def processFile(self, file_path_and_name):
try:
f = open(file_path_and_name, 'r')
text = f.read()
text = re.sub('<[^<]+?>', '', text)
sent_tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
lines = sent_tokenizer.tokenize(text.strip())
text = lines
sentences = []
porter = nltk.PorterStemmer()
for sent in lines:
OG_sent = sent[:]
sent = sent.strip().lower()
line = nltk.word_tokenize(sent)
stemmed_sentence = [porter.stem(word) for word in line]
stemmed_sentence = filter(
lambda x: x != '.' and x != '`' and x != ',' and x != '?'
and x != "'" and x != '!' and x != '''"''' and x != "''"
and x != "'s", stemmed_sentence)
if stemmed_sentence != []:
sentences.append(
sentence.sentence(file_path_and_name, stemmed_sentence,
OG_sent))
return sentences
except IOError:
print 'Oops! File not found', file_path_and_name
return [sentence.sentence(file_path_and_name, [], [])]
def whilsent():
if tk.token.code == CodeTable['WHILE']:
tk.token = next(tk.tg)
else:
raise SyntaxError("SyntaxError at line:{},column:{}".format(tk.token.line, tk.token.column))
conditio.conditio()
if tk.token.code == CodeTable['DO']:
tk.token = next(tk.tg)
else:
raise SyntaxError("SyntaxError at line:{},column:{}".format(tk.token.line, tk.token.column))
sentence.sentence()
def compsent():
# 复合语句
if tk.token.code == CodeTable['BEGIN']:
tk.token = next(tk.tg)
else:
raise SyntaxError("SyntaxError at line:{},column:{}".format(
tk.token.line, tk.token.column))
sentence()
sentsuff()
if tk.token.code == CodeTable['END']:
tk.token = next(tk.tg)
else:
raise SyntaxError("SyntaxError at line:{},column:{}".format(
tk.token.line, tk.token.column))
def init_joy():
global sentences
joy_file = open('joyToken.txt', 'r')
for line in joy_file:
line = stop_words_remove(line)
sentences.append(sentence(line, 'joy'))
joy_file.close()
def init_disgust():
global sentences
disgust_file = open('DisgustToken.txt', 'r')
for line in disgust_file:
line = stop_words_remove(line)
sentences.append(sentence(line, 'disgust'))
disgust_file.close()
def buildQuery(sentences, TF_IDF_w, n):
#sort in descending order of TF-IDF values
scores = TF_IDF_w.keys()
scores = sorted(scores, reverse=True)
i = 0
j = 0
queryWords = []
# print("n, len(scores):", n, len(scores))
# if len(scores) == 1:
# print(TF_IDF_w)
# select top n words
while(i<n):
words = TF_IDF_w[scores[j]]
for word in words:
queryWords.append(word)
i=i+1
if (i>n):
break
j=j+1
# return the top selected words as a sentence
# return sentence.sentence("query", queryWords, queryWords)
return sentence.sentence(queryWords, queryWords)
def getemoj(request):
global run
run = run + 1
text = request.GET['sentence']
line = stop_words_remove(text)
sent = sentence(line, 'NONE')
sent.init_sent_vector()
X = numpy.asarray([sent.get_sentence_vector()])
getemoj.model = load_model('my_model.h5')
predictions = getemoj.model.predict(X)
feels = predictions.tolist()
f = feels[0]
ind = f.index(max(f))
if (ind == 0):
response = "you are feeling really afraid?! why so serious?"
elif (ind == 1):
response = "you are feeling really disgusting"
else:
response = "you are feeling happy that's awesome"
return HttpResponse(response)
def init_fear():
global sentences
fear_file = open('fearToken.txt', 'r')
for line in fear_file:
line = stop_words_remove(line)
sentences.append(sentence(line, 'fear'))
fear_file.close()
def processFile(self, file_path_and_name):
try:
# open file
f = open(file_path_and_name, 'r')
text = f.read()
# remove HTML tags
text = nltk.clean_html(text.replace('\n', ''))
# segement data into a list of sentences
sent_tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
lines = sent_tokenizer.tokenize(text.strip())
# ensure names used are in their complete form
# text = self.use_full_names(lines)
text = lines
# convert sentences to list of words
sentences = []
porter = nltk.PorterStemmer()
# every sentence
for sent in lines:
OG_sent = sent[:]
sent = sent.strip().lower()
line = nltk.word_tokenize(sent)
# convert words to stemmed words before appending to list and returning
stemmed_sentence = [porter.stem(word) for word in line]
stemmed_sentence = filter(
lambda x: x != '.' and x != '`' and x != ',' and x != '?'
