def get_answer(self, question):
# We try to find an answer in all of our services
possible_answers = {}
corrected_question = spell_check(question)
entities = get_entities(corrected_question)
# First, we try to find an answer in our local AIML files
possible_answers["AIML"] = aiml_response(corrected_question)
if not self._multiple_answers and possible_answers["AIML"] != "":
return possible_answers["AIML"]
# We search on Wikipedia only if the question starts with `WHO` or `WHAT`;
# but if it is too ambiguous, we still perform a Google Search.
if "WP" in entities.values():
search_words = ""
for k, v in entities.iteritems():
if v.startswith('NN'):
search_words += k + " "
possible_answers["Wikipedia"] = wikipedia_search(search_words[:-1])
if possible_answers["Wikipedia"] == "":
possible_answers["Google"] = google_search(corrected_question)
if not self._multiple_answers and possible_answers[
"Wikipedia"] != "":
return possible_answers["Wikipedia"]
if not self._multiple_answers and possible_answers["Google"] != "":
return possible_answers["Google"]
# We search on Google only if the question starts with `WHERE`, `WHEN`, `WHOSE`, `WHICH`.
if any(pronoun in entities.values()
for pronoun in ["WDT", "WP$", "WRB"]):
possible_answers["Google"] = google_search(corrected_question)
if not self._multiple_answers and possible_answers["Google"] != "":
return possible_answers["Google"]
if not self._multiple_answers:
return "What? I even don't know how to answer. Maybe you tell me!?"
else:
answer = ""
for k, v in possible_answers.iteritems():
if v != "":
answer += v + " ({})\n".format(k)
answer = answer[:-1]
if answer == "":
answer = "What? I even don't know how to answer. Maybe you tell me!?"
return answer
def perform_google_search():
search_term = "+".join(message.content.split()[1:])
try:
results, moreResultsUrl = google_search.google_search(
search_term)
print results
except TypeError:
print "No results"
message.message.reply("No results were found for " +
message.content[9:])
return
print message.content
print search_term
message.message.reply("Search results for **%s**: " %
message.content[9:])
# sould be configurable
count = len(results) < 3 and len(results) or 3
for i in range(0, count):
res = results[i]
room.send_message(
"[%s](%s)" %
(res['titleNoFormatting'], res['unescapedUrl']))
room.send_message(
chaterize_message(res["content"]).replace("\n", " "))
room.send_message("[See More...](%s)" % moreResultsUrl)
def get(self):
searchstr = self.get_argument("search_text")
page = int(self.get_argument("page"))
rsz = int(self.get_argument("rsz"))
print 'searching:', searchstr
results = google_search(searchstr, page, rsz)
nextpage = 1
if len(results) >= 8:
nextpage = page + 1
self.render("search_results.html", results=results, searchstr = searchstr, page = page, nextpage = nextpage)
def get_scholar_venues():
# Searches for all venues gotten through google scholar on dblp.
venues = db.get_scholar_venues()
for v in venues:
venue = v[0]
if "Journal" in venue:
continue
s = get_html(link_base_dblp + venue)
names = s.findAll("venue")
hits = s.findAll("url")
for h in hits:
if not h.text.startswith("htt"):
hits.remove(h)
solid_hit = 0
for name, hit in zip(names, hits):
title = name.text
url = hit.text
similarity = similar(venue, title)
if similarity >= 0.8:
solid_hit = 1
if "/conf" in url:
db.add_venue(title, url)
break
if solid_hit == 0:
searching_term = "dblp " + venue
search_results = google_search(searching_term)
urls = []
for result in search_results:
url = result['link']
if url.endswith("/index"):
if "/conf" in url:
db.add_venue(venue, url)
break
urls.append(url)
print("No match for", venue)
for url in urls:
# For last resort manual adding
if "https://dblp.org/db/conf" in url:
print(url)
def search_google(self):
"""
search on google and create/update the search_keyword in db.
