def update_questions(self):
"""
Update self.questions with Question instances, which contain
text, answers, user_answer, and explanation attributes. See
the Question class in objects.py for more details. Like
User.update_questions(), note that this can take a while due to
OKCupid displaying only ten questions on each page, potentially
requiring a large number of requests to the server.
"""
count = 0
for category in ['Ethics', 'Sex', 'Religion', 'Lifestyles', 'Dating', 'Other']:
keep_going = True
while keep_going:
questions_data = {
'low': 1 + 10*count,
category: '1',
}
questions_request = self._session.post('http://www.okcupid.com/profile/{0}/questions'.format(self.name), data=questions_data)
tree = html.fromstring(questions_request.content.decode('utf8'))
for div in tree.iter('div'):
if 'id' in div.attrib and re.match(r'question_(\d+)', div.attrib['id']):
explanation = ''
number = re.match(r'question_(\d+)', div.attrib['id']).group(1)
text = helpers.replace_chars(div.xpath(".//p[@class = 'qtext']")[0].text)
answer_eles = div.xpath(".//input[contains(@id,'question_{0}_qans')]".format(number))
answers = []
for ele in answer_eles:
answers.append(ele.attrib['value'])
user_answer_ele = div.xpath(".//span[@id = 'answer_viewer_{0}']".format(number))[0]
user_answer = user_answer_ele.text.strip()
they_approve = None
if 'class' in user_answer_ele.attrib and user_answer_ele.attrib['class'] == 'not_accepted':
they_approve = False
elif len(user_answer):
they_approve = True
answer_target = div.xpath(".//span[@id = 'answer_target_{0}']".format(number))[0]
you_approve = None
if 'class' in answer_target.attrib and answer_target.attrib['class'] == 'not_accepted':
you_approve = False
elif len(user_answer):
you_approve = True
explanation = div.xpath(".//span[@id = 'note_target_{0}']".format(number))[0].text
if explanation is None:
explanation = ''
else:
explanation = helpers.replace_chars(explanation.strip())
if text not in [q.text for q in self.questions]:
self.questions.append(ProfileQuestion(text,
answers,
user_answer,
explanation,
self,
category,
you_approve,
they_approve))
next = tree.xpath("//a[text() = 'Next']")
if not len(next) or 'href' not in next[0].attrib:
keep_going = False
else:
count += 1
def update_questions(self):
"""
Update self.questions with Question instances, which contain
text, user_answer, and explanation attributes. See
the Question class in objects.py for more details. Like
User.update_questions(), note that this can take a while due to
OKCupid displaying only ten questions on each page, potentially
requiring a large number of requests to the server.
"""
keep_going = True
question_number = 0
while keep_going:
questions_data = {
'low': 1 + question_number,
}
get_questions = self._session.post(
'http://www.okcupid.com/profile/{0}/questions'.format(self.name),
data=questions_data)
tree = html.fromstring(get_questions.content.decode('utf8'))
next_wrapper = tree.xpath("//li[@class = 'next']")
question_wrappers = tree.xpath("//div[contains(@id, 'question_')]")
for div in question_wrappers:
if not div.attrib['id'][9:].isdigit():
question_wrappers.remove(div)
for div in question_wrappers:
question_number += 1
explanation = ''
text = helpers.replace_chars(div.xpath(".//div[@class = 'qtext']/p/text()")[0])
user_answer = div.xpath(".//span[contains(@id, 'answer_target_')]/text()")[0].strip()
explanation_span = div.xpath(".//span[@class = 'note']")
if explanation_span[0].text is not None:
explanation = explanation_span[0].text.strip()
self.questions.append(Question(text, user_answer, explanation))
if not len(next_wrapper):
keep_going = False
def update_questions(self):
"""
Update `self.questions` with a sequence of question objects,
whose properties can be found in objects.py. Note that this
can take a while due to OKCupid displaying only ten questions
on each page, potentially requiring a large number of requests.