and x != "'" and x != '!' and x != '''"''' and x != "''"
and x != "'s", stemmed_sentence)
# no empy sentences
if stemmed_sentence != []:
sentences.append(
sentence.sentence(file_path_and_name, stemmed_sentence,
OG_sent))
return sentences
# print error message if file not found
except IOError:
print 'Oops! File not found', file_path_and_name
return [sentence.sentence(file_path_and_name, [], [])]
def __init__(self, parent, sList):
self.numSents = len(sList)
self.parentPaper = parent
self._topic = ''
self.index = 0
self.sentenceList = []
for s in sList:
self.sentenceList.append(sentence(s))
def translated(list_topics, start_idx, end_idx, dir_topic, dir_out, paramter):
#Translate
fileBudget = open("data/budget/budget","r")
budget_string = fileBudget.read().split("\n")
budget = []
for i in range(len(budget_string)-1):
budget.append(int(budget_string[i]))
number_sentence = 0
for i in range(start_idx, end_idx,1):
print("process " + list_topics[i])
file = open(dir_topic + list_topics[i], mode="rb")
parameter[-1] = budget[i]
sentences = file.read().decode("utf8","replace").split('\r\n')
phrase_sentences = []
for line in sentences:
if len(line) != 0:
number_sentence += 1
phrase= []
try:
blob = TextBlob(line)
translated_blob = blob.translate(to='vi')
out_sen = " ".join(translated_blob.tokens)
start = 0
out_sen_tmp = out_sen.lower().split()
end = len(out_sen_tmp)
for nphrase in translated_blob.noun_phrases:
phrase_ele = []
nphrase = nphrase.split()
k = find_index(out_sen_tmp,nphrase,start, end)
start = k + len(nphrase)
for j in range(k,k + len(nphrase),1):
phrase_ele.append(j+1)
phrase.append(phrase_ele)
except:
out_sen = line
phrase = []
if (out_sen != "" ):
sen = sentence.sentence(out_sen,phrase)
phrase_sentences.append(sen)
summarizer = PhraseBaseSummarization.phrase_based_summarization(phrase_sentences)
summary =summarizer.summarizer(parameter)
fileOut = open(dir_out+list_topics[i],"w")
fileOut.write(summary)
print("finish " + list_topics[i])
fileOut.close()
print "no.sentence: ", number_sentence
def create_doc_object(list_of_sentences):
sentences = []
total_tfidf = 0
for sent in list_of_sentences:
words = tb(sent).words
tfidf_dic = tfidf.generate_dictionary_for_specific_words(words)
tfidf_val = tfidf.get_sentence_value(sent)
vector = sentence_to_vec(sent)
sentences.append(
sentence.sentence(sent, tfidf_dic, vector[0], tfidf_val))
total_tfidf += tfidf_val
return doc.documant(sentences, total_tfidf,
sens2vec(sentences, total_tfidf))
def processFile(sample):
# read file from provided folder path
# f = open(file_name,'r')
# text_0 = f.read()
text_0 = sample
# extract content in TEXT tag and remove tags
# text_1 = re.search(r"<TEXT>.*</TEXT>",text_0, re.DOTALL)
# text_1 = re.sub("<TEXT>\n","",text_1.group(0))
# text_1 = re.sub("\n</TEXT>","",text_1)
# # replace all types of quotations by normal quotes
# text_1 = re.sub("\n"," ",text_1)
# text_1 = re.sub("\"","\"",text_1)
# text_1 = re.sub("''","\"",text_1)
# text_1 = re.sub("``","\"",text_1)
# text_1 = re.sub(" +"," ",text_1)
# segment data into a list of sentences
# sentence_token = nltk.data.load('tokenizers/punkt/english.pickle')
# lines = sentence_token.tokenize(text_1.strip())
lines = split_sentences(text_0 + "\n")
# setting the stemmer
sentences = []
porter = nltk.PorterStemmer()
# modelling each sentence in file as sentence object
for line in lines:
# original words of the sentence before stemming
originalWords = line[:]
line = line.strip().lower()
# word tokenization
sent = nltk.word_tokenize(line)
# stemming words
stemmedSent = [porter.stem(word) for word in sent]
