def api_request(self, command, **kwargs):
if 'data' in kwargs:
data = to_unicode(kwargs.pop('data')).encode('utf-8', 'replace')
else:
data = None
headers = {}
if not command.startswith('applications'):
today = local2utc(datetime.now()).strftime('%Y-%m-%d')
token = sha256(self.username + self.APITOKEN + today).hexdigest()
headers['Authorization'] = 'Basic %s' % (b64encode('%s:%s' % (self.username, self.password)))
headers['X-Platform'] = 'android'
headers['X-Client-Version'] = self.APIVERSION
headers['X-AUM-Token'] = token
url = self.buildurl(self.absurl('/api/%s' % command), **kwargs)
if isinstance(url, unicode):
url = url.encode('utf-8')
req = self.request_class(url, data, headers)
buf = self.openurl(req).read()
try:
r = json.loads(buf)
except ValueError:
raise ValueError(buf)
return r
def __init__(self, browser, url, tree):
Content.__init__(self, browser)
self.url = url
self.id = url2id(self.url)
if tree is None:
return
header = tree.find("header")
self.title = u" — ".join([a.text for a in header.find("h1").xpath(".//a")])
try:
a = self.browser.parser.select(header, "a[rel=author]", 1)
except BrokenPageError:
self.author = "Anonyme"
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib["href"].split("/")[2])
self.body = self.browser.parser.tostring(self.browser.parser.select(tree, "div.content", 1))
try:
self.date = datetime.strptime(
self.browser.parser.select(header, "time", 1).attrib["datetime"].split("+")[0], "%Y-%m-%dT%H:%M:%S"
)
self.date = local2utc(self.date)
except BrokenPageError:
pass
for form in self.browser.parser.select(tree.find("footer"), "form.button_to"):
if form.attrib["action"].endswith("/for"):
self.relevance_url = form.attrib["action"].rstrip("for").rstrip("against")
self.relevance_token = self.browser.parser.select(form, "input[name=authenticity_token]", 1).attrib[
"value"
]
self.score = int(self.browser.parser.select(tree, "div.figures figure.score", 1).text)
def parse(self):
self.url = "%s#%s" % (self.preurl, self.div.attrib["id"])
self.title = unicode(self.browser.parser.select(self.div.find("h2"), "a.title", 1).text)
try:
a = self.browser.parser.select(self.div.find("p"), "a[rel=author]", 1)
except BrokenPageError:
self.author = "Anonyme"
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib["href"].split("/")[2])
self.date = datetime.strptime(
self.browser.parser.select(self.div.find("p"), "time", 1).attrib["datetime"].split("+")[0],
"%Y-%m-%dT%H:%M:%S",
)
self.date = local2utc(self.date)
content = self.div.find("div")
try:
signature = self.browser.parser.select(content, "p.signature", 1)
except BrokenPageError:
# No signature.
pass
else:
content.remove(signature)
self.signature = self.browser.parser.tostring(signature)
self.body = self.browser.parser.tostring(content)
self.score = int(self.browser.parser.select(self.div.find("p"), "span.score", 1).text)
forms = self.browser.parser.select(self.div.find("footer"), "form.button_to")
if len(forms) > 0:
self.relevance_url = forms[0].attrib["action"].rstrip("for").rstrip("against")
self.relevance_token = self.browser.parser.select(forms[0], "input[name=authenticity_token]", 1).attrib[
"value"
]
def __init__(self, browser, url, tree):
Content.__init__(self, browser)
self.url = url
self.id = url2id(self.url)
if tree is None:
return
header = tree.find('header')
self.title = u' — '.join([a.text for a in header.find('h1').xpath('.//a')])
try:
a = self.browser.parser.select(header, 'a[rel=author]', 1)
except BrokenPageError:
self.author = 'Anonyme'
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib['href'].split('/')[2])
self.body = self.browser.parser.tostring(self.browser.parser.select(tree, 'div.content', 1))
try:
self.date = datetime.strptime(self.browser.parser.select(header, 'time', 1).attrib['datetime'].split('+')[0],
'%Y-%m-%dT%H:%M:%S')
self.date = local2utc(self.date)
except BrokenPageError:
pass
for form in self.browser.parser.select(tree.find('footer'), 'form.button_to'):
if form.attrib['action'].endswith('/for'):
self.relevance_url = form.attrib['action'].rstrip('for').rstrip('against')
self.relevance_token = self.browser.parser.select(form, 'input[name=authenticity_token]', 1).attrib['value']
self.score = int(self.browser.parser.select(tree, 'div.figures figure.score', 1).text)
