def get_public_problems(profile, verbose=True):
"""Extract public submissions from the given profile."""
problems = []
urlbase = ('https://www.urionlinejudge.com.br/judge/pt/profile/'
'%s/page:%s/sort:run_id/direction:asc')
refreshed = set()
i = 0
while True:
i += 1
url = urlbase % (profile, i)
# Read html or break if encounter an error page
try:
html = htmlopen(url, verbose=verbose)
except RuntimeError as ex:
if ex.args[0] == 404:
break
raise
# Process table
tbody = html.xpath('//table/tbody')[0]
transaction = []
for tr in tbody:
data = [x.text_content().strip() for x in tr]
if not data[0] or len(data) == 1:
break
# Convert some columns..
data[0] = int(data[0]) # problem profile
data[2] = _ranking(data[2]) # ranking
data[5] = float(data[5]) # time
data[6] = _todatetime(data[6]) # date
problem = Problem(*data)
transaction.append(problem)
# Add transaction to problem list
if len(transaction) == 28:
problems.extend(transaction)
elif url not in refreshed:
# Force refresh
urlopen(url, verbose=verbose, expires=120)
refreshed.add(url)
i -= 1
else:
problems.extend(transaction)
break
# Create dataframe
return pd.DataFrame(problems, columns=problem_fields)
def get_public_profile(profile, verbose=True):
"""View all non-problem related information in the public profile."""
url = ('https://www.urionlinejudge.com.br/judge/pt/profile/'
'%s/page:1/sort:run_id/direction:asc') % profile
html = htmlopen(url, verbose=verbose)
username = html.xpath('//div[@class="pb-username"]')[0].text_content()
info = html.xpath('//ul[@class="pb-information"]/li')
data = (x.text_content().strip() for x in info)
data = (x.partition(':') for x in data)
data = dict((k.lower(), v) for (k, _, v) in data)
data['username'] = username
# Make conversions
dd, mm, yyyy = map(int, data.pop('desde').split('/'))
data['date'] = datetime.date(yyyy, mm, dd)
data['ranking'] = int(data.pop('posição')[:-1])
data['solved'] = int(data.pop('resolvido'))
data['submissions'] = int(data.pop('submissões'))
data['tried'] = int(data.pop('tentado'))
data['country'] = data.pop('país')
data['university'] = data.pop('universidade').strip()
return Profile(**data)
def __htmlopen(self, url, **kwds):
if url in urlcache():
return htmlopen(url, **kwds)
else:
return htmlopen(url, session=self.session, **kwds)