# stemmedSent = filter(lambda x: x!='.'and x!='`'and x!=','and x!='?'and x!="'"
# and x!='!' and x!='''"''' and x!="''" and x!="'s", stemmedSent)
# list of sentence objects
if stemmedSent != []:
# sentences.append(sentence.sentence(file_name, stemmedSent, originalWords))
sentences.append(sentence.sentence(stemmedSent, originalWords))
return sentences
def processFile(file_name):
# read file from provided folder path
f = open(file_name,'r')
text_0 = f.read()
# extract content in TEXT tag and remove tags
text_1 = re.search(r"<TEXT>.*</TEXT>",text_0, re.DOTALL)
text_1 = re.sub("<TEXT>\n","",text_1.group(0))
text_1 = re.sub("\n</TEXT>","",text_1)
# replace all types of quotations by normal quotes
text_1 = re.sub("\n"," ",text_1)
text_1 = re.sub("\"","\"",text_1)
text_1 = re.sub("''","\"",text_1)
text_1 = re.sub("``","\"",text_1)
text_1 = re.sub(" +"," ",text_1)
# segment data into a list of sentences
sentence_token = nltk.data.load('tokenizers/punkt/english.pickle')
lines = sentence_token.tokenize(text_1.strip())
# setting the stemmer
sentences = []
porter = nltk.PorterStemmer()
# modelling each sentence in file as sentence object
for line in lines:
# original words of the sentence before stemming
originalWords = line[:]
line = line.strip().lower()
# word tokenization
sent = nltk.word_tokenize(line)
# stemming words
stemmedSent = [porter.stem(word) for word in sent]
stemmedSent = filter(lambda x: x!='.'and x!='`'and x!=','and x!='?'and x!="'"
and x!='!' and x!='''"''' and x!="''" and x!="'s", stemmedSent)
# list of sentence objects
if stemmedSent != []:
sentences.append(sentence.sentence(file_name, stemmedSent, originalWords))
return sentences
def readGoldenFile(self,wordfile, tagfile, numSentences):
""" read training file """
wf = open(wordfile,'rt')
tf = open(tagfile,'rt')
allSentences = []
wlines = wf.readlines()
tlines = tf.readlines()
# build and save sentence objects
for i in range(0,numSentences):
# new sentence
words = wlines[i].split()
tags = tlines[i].split()
s = sentence.sentence(words,tags)
allSentences.append(s)
wf.close()
tf.close()
# save data
self.allSentences = allSentences
self.sentenceNum = len(allSentences)
def buildQuery(sentences, TF_IDF_w, n):
#sort in descending order of TF-IDF values
scores = TF_IDF_w.keys()
scores.sort(reverse=True)
i = 0
j = 0
queryWords = []
# select top n words
while (i < n):
words = TF_IDF_w[scores[j]]
for word in words:
queryWords.append(word)
i = i + 1
if (i > n):
break
j = j + 1
# return the top selected words as a sentence
return sentence.sentence("query", queryWords, queryWords)
def buildQuery(sentences, TF_IDF_w, n):
#sort in descending order of TF-IDF values
scores = TF_IDF_w.keys()
scores.sort(reverse=True)
i = 0
j = 0
queryWords = []
# select top n words
while(i<n):
words = TF_IDF_w[scores[j]]
for word in words:
queryWords.append(word)
i=i+1
if (i>n):
break
j=j+1
# return the top selected words as a sentence
return sentence.sentence("query", queryWords, queryWords)
def processFile(file_name):
# read file from provided folder path
f = open(file_name, 'r')
text_1 = f.read()
# tách câu
sentence_token = nltk.data.load('tokenizers/punkt/english.pickle')
lines = sentence_token.tokenize(text_1.strip())
# setting the stemmer
sentences = []
# modelling each sentence in file as sentence object
for i in range(len(lines)):
line = lines[i]
# giữ lại câu gốc
originalWords = line[:]
# chuyển đối tất cả chữ hoa trong chuỗi sang kiểu chữ thường "Good Mike" => "good mike"
line = line.strip().lower()