def api_request(self, command, **kwargs):
if 'data' in kwargs:
data = to_unicode(kwargs.pop('data')).encode('utf-8', 'replace')
else:
data = None
headers = {}
if not command.startswith('applications'):
today = local2utc(datetime.now()).strftime('%Y-%m-%d')
token = sha256(self.username + self.APITOKEN + today).hexdigest()
headers['Authorization'] = 'Basic %s' % (b64encode(
'%s:%s' % (self.username, self.password)))
headers['X-Platform'] = 'android'
headers['X-Client-Version'] = self.APIVERSION
headers['X-AUM-Token'] = token
url = self.buildurl(self.absurl('/api/%s' % command), **kwargs)
if isinstance(url, unicode):
url = url.encode('utf-8')
req = self.request_class(url, data, headers)
buf = self.openurl(req).read()
try:
r = json.loads(buf)
except ValueError:
raise ValueError(buf)
return r
def __init__(self, browser, url, tree):
super(Article, self).__init__(browser)
self.url = url
self.id = url2id(self.url)
if tree is None:
return
header = tree.find('header')
self.title = u' — '.join([a.text for a in header.find('h1').xpath('.//a')])
try:
a = header.xpath('.//a[@rel="author"]')[0]
except IndexError:
self.author = 'Anonyme'
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib['href'].split('/')[2])
self.body = lxml.html.tostring(tree.xpath('.//div[has-class("content")]')[0]).decode('utf-8')
try:
self.date = datetime.strptime(header.xpath('.//time')[0].attrib['datetime'].split('+')[0],
'%Y-%m-%dT%H:%M:%S')
self.date = local2utc(self.date)
except IndexError:
pass
for form in tree.find('footer').xpath('//form[has-class("button_to")]'):
if form.attrib['action'].endswith('/for'):
self.relevance_url = form.attrib['action'].rstrip('for').rstrip('against')
self.relevance_token = form.xpath('.//input[@name="authenticity_token"]')[0].attrib['value']
self.score = int(tree.xpath('.//div[has-class("figures")]//figure[has-class("score")]')[0].text)
def iter_threads(self):
for thread in self.browser.get_threads():
t = Thread(thread['id'])
t.flags = Thread.IS_DISCUSSION
t.title = u'Discussion with %s' % thread['name']
t.date = local2utc(datetime.datetime.fromtimestamp(thread['last_message']['utc_timestamp']))
yield t
def parse(self):
self.url = '%s#%s' % (self.preurl, self.div.attrib['id'])
self.title = unicode(self.browser.parser.select(self.div.find('h2'), 'a.title', 1).text)
try:
a = self.browser.parser.select(self.div.find('p'), 'a[rel=author]', 1)
except BrokenPageError:
self.author = 'Anonyme'
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib['href'].split('/')[2])
self.date = datetime.strptime(self.browser.parser.select(self.div.find('p'), 'time', 1).attrib['datetime'].split('+')[0],
'%Y-%m-%dT%H:%M:%S')
self.date = local2utc(self.date)
content = self.div.find('div')
try:
signature = self.browser.parser.select(content, 'p.signature', 1)
except BrokenPageError:
# No signature.
pass
else:
content.remove(signature)
self.signature = self.browser.parser.tostring(signature)
self.body = self.browser.parser.tostring(content)
self.score = int(self.browser.parser.select(self.div.find('p'), 'span.score', 1).text)
forms = self.browser.parser.select(self.div.find('footer'), 'form.button_to')
if len(forms) > 0:
self.relevance_url = forms[0].attrib['action'].rstrip('for').rstrip('against')
self.relevance_token = self.browser.parser.select(forms[0], 'input[name=authenticity_token]', 1).attrib['value']
def iter_threads(self):
for thread in self.browser.get_threads():
t = Thread(thread['id'])
t.flags = Thread.IS_DISCUSSION
t.title = u'Discussion with %s' % thread['name']
t.date = local2utc(datetime.datetime.fromtimestamp(thread['last_message']['utc_timestamp']))
yield t
def parse(self):
self.url = '%s#%s' % (self.preurl, self.div.attrib['id'])
self.title = unicode(self.browser.parser.select(self.div.find('h2'), 'a.title', 1).text)
try:
a = self.browser.parser.select(self.div.find('p'), 'a[rel=author]', 1)
except BrokenPageError:
self.author = 'Anonyme'
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib['href'].split('/')[2])
self.date = datetime.strptime(self.browser.parser.select(self.div.find('p'), 'time', 1).attrib['datetime'].split('+')[0],
'%Y-%m-%dT%H:%M:%S')
self.date = local2utc(self.date)
content = self.div.find('div')
try:
signature = self.browser.parser.select(content, 'p.signature', 1)
except BrokenPageError:
# No signature.
pass
else:
content.remove(signature)
self.signature = self.browser.parser.tostring(signature)
self.body = self.browser.parser.tostring(content)
self.score = int(self.browser.parser.select(self.div.find('p'), 'span.score', 1).text)
forms = self.browser.parser.select(self.div.find('footer'), 'form.button_to')
if len(forms) > 0:
self.relevance_url = forms[0].attrib['action'].rstrip('for').rstrip('against')
self.relevance_token = self.browser.parser.select(forms[0], 'input[name=authenticity_token]', 1).attrib['value']
def __init__(self, browser, url, tree):
Content.__init__(self, browser)
self.url = url
self.id = url2id(self.url)
if tree is None:
return
header = tree.find('header')
self.title = u' — '.join([a.text for a in header.find('h1').xpath('.//a')])
try:
a = self.browser.parser.select(header, 'a[rel=author]', 1)
except BrokenPageError:
self.author = 'Anonyme'
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib['href'].split('/')[2])
self.body = self.browser.parser.tostring(self.browser.parser.select(tree, 'div.content', 1))
try:
self.date = datetime.strptime(self.browser.parser.select(header, 'time', 1).attrib['datetime'].split('+')[0],
'%Y-%m-%dT%H:%M:%S')
self.date = local2utc(self.date)
except BrokenPageError:
pass
for form in self.browser.parser.select(tree.find('footer'), 'form.button_to'):
if form.attrib['action'].endswith('/for'):
self.relevance_url = form.attrib['action'].rstrip('for').rstrip('against')
self.relevance_token = self.browser.parser.select(form, 'input[name=authenticity_token]', 1).attrib['value']
self.score = int(self.browser.parser.select(tree, 'div.figures figure.score', 1).text)
def get_thread_mails(self):
mails = {
'member' : {},
'messages' : [],
}
try:
mails['member']['pseudo'] = self.parser.tocleanstring(self.document.getroot().cssselect('div#message_heading div.username span.name')[0])
except IndexError:
mails['member']['pseudo'] = 'Unknown'
for li in reversed(self.document.xpath('//ul[@id="thread"]//li[contains(@id, "message_")]')):
try:
txt = self.parser.tostring(li.xpath('.//div[@class="message_body"]')[0])
except IndexError:
continue # 'Match' message
txt = html2text(txt).strip()
m = re.search(r'(\d+), ', li.xpath('.//span[@class="timestamp"]//script')[0].text)
assert m
date = local2utc(datetime.fromtimestamp(int(m.group(1))))
id_from = li.find('a').attrib['href'].split('/')[-1].split('?')[0]
mails['messages'].append({
'date' : date,
'message' : unicode(txt),
'id_from' : unicode(id_from),
})
return mails
def parse(self):
self.url = '%s#%s' % (self.preurl, self.div.attrib['id'])
self.title = unicode(self.div.find('h2').xpath('.//a[has-class("title")]')[0].text)
try:
a = self.div.find('p').xpath('.//a[@rel="author"]')[0]
except IndexError:
self.author = 'Anonyme'
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib['href'].split('/')[2])
self.date = datetime.strptime(self.div.find('p').xpath('.//time')[0].attrib['datetime'].split('+')[0],
'%Y-%m-%dT%H:%M:%S')
self.date = local2utc(self.date)
content = self.div.find('div')
try:
signature = content.xpath('.//p[has-class("signature")]')[0]
except IndexError:
# No signature.
pass
else:
content.remove(signature)
self.signature = lxml.html.tostring(signature).decode('utf-8')
self.body = lxml.html.tostring(content).decode('utf-8')
self.score = int(self.div.find('p').xpath('.//span[has-class("score")]')[0].text)
forms = self.div.find('footer').xpath('.//form[has-class("button_to")]')
if len(forms) > 0:
self.relevance_url = forms[0].attrib['action'].rstrip('for').rstrip('against')
self.relevance_token = forms[0].xpath('.//input[@name="authenticity_token"]')[0].attrib['value']
def iter_threads(self):
for thread in self.browser.get_threads():
t = Thread(thread['id'])
t.flags = Thread.IS_DISCUSSION
for user in thread['participants']:
if user['user']['id'] != self.browser.my_id:
t.title = u'Discussion with %s' % user['user']['display_name']
t.date = local2utc(parse_date(thread['modification_date']))
yield t
def parse_date(s):
s = s.replace(u'Fév', 'Feb') \
.replace(u'Avr', 'Apr') \
.replace(u'Mai', 'May') \
.replace(u'Juin', 'Jun') \
.replace(u'Juil', 'Jul') \
.replace(u'Aoû', 'Aug') \
.replace(u'Ao\xfbt', 'Aug') \
.replace(u'Déc', 'Dec')
return local2utc(_parse_dt(s))
def iter_threads(self):
for thread in self.browser.get_threads():
t = Thread(thread['id'])
t.flags = Thread.IS_DISCUSSION
for user in thread['participants']:
if user['user']['id'] != self.browser.my_id:
t.title = u'Discussion with %s' % user['user'][
'display_name']
t.date = local2utc(parse_date(thread['modification_date']))
yield t
def build_request(self, url, *args, **kwargs):
headers = kwargs.setdefault('headers', {})
if 'applications' not in url:
today = local2utc(datetime.now()).strftime('%Y-%m-%d')
token = sha256((self.username + self.APITOKEN + today).encode('utf-8')).hexdigest()
headers['Authorization'] = 'Basic %s' % (b64encode(b'%s:%s' % (self.username.encode('utf-8'), self.password.encode('utf-8')))).decode('utf-8')
headers['X-Platform'] = 'android'
headers['X-Client-Version'] = self.APIVERSION
headers['X-AUM-Token'] = token
return super(AuMBrowser, self).build_request(url, *args, **kwargs)
def get_thread(self, thread):
if not isinstance(thread, Thread):
thread = Thread(thread)
thread.flags = Thread.IS_DISCUSSION
user = self.browser.get_user(thread.id)
thread.title = u'Discussion with %s' % user['name']
contact = self.storage.get('contacts', thread.id, default={'lastmsg': 0})
signature = u'Age: %s' % user['age']
signature += u'\nLast online: %s' % user['last_online']
signature += u'\nPhotos:\n\t%s' % '\n\t'.join([user['photo_host'] + photo['large'] for photo in user['photos']])
child = None
for msg in self.browser.get_thread_messages(thread.id):
flags = 0
if int(contact['lastmsg']) < msg['utc_timestamp']:
flags = Message.IS_UNREAD
if msg['type'] == 'msg':
content = unicode(msg['msg'])
elif msg['type'] == 'new_challenge':
content = u'A new challenge has been proposed!'
elif msg['type'] == 'serie':
content = u"I've played"
elif msg['type'] == 'end_game':
content = u'%s is the winner! (%s VS %s)' % (self.browser.my_name if msg['score']['w'] == self.browser.my_id else user['name'], msg['score']['s'][0], msg['score']['s'][1])
else:
content = u'Unknown action: %s' % msg['type']
msg = Message(thread=thread,
id=msg['utc_timestamp'],
title=thread.title,
sender=unicode(self.browser.my_name if msg['from'] == self.browser.my_id else user['name']),
receivers=[unicode(self.browser.my_name if msg['from'] != self.browser.my_id else user['name'])],
date=local2utc(datetime.datetime.fromtimestamp(msg['utc_timestamp'])),
content=content,
children=[],
parent=None,
signature=signature if msg['from'] != self.browser.my_id else u'',
flags=flags)
if child:
msg.children.append(child)
child.parent = msg
child = msg
thread.root = child
return thread
def get_thread(self, thread):
if not isinstance(thread, Thread):
thread = Thread(thread)
thread.flags = Thread.IS_DISCUSSION
user = self.browser.get_user(thread.id)
thread.title = u'Discussion with %s' % user['name']
contact = self.storage.get('contacts', thread.id, default={'lastmsg': 0})
signature = u'Age: %s' % user['age']
signature += u'\nLast online: %s' % user['last_online']
signature += u'\nPhotos:\n\t%s' % '\n\t'.join([user['photo_host'] + photo['large'] for photo in user['photos']])
child = None
for msg in self.browser.get_thread_messages(thread.id):
flags = 0
if int(contact['lastmsg']) < msg['utc_timestamp']:
flags = Message.IS_UNREAD
if msg['type'] == 'msg':
content = unicode(msg['msg'])
elif msg['type'] == 'new_challenge':
content = u'A new challenge has been proposed!'