:return: Top 5 links
"""
google_query = self.command[1].lower()
# search on Google API
results = google_search(google_query)
if results:
# Check if the data already exists and then create/update the data.
self.create_update_db(google_query)
results = ' \n'.join(results)
return "The Top five links for your search are -\n{links}".format(
links=results)
else:
return "Sorry, there are no matching Links found"
def search():
status_code = 200
query = request.args.get('query')
info = {}
if 'timespan' not in cache:
cache['timespan'] = {}
if 'timespan' in cache and query in cache['timespan']:
info = cache['timespan'][query]
else:
info = google_search.google_search(query)
cache['timespan'][query] = info
with open('cache.json', 'w') as f:
json.dump(cache, f, indent=4, sort_keys=True, default=str)
is_left = True
for data in info['results']:
data['position'] = 'left' if is_left else 'right'
is_left = not is_left
return render_template("index.html", **info)
def main(args):
#these will be default until command-line arguments are supported
query = 'hummus'
website = 'allrecipes.com'
#Set our allrecipes search result url. in this case we are searching for hummus recipes
search_results_url='http://allrecipes.com/search/default.aspx?qt=k&wt='+query+'&rt=r&origin=Home%20Page'
#create browser
br=create_browser()
gs = google_search.google_search(website,query,MAX_SEARCH_ENTRIES)
gs.loop_until_complete()
results_list = gs.results_list
results_list = [x for x in results_list if allrecipes.validate_url(x)]
#get list of recipe url's from search results
if False:
results_list = read_search_results(search_results_url,br)
print "READING INFO FROM RECIPE PAGES"
#read info from all the recipe pages in to a dict
recipe_info_dict = read_recipe_pages(results_list,br,query)
write_info_to_database(recipe_info_dict)
def perform_google_search():
search_term = "+".join(message.content.split()[1:])
try:
results, moreResultsUrl = google_search.google_search(search_term)
print results
except TypeError:
print "No results"
message.message.reply("No results were found for " + message.content[9:])
return
print message.content
print search_term
message.message.reply("Search results for **%s**: " % message.content[9:])
# sould be configurable
count = len(results) < 3 and len(results) or 3
for i in range(0, count):
res = results[i]
room.send_message("[%s](%s)" % (res['titleNoFormatting'], res['unescapedUrl']))
room.send_message(chaterize_message(res["content"]).replace("\n", " "))
room.send_message("[See More...](%s)" % moreResultsUrl)
def main():
query_string = raw_input("Enter keywords seperated by space \n -->")
keywords = query_string.split(' ')
print 'Performing Bing search...'
bing_images_links = [image['url'] for image in bing_search(keywords)]
print 'Done.'
print 'Performing Giphy search...'
gifs_links = giphy_search(keywords)
print 'Performing Google search...'
google_images_links = google_search(keywords)
print 'Performing Twitter tweets search...'
twitter_text = [tweet[2] for tweet in get_tweets(keywords, media=False)]
print 'Performing Twitter media search...'
twitter_images_links = []
try:
for tweet in get_tweets(keywords, media=True):
if len(tweet)==4:
twitter_images_links.append(tweet[3])
except:
pass
print 'Done performing searches.'
print 'Printing out the results'
print_links('Bing', bing_images_links)
print_links('Giphy', gifs_links)
print_links('Google', google_images_links)
print_links('Twitter', twitter_text)
print_links('Twitter Media', twitter_images_links)
async def bot_google_search(ctx, *args):
# validate user input
if not args:
await ctx.send(env.INPUT_REQUIRED_MSG)
else:
search_keyword = ' '.join(args)
# validate if input is less than or equal to max length
if len(search_keyword) > env.SEARCH_KEY_MAX_LENGTH:
await ctx.send(env.MAX_INPUT_LENGTH_MSG)
else:
# create entry in database
db.create_search_history(ctx.message.author.id, search_keyword)
# get result from google api
results = gs.google_search(search_keyword)
if not results:
await ctx.send(env.NO_SEARCH_RESULT_MSG)
for result in results:
embed = discord.Embed(title=result.get('title'),
url=result.get('link'),
description=result.get('description'),
color=discord.Color.purple())
await ctx.send(embed=embed)
def baike_search(usersay):
usersay = usersay + u' 百科'
baike_search = google_search(usersay)
return google_search
def baike_search(usersay):
usersay = usersay + u' 百科'
baike_search = google_search(usersay)
return baike_search
import pprint, google_search
pprint.pprint(google_search.google_search("高中 网站", 5))