# tách từ
stemmedSent = ViTokenizer.tokenize(line).split()
stemmedSent = list(
filter(
lambda x: x != '.' and x != '`' and x != ',' and x != '?' and x
!= "'" and x != ":" and x != '!' and x != '''"''' and x != "''"
and x != '-' and x not in stop_word, stemmedSent))
if ((i + 1) == len(lines)) and (len(stemmedSent) <= 5):
break
# list of sentence objects
if stemmedSent != []:
sentences.append(
sentence.sentence(file_name, stemmedSent, originalWords))
return sentences
def makeQuery(self, n, sentences, idfs):
scored_words = self.TF_IDF(sentences, idfs)
best_words = self.getBestWords(n, scored_words)
return sentence.sentence("query", best_words, [])
web.config.debug = False
render = web.template.render('templates/')
urls = ('/', 'index', '/image', 'image', '/history', 'history',
'/imagedemo/(.*)', 'imagedemo')
app = web.application(urls, globals())
session = web.session.Session(app,
web.session.DiskStore('sessions'),
initializer={
'login': 0,
'user': 0
})
user_control = userControl.user(constant.DATABASE_FILE)
sen = sentence.sentence(constant.DATABASE_FILE)
image_pool = image.image(constant.DATABASE_FILE)
def check():
if session.login == 0:
judge = 4
resp = {'judge': judge}
return True, render.index(resp)
else:
return False, None
class index():
def GET(self):
judge = 4
else:
location = prevlocation
newlocation = False
location = location.replace(' ','-') #Google requires a '-' in 2-word city names
if location != '':
if newlocation: #If location hasn't changed, don't fetch data again. It's already available
print 'Fetching weather information from Google...'
# Call Google weather to get current weather conditions
google_result = weather.get_weather(location)
# Print today's weather report. Replace this with code to deduce the user request
if time == '' or time == 'today' :
printstring = sentence.sentence(google_result['current_conditions']['condition'], time)
print printstring, time
else :
if time == 'tomorrow':
printstring = sentence.sentence(google_result['forecasts'][1]['condition'], time)
print printstring, time
else:
found = False
for i in range(4):
if google_result['forecasts'][i]['day_of_week'] == newtime:
printstring = sentence.sentence(google_result['forecasts'][i]['condition'], time)
print printstring, "on", time
found = True
if not found:
print "Forecast for " + time + " is not available currently."
logging.basicConfig(
format="[%(asctime)s - %(filename)s:line %(lineno)s] %(message)s",
datefmt='%d %b %H:%M:%S',
level=logging.INFO)
logger.setLevel(logging.INFO)
from setup import START
from constant import *
web.config.debug = False
render = web.template.render('adminTemplates/')
urls = ('/', 'index', '/image', 'image', '/history', 'history', '/single',
'single')
sen = sentence.sentence(DATABASE_FILE)
user_control = userControl.user(DATABASE_FILE)
image_pool = image.image(DATABASE_FILE)
class index:
def GET(self):
i = web.input()
start = i.get("start")
end = i.get("end")
if start is None:
start = START
if end is None:
end = time.time()
def chat():
# keyword conditions
condnext = False
condweather = False
condtime = False
condlocation = False
condtemp = False
condkey = False
condresponse = False
foundinfo = False
condtrain = False
condcountry = False
condspellcheck = True
# global variables
conversation = []
location = ''
prevlocation = location
time = 'today'
key = ''
keytemplate = []
fulltime = ''
numdays = ''
logstr = ''
printstr = ''
responsedict = {} # Dictionary to hold all inputs without predefined responses. This dictionary will be written into predefined_responses.txt before exiting the program.