elif msg['type'] == 'serie':
content = u"I've played"
elif msg['type'] == 'end_game':
content = u'%s is the winner! (%s VS %s)' % (self.browser.my_name if msg['score']['w'] == self.browser.my_id else user['name'], msg['score']['s'][0], msg['score']['s'][1])
else:
content = u'Unknown action: %s' % msg['type']
msg = Message(thread=thread,
id=msg['utc_timestamp'],
title=thread.title,
sender=unicode(self.browser.my_name if msg['from'] == self.browser.my_id else user['name']),
receivers=[unicode(self.browser.my_name if msg['from'] != self.browser.my_id else user['name'])],
date=local2utc(datetime.datetime.fromtimestamp(msg['utc_timestamp'])),
content=content,
children=[],
parent=None,
signature=signature if msg['from'] != self.browser.my_id else u'',
flags=flags)
if child:
msg.children.append(child)
child.parent = msg
child = msg
thread.root = child
return thread
def get_thread(self, thread):
if not isinstance(thread, Thread):
thread = Thread(thread)
thread.flags = Thread.IS_DISCUSSION
info = self.browser.get_thread(thread.id)
for user in info['participants']:
if user['user']['id'] == self.browser.my_id:
me = HappnContact(user['user'])
else:
other = HappnContact(user['user'])
thread.title = u'Discussion with %s' % other.name
contact = self.storage.get('contacts',
thread.id,
default={'lastmsg': 0})
child = None
for msg in info['messages']:
flags = 0
if int(contact['lastmsg']) < int(msg['id']):
flags = Message.IS_UNREAD
if msg['sender']['id'] == me.id:
sender = me
receiver = other
else:
sender = other
receiver = me
msg = Message(thread=thread,
id=msg['id'],
title=thread.title,
sender=sender.name,
receivers=[receiver.name],
date=local2utc(parse_date(msg['creation_date'])),
content=msg['message'],
children=[],
parent=None,
signature=sender.get_text(),
flags=flags)
if child:
msg.children.append(child)
child.parent = msg
child = msg
thread.root = child
return thread
def parse_dt(s):
now = datetime.datetime.now()
if s is None:
return local2utc(now)
if 'minutes ago' in s:
m = int(s.split()[0])
d = now - datetime.timedelta(minutes=m)
elif u'–' in s:
# Date in form : "Yesterday – 20:45"
day, hour = s.split(u'–')
day = day.strip()
hour = hour.strip()
if day == 'Yesterday':
d = now - datetime.timedelta(days=1)
elif day == 'Today':
d = now
hour = _parse_dt(hour)
d = datetime.datetime(d.year, d.month, d.day, hour.hour, hour.minute)
else:
#if ',' in s:
# Date in form : "Dec 28, 2011")
d = _parse_dt(s)
return local2utc(d)
def parse_dt(s):
now = datetime.datetime.now()
if s is None:
return local2utc(now)
if 'minutes ago' in s:
m = int(s.split()[0])
d = now - datetime.timedelta(minutes=m)
elif u'–' in s:
# Date in form : "Yesterday – 20:45"
day, hour = s.split(u'–')
day = day.strip()
hour = hour.strip()
if day == 'Yesterday':
d = now - datetime.timedelta(days=1)
elif day == 'Today':
d = now
hour = _parse_dt(hour)
d = datetime.datetime(d.year, d.month, d.day, hour.hour, hour.minute)
else:
#if ',' in s:
# Date in form : "Dec 28, 2011")
d = _parse_dt(s)
return local2utc(d)
def get_thread(self, thread):
if not isinstance(thread, Thread):
thread = Thread(thread)
thread.flags = Thread.IS_DISCUSSION
info = self.browser.get_thread(thread.id)
for user in info['participants']:
if user['user']['id'] == self.browser.my_id:
me = HappnContact(user['user'])
else:
other = HappnContact(user['user'])
thread.title = u'Discussion with %s' % other.name
contact = self.storage.get('contacts', thread.id, default={'lastmsg': 0})
child = None
for msg in info['messages']:
flags = 0
if int(contact['lastmsg']) < int(msg['id']):
flags = Message.IS_UNREAD
if msg['sender']['id'] == me.id:
sender = me