"""
keep_going = True
question_number = 0
while keep_going:
questions_data = {
'low': 1 + question_number,
}
get_questions = self._session.post(
'http://www.okcupid.com/profile/{0}/questions'.format(self.username),
data=questions_data)
tree = html.fromstring(get_questions.content.decode('utf8'))
next_wrapper = tree.xpath("//li[@class = 'next']")
# Get a list of each question div wrapper, ignore the first because it's an unanswered question
question_wrappers = tree.xpath("//div[contains(@id, 'question_')]")[1:]
for div in question_wrappers:
if not div.attrib['id'][9:].isdigit():
question_wrappers.remove(div)
for div in question_wrappers:
question_number += 1
explanation = ''
text = helpers.replace_chars(div.xpath(".//div[@class = 'qtext']/p/text()")[0])
user_answer = div.xpath(".//li[contains(@class, 'mine')]/text()")[0]
explanation_p = div.xpath(".//p[@class = 'value']")
if explanation_p[0].text is not None:
explanation = explanation_p[0].text
self.questions.append(Question(text, user_answer, explanation))
if not len(next_wrapper):
keep_going = False
def update_questions(self):
"""
Update `self.questions` with a sequence of question objects,
whose properties can be found in objects.py. Note that this
can take a while due to OKCupid displaying only ten questions
on each page, potentially requiring a large number of requests.
"""
count = 0
question_number = 0
keep_going = True
while keep_going:
questions_data = {
'low': 1 + 10*count,
}
get_questions = self._session.post('http://www.okcupid.com/profile/{0}/questions'.format(self.username), data=questions_data)
time_start = time.clock()
tree = html.fromstring(get_questions.content.decode('utf8'))
for div in tree.iter('div'):
if 'id' in div.attrib and re.match(r'question_(\d+)', div.attrib['id']):
question_number += 1
explanation = ''
number = re.match(r'question_(\d+)', div.attrib['id']).group(1)
text = helpers.replace_chars(div.xpath(".//p[@class = 'qtext']")[0].text)
answer_eles = div.xpath(".//li")
answers = {}
# Use a dictionary/regex for the answer values
# because occasionally the numbers are not sequential
for ele in answer_eles:
value = re.match(r'self_answers_\d+_(\d+)', ele.attrib['id']).group(1)
answers[value] = ele.text
acceptable_answers = [ele.text for ele in answer_eles if ele.attrib['class'] in (' match', 'mine match')]
importance_no = div.xpath(".//input[@id = 'question_{0}_importance']/@value".format(number))[0]
if importance_no == '5':
importance = 'Irrelevant'
elif importance_no == '4':
importance = 'A little important'
elif importance_no == '3':
importance = 'Somewhat important'
elif importance_no == '2':
importance = 'Very important'
elif importance_no == '1':
importance = 'Mandatory'
explanation_p = div.xpath(".//p[@class = 'explanation']")
if explanation_p[0].text is not None:
explanation = explanation_p[0].text
answer_int = int(div.xpath(".//input[@id = 'question_{0}_answer']/@value".format(number))[0])
if question_number > 1 and text not in [q.text for q in self.questions]:
user_answer = answers[str(answer_int)]
self.questions.append(UserQuestion(text, answers, user_answer, explanation, self, acceptable_answers, importance))
next = tree.xpath("//a[text() = 'Next']")
if not len(next) or 'href' not in next[0].attrib:
keep_going = False
else:
count += 1
def update_questions(self):
"""
Update self.questions with Question instances, which contain
text, answers, user_answer, and explanation attributes. See
the Question class in objects.py for more details. Like
User.update_questions(), note that this can take a while due to
OKCupid displaying only ten questions on each page, potentially
requiring a large number of requests to the server.
