonder Says <*****@*****.**>'
subject = s
messageText = random.choice([
"Today's subject is ", "This week's subject is ",
"Today let's have a look at ", "Today's topic is ", "Today let's discuss "
])
messageText += s + '.<br /><br />Did you know...?<br /><br />'
imageUrls = im.getNImages(s, 5)
sentences = SummarizeUrl(wikipediaUrl)
if sentences is None:
sentences = ['Little is known about ' + s + '.'] * 5
for i in xrange(5):
messageText += sentences[i].encode('utf-8') + '<br /><br />'
messageText += '<center><img style="max-height: 480px; max-width: 480px;" alt="one of the best pictures of ' + s + '"" src="' + imageUrls[
i].encode('utf-8') + '" /></center><br /><br />'
messageText += "That's all for now. Have a great week!<br /><br />Long live the queen,<br />Steve"
# feel free to write back with feedback or suggestions on how to improve my newsletter
message = sendEmail.CreateHTMLMessage(sender, to, subject, messageText)
sendEmail.SendMessage(quickstart.gmail_service, 'me', message)
# End of get_json_files()
if __name__ == "__main__":
args = parse_arguments()
json_files = get_json_files(args.dataset)
total_summaries = []
# Summarizing from URL
print("Summarizing from URLs")
for current_file in [json_files[0]]:
with open(current_file, 'r') as json_file:
for line in json_file:
record = json.loads(line)
url = record['URL_s']
print("Summarizing...", url)
summary_sentences = SummarizeUrl(url)
if summary_sentences:
total_summaries.append(" ".join(summary_sentences))
print("Done processing one file")
print("Finished first pass through all records")
print("Recombining and summarizing...")
while len(total_summaries) > 15:
summaries_to_join = int(len(total_summaries) / 15)
if summaries_to_join == 1:
break
if summaries_to_join > 20:
summaries_to_join = 20
combined_summaries = [
" ".join(total_summaries[i:i + summaries_to_join])
for i in range(0, len(total_summaries), summaries_to_join)
def longSummaryFromQuery(query):
g = d.get_zci(query, True)
if "(" in g:
return Summarize(query, g, 15)
else:
return SummarizeUrl(g, 15)
#rss = 'https://www.incibe-cert.es/bfeed/avisos-sci/all'
b = float('0.0')
lenlista = len(listaulr)
for rss in listaulr :
print((b/lenlista)*100)
if hasattr(ssl, '_create_unverified_context'):
ssl._create_default_https_context = ssl._create_unverified_context
d = feedparser.parse(rss.replace('"','')) #<<WORKS!!
if (len(d.entries) > 10) :
ent = 10
else:
ent = len(d.entries)
coda = Coda('c5d228f1-a38e-4956-b6e6-f6ceaac089d0')
for i in range(0,int(ent)) :
try:
summaries = SummarizeUrl(d.entries[i].link)
payload = {
'rows': [
{
'cells': [
{'column': 'c-_zKlSZgEHp', 'value': d.entries[i].title}, {'column': 'c-FaaJIgjDAb', 'value': d.entries[i].link}, {'column': 'c-MM-bPQUWUb', 'value': summaries},
],
},
],
"keyColumns" : [
"c-_zKlSZgEHp"
]
}
print(payload)
coda.upsert_row('G6ody-h41g','grid-30lv9fOPY1',payload)
except:
def wholeArticleFromQuery(query):
g = d.get_zci(query, True)
if "(" in g:
return Summarize(query, g, 1500)
else:
return SummarizeUrl(g, 1500)
def tease_url(url):
summaries = SummarizeUrl(url)
print(summaries)
def summarize():
url = urllib2.unquote(request.args.get('u') or '')
summaries = SummarizeUrl(url)
return jsonify(result=summaries)