# read data files
citylist = readfile.readfile('cities.txt')
keylist = readfile.readfile('keywords.txt')
timelist = readfile.readfile('time.txt')
condlist = readfile.readfile('conditions.txt')
numlist = readfile.readfile('numbers.txt')
countrylist = readfile.readfile('countries.txt')
exitlist = ['exit', 'quit', 'bye', 'ok']
# Greeting message
printstr = 'Hello! You can ask me questions about the weather in any major city in the world. What would you like to know?'
print printstr
logstr += '\n\n' + printstr
# Start main loop
while True :
foundinfo = False
condtrain = False
condcountry = False
# read input from user
input = raw_input('\nMe > ')
logstr += '\nMe > ' + input + '\nBot > '
if input in exitlist:
if input == 'ok':
exitans = raw_input("Do you want to quit? (y/n)")
if exitans in ('y','Y','Yes','YES','yes'):
break
else:
continue
break
if input == 'disable spellcheck':
condspellcheck = False
continue
if input == 'enable spellcheck':
condspellcheck = True
continue
condcorrected = False
if condspellcheck:
corrected_input = ''
for i in input.split():
str = spellcheck.correct(i)
if str != i:
condcorrected = True
corrected_input += str + ' '
if condcorrected:
print 'did you mean: \"' + corrected_input + '\"?'
input = corrected_input
currentstring = input.split()
conversation.append(currentstring)
# Start searching input for each of the keywords
if input == 'train':
condtrain = True
printstr = 'Entering training mode. Enter input and response seperated by a "|": input|response. Type "exit" to quit training mode'
print printstr
logstr += '\n' + printstr + '\n'
while True:
traininput = raw_input('>')
if traininput == 'exit':
break
if traininput.find('|') < 0:
printstr = 'Format error: use input|response'
print printstr
logstr += '\n' + printstr + '\n'
continue
traininput = traininput.split('|')
responsedict[traininput[0]] = traininput[1]
if condtrain:
continue
for i in countrylist:
for j in currentstring:
if lower(i[0]) == lower(j):
printstr = 'Which city in ' + i[0] + '?'
condcountry = True
foundinfo = True
break
if condcountry:
print printstr
logstr += printstr
continue
if 'next' in input:
foundinfo = True
condnext = True
condtime = False
numdays = currentstring[currentstring.index('next') + 1]
for i in numlist:
if numdays == i[0]:
numdays = i[1]
break
if re.match('[0-9]*$',numdays):
numdays = int(numdays)
else:
numdays = ''
if 'weather' in input:
foundinfo = True
condweather = True
condkey = False
condtemp = False
key = ''
keytemplate = []
# get key from input
for i in keylist:
if i[0] in input:
if 'sunday' in lower(input) and i[0] == 'sun':
break
else:
foundinfo = True
condkey = True
condweather = False
condtemp = False
key = i[0]
keytemplate = i
break
# get time from input
for i in timelist:
if lower(i[0]) in input:
foundinfo = True
condtime = True
numdays = ''
if lower(i[0]) != 'today' and lower(i[0]) != 'tomorrow':
time = i[1]
fulltime = i[0]
break
else:
time = i[0]
fulltime = time
break
if fulltime == '':
fulltime = time
if numdays != '':
condtime = True
if numdays > 4:
printstr = 'Forecast is available only for the next 4 days.'
print printstr
logstr += '\n' + printstr + '\n'
else:
time = ''
fulltime = ''
count = numdays
# get location from input
for i in citylist:
if lower(i[0]) in input:
foundinfo = True
condlocation = True
location = i[0]
break
# find if a new location has been mentioned. if not, don't fetch data again
if location != prevlocation:
newlocation = True
condlocation = True
prevlocation = location
else:
newlocation = False
if location is '':
if prevlocation is '':
condlocation = False
else:
location = prevlocation
newlocation = False
location = location.replace(' ','-') #Google requires a '-' in 2-word city names
result = False
# get temperature from input
if 'temperature' in input:
foundinfo = True
condtemp = True
# User gave no infomation about weather. Switching to general predefined response based chat
if not foundinfo:
response = predefined_responses.respond(input, responsedict)
if response == '':
printstr = "I don't know what that means. If I asked you the same question, what would you reply?"