receiver = other
else:
sender = other
receiver = me
msg = Message(thread=thread,
id=msg['id'],
title=thread.title,
sender=sender.name,
receivers=[receiver.name],
date=local2utc(parse_date(msg['creation_date'])),
content=msg['message'],
children=[],
parent=None,
signature=sender.get_text(),
flags=flags)
if child:
msg.children.append(child)
child.parent = msg
child = msg
thread.root = child
return thread
def __init__(self, browser, url, tree):
super(Article, self).__init__(browser)
self.url = url
self.id = url2id(self.url)
if tree is None:
return
header = tree.find('header')
self.title = u' — '.join(
[a.text for a in header.find('h1').xpath('.//a')])
try:
a = header.xpath('.//a[@rel="author"]')[0]
except IndexError:
self.author = 'Anonyme'
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib['href'].split('/')[2])
self.body = lxml.html.tostring(
tree.xpath('.//div[has-class("content")]')[0]).decode('utf-8')
try:
self.date = datetime.strptime(
header.xpath('.//time')[0].attrib['datetime'].split('+')[0],
'%Y-%m-%dT%H:%M:%S')
self.date = local2utc(self.date)
except IndexError:
pass
for form in tree.find('footer').xpath(
'//form[has-class("button_to")]'):
if form.attrib['action'].endswith('/for'):
self.relevance_url = form.attrib['action'].rstrip(
'for').rstrip('against')
self.relevance_token = form.xpath(
'.//input[@name="authenticity_token"]')[0].attrib['value']
self.score = int(
tree.xpath(
'.//div[has-class("figures")]//figure[has-class("score")]')
[0].text)
def get_thread_mails(self):
mails = {
'member': {},
'messages': [],
}
try:
mails['member']['pseudo'] = self.parser.tocleanstring(
self.document.getroot().cssselect(
'#message_heading div.username span.name')[0])
except IndexError:
mails['member']['pseudo'] = 'Unknown'
for li in reversed(
self.document.xpath(
'//ul[@id="thread"]//li[contains(@id, "message_")]')):
try:
txt = self.parser.tostring(
li.xpath('.//div[@class="message_body"]')[0])
except IndexError:
continue # 'Match' message
txt = html2text(txt).strip()
m = re.search(
r'(\d+), ',
li.xpath('.//span[@class="timestamp"]//script')[0].text)
assert m
date = local2utc(datetime.fromtimestamp(int(m.group(1))))
id_from = li.find('a').attrib['href'].split('/')[-1].split('?')[0]
mails['messages'].append({
'date': date,
'message': unicode(txt),
'id_from': unicode(id_from),
})
return mails
def parse(self):
self.url = '%s#%s' % (self.preurl, self.div.attrib['id'])
self.title = unicode(
self.div.find('h2').xpath('.//a[has-class("title")]')[0].text)
try:
a = self.div.find('p').xpath('.//a[@rel="author"]')[0]
except IndexError:
self.author = 'Anonyme'
self.username = None
else:
self.author = unicode(a.text)
self.username = unicode(a.attrib['href'].split('/')[2])
self.date = datetime.strptime(
self.div.find('p').xpath('.//time')[0].attrib['datetime'].split(
'+')[0], '%Y-%m-%dT%H:%M:%S')
self.date = local2utc(self.date)
content = self.div.find('div')
try:
signature = content.xpath('.//p[has-class("signature")]')[0]
except IndexError:
# No signature.
pass
else:
content.remove(signature)
self.signature = lxml.html.tostring(signature).decode('utf-8')
self.body = lxml.html.tostring(content).decode('utf-8')
self.score = int(
self.div.find('p').xpath('.//span[has-class("score")]')[0].text)
forms = self.div.find('footer').xpath(
'.//form[has-class("button_to")]')
if len(forms) > 0:
self.relevance_url = forms[0].attrib['action'].rstrip(
'for').rstrip('against')
self.relevance_token = forms[0].xpath(
'.//input[@name="authenticity_token"]')[0].attrib['value']
def parse_dt(s):
d = _parse_dt(s)
return local2utc(d)
def parse_dt(s):
d = _parse_dt(s)
return local2utc(d)