def getLinks(url, soup):
"""
Returns a list of urls
"""
currentWords = []
listURLs = []
wordsInPage = soup.get_text()
for word in wordsInPage.split():
currentWords.append(word)
if word not in db:
db[word] = [(util.clean_words(soup.title.text), url)]
elif word not in currentWords:
db[word].append((util.clean_words(soup.title.text), url))
for link in soup.find_all("a"):
newUrl = urllib.parse.urljoin(url, link.get("href"))
listURLs.append(newUrl)
return listURLs
def search():
"""Returns the results page."""
query = util.clean_words(request.args.get("query", ""))
if query == "":
return render_template("index.html")
else:
results = moogle.answer(app.db, query)
return render_template("search.html", authors=moogle.authors(), query=query, results=results)
def sanitizeText(text):
try:
text = util.clean_words(text)
except:
text.encode('utf-8')
text = text.split(' ')
filter(None, text)
return [word for word in text if word not in STOP_WORDS]
def make_data(self,
trainfilename,
maxseqlen=None,
maxclauselen=None,
label_ind=None,
train=False):
use_attention = self.params["use_attention"]
batch_size = self.params["batch_size"]
str_seqs, label_seqs = read_passages(trainfilename, is_labeled=train)
print("Filtering data")
str_seqs = clean_words(str_seqs)
label_seqs = to_BIO(label_seqs)
if not label_ind:
self.label_ind = {"none": 0}
else:
self.label_ind = label_ind
seq_lengths = [len(seq) for seq in str_seqs]
if self.maxseqlen is None:
if maxseqlen:
self.maxseqlen = maxseqlen
elif self.params["maxseqlen"] is not None:
self.maxseqlen = self.params["maxseqlen"]
else:
self.maxseqlen = max(seq_lengths)
if self.maxclauselen is None:
if maxclauselen:
self.maxclauselen = maxclauselen
elif self.params["maxclauselen"] is not None:
self.maxclauselen = self.params["maxclauselen"]
elif use_attention:
sentence_lens = []
for str_seq in str_seqs:
for seq in str_seq:
tokens = self.tokenizer.tokenize(seq.lower())
sentence_lens.append(len(tokens))
self.maxclauselen = np.round(
np.mean(sentence_lens) +
3 * np.std(sentence_lens)).astype(int)
if len(self.label_ind) <= 1:
for str_seq, label_seq in zip(str_seqs, label_seqs):
for label in label_seq:
if label not in self.label_ind:
# Add new labels with values 0,1,2,....
self.label_ind[label] = len(self.label_ind)
self.rev_label_ind = {i: l for (l, i) in self.label_ind.items()}
discourse_generator = BertDiscourseGenerator(self.bert, self.tokenizer,
str_seqs, label_seqs,
self.label_ind,
batch_size, use_attention,
self.maxseqlen,
self.maxclauselen, train)
return seq_lengths, discourse_generator # One-hot representation of labels
def crawler_aux(db, url, maxdist):
if maxdist > 0:
try:
response = urllib.request.urlopen(url)
page = response.read()
soup = BeautifulSoup(page, "html.parser")
db[url] = (clean_words(soup.title.text),
set(
clean_words(soup.get_text() +
soup.title.text).split(" ")))
for link in soup.find_all("a"):
link_str = urllib.parse.urljoin(url, link.get("href"))
if link_str not in db and link_str[len(link_str) -
5:] == '.html':
crawler_aux(db, link_str, maxdist - 1)
except Exception as e:
print("Exception found while reading the webpage {}\n{}\n".format(
url, e))
def make_data(self,
trainfilename,
maxseqlen=None,
maxclauselen=None,
label_ind=None,
train=False):
use_attention = self.params["use_attention"]
maxseqlen = self.params["maxseqlen"]
maxclauselen = self.params["maxclauselen"]
batch_size = self.params["batch_size"]
str_seqs, label_seqs = read_passages(trainfilename, is_labeled=train)
print("Filtering data")
str_seqs = clean_words(str_seqs)
label_seqs = to_BIO(label_seqs)
if not label_ind:
self.label_ind = {"none": 0}
else:
self.label_ind = label_ind
seq_lengths = [len(seq) for seq in str_seqs]
if not maxseqlen:
maxseqlen = max(seq_lengths)
if not maxclauselen:
if use_attention:
clauselens = []
for str_seq in str_seqs:
clauselens.extend(
[len(clause.split()) for clause in str_seq])
maxclauselen = np.round(
np.mean(clauselens) + 3 * np.std(clauselens)).astype(int)
X = []
Y = []
Y_inds = []
init_word_rep_len = len(self.rep_reader.word_rep) # Vocab size
if len(self.label_ind) <= 1:
for str_seq, label_seq in zip(str_seqs, label_seqs):
for label in label_seq:
if label not in self.label_ind:
# Add new labels with values 0,1,2,....
self.label_ind[label] = len(self.label_ind)
self.rev_label_ind = {i: l for (l, i) in self.label_ind.items()}
discourse_generator = DiscourseGenerator(self.rep_reader, str_seqs,
label_seqs, self.label_ind,
batch_size, use_attention,
maxseqlen, maxclauselen,
train, self.input_size)
self.maxseqlen = maxseqlen
self.maxclauselen = maxclauselen
return seq_lengths, discourse_generator # One-hot representation of labels
def main():
parser = argparse.ArgumentParser(
description="Answer module for the µoogle project",
epilog=moogle.authors(),
)
parser.add_argument("-q", "--query", type=str, help="query (use quotes for more than one word")
parser.add_argument("-d", "--database", type=str, help="filename of the database", default="moogle.dat")
args = parser.parse_args(sys.argv[1:])
db = moogle.load(args.database)
query = util.clean_words(args.query)
answer = moogle.answer(db, query)
pprint.pprint(answer)
def calc(self, input_str, points):
# Trivial case when there are no words.
if input_str is None:
return
try:
clear = util.clean_words(
input_str) # Clean the words from "litter".
words = clear.split(" ") # Split the string into several words.
# We check for every word of the string.
for word in words:
if word == '':
continue
sum_points = self.words_weight.get(
word) # Initially set to 0, we will accumulate the score.
if word in self.CONST_STOP_WORDS: # If the word is actually a STOP WORD...
x = points * 0.1
else:
x = points
if sum_points is None:
self.words_weight[word] = x
else:
self.words_weight[word] += x
except ValueError as e:
print(e)
def crawler(url, maxdist):
"""
Crawls the web starting from url,
following up to maxdist links
and returns the built database.
"""
# Please implement this function
dict = {} # db: diccionari (paraula apareguda -> [(titol,url) de web)]
webs = [] # llista de webs a tractar
webs.append(tuple((url, 0))) # la url original, amb profunditat 0
# A webs guardarem totes les webs a guardar, amb parells (url,depth)
# Realment, no podem saber si un link es HTML o no fins que no l'obrim
# i mirem el content type. AMb l'extensio no n'hi ha prou.
# Per exemple, apple.com/education es HTML perque el servidor serveix
# el corresponent index.html
# for (web, depth) in webs:
i = 0
used_webs = set([url]) # webs ja visitades
# utilitzem while en comptes d'un iterador perque la llista es
# dinamica (es fan appends dins de l'iteracio), i llavors queda
# mes clar aixi.
while i < len(webs):
(web, depth) = webs[i]
try:
# timeout de 3, el valor per defecte es massa gran
response = urllib.request.urlopen(web, timeout=3)
content_type = response.info().get('Content-Type')
# Posem aquesta condicio en comptes de
# content_type = 'text/html', perque algunes webs
# indiquen mes coses al content type
if "html" not in content_type:
break
page = response.read()
soup = BeautifulSoup(page, "html.parser")
# Si salta alguna excepcio sera la corresponent a un error
# a l'establir la connexio (de sockets, handshake...) o
# HTTP (404 not found, 403 forbidden...)
except Exception as e:
print(e)
i += 1
continue
try:
# Com hem dit, cada paraula trobada en alguna de les
# webs tindra una entrada al diccionari amb el
# conjunt de les webs (titol,url) on aparegui
title = clean_words(soup.title.text)
text = clean_words(soup.get_text())
content = title + text
added_words = set([])
# Per totes les paraules del contingut de la web
for word in content.split():
if word in added_words:
continue
if word not in dict:
dict[word] = set([tuple((title, web))])
else:
dict[word].add(tuple((title, web)))
added_words.add(word)
# Es podria produir una excepcio si una web no tingues
# contingut, o si per algun caracter especial falles
# clean_words
except Exception as e:
print(e)
# Afegim tots els links no repetits i amb protocol hhtp
# o https
# Si ja estem a la maxima profunditat, no mirarem cap link
# Sumarem 1 a la profunditat dels fills, respecte la del pare
if depth < maxdist:
for link in soup.find_all("a"):
try:
newurl = urllib.parse.urljoin(web, link.get("href"))
(newweb, fragment) = urllib.parse.urldefrag(newurl)
if newweb.startswith("http") and newweb not in used_webs:
used_webs.add(newweb)
webs.append(tuple((newweb, depth + 1)))
# urljoin pot provocar una excepcio si la url
# esta mal formada
except Exception as e:
print(e)
continue
i += 1
return dict
def get_title(self):
return util.clean_words(self.title)