print printstr
logstr += printstr
responseinput = raw_input('Me > ')
logstr += '\nMe > ' + responseinput
if not responseinput in ('exit', 'quit'):
responsedict[input] = responseinput
print 'response learnt'
else:
printstr = response
print printstr
logstr += printstr
continue
if condlocation:
if newlocation: #If location hasn't changed, don't fetch data again. It's already available
printstr = 'Fetching weather information from Google...'
print printstr
logstr += printstr
# Call Google weather to get current weather conditions
google_result = weather.get_weather(location)
if google_result == {}:
print 'Could not get data from google.'
continue
# We have a valid location. Get further information
# User has asked about temperature. Return temperature information and continue
if condtemp:
printstr = temperature.temperature(google_result, time)
print printstr
logstr += printstr
continue
# User has asked about a specific weather condition. Print information. There are 2 possibilities:
# 1. Find the condition in the next n days
# 2. Find the condition in a specified day
if condkey:
# 1. User has asked about a specific condition in the 'next x days'. Return appropriate response
printstr = ''
timecounter = 0
day_of_week = ''
condition = ''
if numdays != '':
for i in google_result['forecasts']:
count -= 1
if count < 0:
break
if key in lower(i['condition']):
result = True
day_of_week = i['day_of_week']
condition = i['condition']
break
for i in timelist:
if i[0] != 'today' and i[0] != 'tomorrow':
if i[1] == day_of_week:
fulltime = i[0]
break
if result:
printstr = keytemplate[3] + keytemplate[0] + ' on ' + fulltime
else:
printstr = keytemplate[4] + keytemplate[0] + ' in the next ' + str(numdays) + ' days.'
print printstr
logstr += printstr
continue
# 2. User has asked about a particular condition on a particular day. Return appropriate response
if time != 'today' and time != 'tomorrow':
for i in google_result['forecasts']:
if i['day_of_week'] == time:
if key in lower(i['condition']):
printstr = keytemplate[3] + keytemplate[0] + ' on'
else:
printstr = keytemplate[4] + keytemplate[0] + ' on'
elif time == 'today':
fulltime = time
if key in lower(google_result['current_conditions']['condition']):
printstr = keytemplate[1] + keytemplate[0]
else:
printstr = keytemplate[2] + keytemplate[0]
elif time == 'tomorrow':
fulltime = time
if key in lower(google_result['forecasts'][1]['condition']):
printstr = keytemplate[3] + keytemplate[0]
else:
printstr = keytemplate[4] + keytemplate[0]
printstr = printstr + ' ' + fulltime
print printstr
logstr += printstr
continue
# User is asking about today's weather. Print details
elif time == '' or time == 'today' :
printstr = sentence.sentence(google_result['current_conditions']['condition'], time)
printstr += ' ' + fulltime + '. ' + google_result['current_conditions']['humidity'] + ' '
if google_result['current_conditions'].has_key('wind_condition'):
printstr += google_result['current_conditions']['wind_condition']
print printstr
logstr += printstr
continue
# User is asking about weather of a particular day. Print details
elif time == 'tomorrow':
printstr = sentence.sentence(google_result['forecasts'][1]['condition'], time)
printstr += ' ' + fulltime
print printstr
logstr += printstr
else:
found = False
for i in range(4):
if google_result['forecasts'][i]['day_of_week'] == time:
printstr = sentence.sentence(google_result['forecasts'][i]['condition'], time)
printstr += " on" + ' ' + fulltime
print printstr
logstr += printstr
found = True
if not found:
printstr = "Forecast for " + time + " is not available currently."
print printstr
logstr += printstr
continue
else:
printstr = 'What\'s the location?'
print printstr
logstr += printstr
# End of outermost while loop.
# Print message before exiting program
dictcount = 0
for i in responsedict:
dictcount += 1
if dictcount > 0:
printstr = 'Writing new entries to database...'
print printstr
logstr += printstr
datafile = file('predefined_responses.txt', 'a')
for i in responsedict.keys():
trimmedi = re.sub('[^a-zA-Z0-9 ]+','', i)
string = trimmedi + '|' + responsedict[i] + '\n'
datafile.write(string)
log.log(logstr)
print 'Ending the program...'
print 'Bye!'