"""
count = 0
for category in [
'Ethics', 'Sex', 'Religion', 'Lifestyles', 'Dating', 'Other'
]:
keep_going = True
while keep_going:
questions_data = {
'low': 1 + 10 * count,
category: '1',
}
questions_request = self._session.post(
'http://www.okcupid.com/profile/{0}/questions'.format(
self.name),
data=questions_data)
tree = html.fromstring(
questions_request.content.decode('utf8'))
for div in tree.iter('div'):
if 'id' in div.attrib and re.match(r'question_(\d+)',
div.attrib['id']):
explanation = ''
number = re.match(r'question_(\d+)',
div.attrib['id']).group(1)
text = helpers.replace_chars(
div.xpath(".//p[@class = 'qtext']")[0].text)
answer_eles = div.xpath(
".//input[contains(@id,'question_{0}_qans')]".
format(number))
answers = []
for ele in answer_eles:
answers.append(ele.attrib['value'])
user_answer_ele = div.xpath(
".//span[@id = 'answer_viewer_{0}']".format(
number))[0]
user_answer = user_answer_ele.text.strip()
they_approve = None
if 'class' in user_answer_ele.attrib and user_answer_ele.attrib[
'class'] == 'not_accepted':
they_approve = False
elif len(user_answer):
they_approve = True
answer_target = div.xpath(
".//span[@id = 'answer_target_{0}']".format(
number))[0]
you_approve = None
if 'class' in answer_target.attrib and answer_target.attrib[
'class'] == 'not_accepted':
you_approve = False
elif len(user_answer):
you_approve = True
explanation = div.xpath(
".//span[@id = 'note_target_{0}']".format(
number))[0].text
if explanation is None:
explanation = ''
else:
explanation = helpers.replace_chars(
explanation.strip())
if text not in [q.text for q in self.questions]:
self.questions.append(
ProfileQuestion(text, answers, user_answer,
explanation, self, category,
you_approve, they_approve))
next = tree.xpath("//a[text() = 'Next']")
if not len(next) or 'href' not in next[0].attrib:
keep_going = False
else:
count += 1
def update_questions(self):
"""
Update `self.questions` with a sequence of question objects,
whose properties can be found in objects.py. Note that this
can take a while due to OKCupid displaying only ten questions
on each page, potentially requiring a large number of requests.
"""
count = 0
question_number = 0
keep_going = True
while keep_going:
questions_data = {
'low': 1 + 10 * count,
}
get_questions = self._session.post(
'http://www.okcupid.com/profile/{0}/questions'.format(
self.username),
data=questions_data)
time_start = time.clock()
tree = html.fromstring(get_questions.content.decode('utf8'))
for div in tree.iter('div'):
if 'id' in div.attrib and re.match(r'question_(\d+)',
div.attrib['id']):
question_number += 1
explanation = ''
number = re.match(r'question_(\d+)',
div.attrib['id']).group(1)
text = helpers.replace_chars(
div.xpath(".//p[@class = 'qtext']")[0].text)
answer_eles = div.xpath(".//li")
answers = {}
# Use a dictionary/regex for the answer values
# because occasionally the numbers are not sequential
for ele in answer_eles:
value = re.match(r'self_answers_\d+_(\d+)',
ele.attrib['id']).group(1)
answers[value] = ele.text
acceptable_answers = [
ele.text for ele in answer_eles
if ele.attrib['class'] in (' match', 'mine match')
]
importance_no = div.xpath(
".//input[@id = 'question_{0}_importance']/@value".
format(number))[0]
if importance_no == '5':
importance = 'Irrelevant'
elif importance_no == '4':
importance = 'A little important'
elif importance_no == '3':
importance = 'Somewhat important'
elif importance_no == '2':
importance = 'Very important'
elif importance_no == '1':
importance = 'Mandatory'
explanation_p = div.xpath(".//p[@class = 'explanation']")
if explanation_p[0].text is not None:
explanation = explanation_p[0].text
answer_int = int(
div.xpath(
".//input[@id = 'question_{0}_answer']/@value".
format(number))[0])
if question_number > 1 and text not in [
q.text for q in self.questions
]:
user_answer = answers[str(answer_int)]
self.questions.append(
UserQuestion(text, answers, user_answer,
explanation, self, acceptable_answers,
importance))
next = tree.xpath("//a[text() = 'Next']")
if not len(next) or 'href' not in next[0].attrib:
keep_going = False
else:
count += 1