# End of function chat()
def bot():
conversation = []
location = ''
time = 'today'
key = ''
keytemplate = []
fulltime = ''
numdays = ''
citylist = readfile.readfile('cities.txt')
keylist = readfile.readfile('keywords.txt')
timelist = readfile.readfile('time.txt')
condlist = readfile.readfile('conditions.txt')
numlist = readfile.readfile('numbers.txt')
exitlist = ['exit', 'quit', 'bye', 'ok']
print 'Hello! You can ask me questions about the weather in any major city in the world. What would you like to know?'
while True :
input = raw_input('Me > ')
if input in exitlist:
break
currentstring = input.split()
conversation.append(currentstring)
if 'next' in currentstring:
numdays = currentstring[currentstring.index('next') + 1]
for i in numlist:
if numdays == i[0]:
numdays = i[1]
break
if re.match('[0-9]*$',numdays):
numdays = int(numdays)
else:
numdays = ''
if 'weather' in currentstring:
key = ''
keytemplate = []
# get key from input
for i in keylist:
if i[0] in input:
key = i[0]
keytemplate = i
break
# get time from input
for i in timelist:
if lower(i[0]) in input:
numdays = ''
if lower(i[0]) != 'today' and lower(i[0]) != 'tomorrow':
time = i[1]
fulltime = i[0]
break
else:
time = i[0]
fulltime = time
break
if fulltime == '':
fulltime = time
if numdays != '':
if numdays > 4:
print 'Forecast is available only for the next 4 days.'
else:
time = ''
fulltime = ''
count = numdays
prevlocation = location
#We store previous location to avoid re-fetching data if the location hasn't been changed
# Below, we check if any token in the input matches a city name, and if so, set location to that city
newlocation = False
# get location from input
foundLocation = False
for i in citylist:
if lower(i[0]) in input:
location = i[0]
foundLocation = True
break
#if not foundLocation:
#if location != '':
#print "I didn't find any city name in your input. I'll get you information about " + location
# find if a new location has been mentioned. if not, don't fetch data again
if location is not prevlocation:
newlocation = True
if location is '':
if prevlocation is '':
print 'City not found'
else:
location = prevlocation
newlocation = False
location = location.replace(' ','-') #Google requires a '-' in 2-word city names
result = False
if location is not '':
if newlocation: #If location hasn't changed, don't fetch data again. It's already available
print 'Fetching weather information from Google...'
# Call Google weather to get current weather conditions
google_result = weather.get_weather(location)
if 'temperature' in currentstring:
print temperature.temperature(google_result, time)
continue
printed = False
if key is not '':
printstring = ''
timecounter = 0
day_of_week = ''
condition = ''
if numdays != '':
for i in google_result['forecasts']:
count -= 1
if count < 0:
break
if key in lower(i['condition']):
result = True
day_of_week = i['day_of_week']
condition = i['condition']
break
for i in timelist:
if i[0] != 'today' and i[0] != 'tomorrow':
if i[1] == day_of_week:
fulltime = i[0]
break
if result:
printstring = keytemplate[3] + keytemplate[0] + ' on ' + fulltime
else:
printstring = keytemplate[4] + keytemplate[0] + ' in the next ' + str(numdays) + ' days.'
print printstring
printed = True
if not printed:
if time != 'today' and time != 'tomorrow':
for i in google_result['forecasts']:
if i['day_of_week'] == time:
if key in lower(i['condition']):
printstring = keytemplate[3] + keytemplate[0] + ' on'
else:
printstring = keytemplate[4] + keytemplate[0] + ' on'
elif time == 'today':
fulltime = time
if key in lower(google_result['current_conditions']['condition']):
printstring = keytemplate[1] + keytemplate[0]
else:
printstring = keytemplate[2] + keytemplate[0]
elif time == 'tomorrow':
fulltime = time
if key in lower(google_result['forecasts'][1]['condition']):
printstring = keytemplate[3] + keytemplate[0]
else:
printstring = keytemplate[4] + keytemplate[0]
print printstring, fulltime
elif time == '' or time == 'today' :
printstring = sentence.sentence(google_result['current_conditions']['condition'], time)
print printstring, fulltime, google_result['current_conditions']['humidity'], google_result['current_conditions']['wind_condition']
else :
if time == 'tomorrow':
printstring = sentence.sentence(google_result['forecasts'][1]['condition'], time)
print printstring, fulltime
else:
found = False
for i in range(4):
if google_result['forecasts'][i]['day_of_week'] == time:
printstring = sentence.sentence(google_result['forecasts'][i]['condition'], time)
print printstring, "on", fulltime
found = True
if not found:
print "Forecast for " + time + " is not available currently."
else:
print 'What\'s the location?'
#end of outermost while loop
print 'ending the program...'
print 'bye!'
while True:
lookup_word = input("Word: ")
print()
result = dictionary.meaning(lookup_word)
if result != None:
# definition
for field in result:
cprint(field, "red", "on_white")
print()
for count, meaning in enumerate(result[field]):
print("{:>4}. {}".format(count + 1, meaning))
print("\n")
# sentenc
cprint("Sentence", "red", "on_white")
print()
sentence(lookup_word)
else:
print("no words match [{}]".format(lookup_word))
print("_" * 105, "\n")
def makeQuery(self, n, sentences, idfs):
scored_words = self.TF_IDF(sentences, idfs)
best_words = self.getBestWords(n, scored_words)
return sentence.sentence("query", best_words, [])
def processFile(file_name):
# read file from provided folder path
#f = open(file_name,'r')
text_0 =file_name
#print 'Text is :-'
# replace all types of quotations by normal quotes
text_1 = re.sub("\n"," ",text_0)
text_1 = re.sub("\"","\"",text_1)
text_1 = re.sub("''","\"",text_1)
text_1 = re.sub("``","\"",text_1)
text_1 = re.sub(" +"," ",text_1)
text_1 = text_1.replace("<TEXT>","")
#print text_1
global article
article = text_1
#print 'Keywords in article are : '
#print keywords(text_1)
global data1
data1=keywords(text_1)
#data1=data1.encode('ascii','ignore')
data1=data1.replace('\n'," ")
#print type(data1)
keyword_reges1=re.compile(r'[\S]+')
data1= keyword_reges1.findall(data1)
article_low = text_1.lower()
#print article_low
art_occ = 0
for x in range(0,len(data1)):
#print data1[x],'-',article_low.count(data1[x])
art_occ = art_occ + article_low.count(data1[x])
# Testing
print ("Total Occurences of Keywords in Artice : ")
print (art_occ)
global occ
occ = art_occ
art_occ = 0
print ('No of words in articles are : ')
print (len(text_1.split()))
print ('No of keywords in articles are : ')
print (len(data1))
# segment data into a list of sentences
sentence_token = nltk.data.load('tokenizers/punkt/english.pickle')
lines = sentence_token.tokenize(text_1.strip())
# setting the stemmer
sentences = []
porter = nltk.PorterStemmer()
# modelling each sentence in file as sentence object
for line in lines:
# original words of the sentence before stemming
originalWords = line[:]
line = line.strip().lower()
# word tokenization
sent = nltk.word_tokenize(line)
# stemming words
stemmedSent = [porter.stem(word) for word in sent]
stemmedSent = filter(lambda x: x!='.'and x!='`'and x!=','and x!='?'and x!="'"
and x!='!' and x!='''"''' and x!="''" and x!="'s", stemmedSent)
# list of sentence objects
stemmedSent=list(stemmedSent)
if stemmedSent != []:
sentences.append(sentence.sentence(file_name, stemmedSent, originalWords))
return sentences
