def __init__(self, **kwargs):
super(Statistic, self).__init__(**kwargs)
if not self.name or not self.url:
raise InitModuleException()
match = re.search(r'\b[0-9]{4}\b', self.key)
if not match:
raise InitModuleException('Not found year')
self.year = int(match.group())
def __init__(self, **kwargs):
super().__init__(**kwargs)
cid = self.key
if ':' in cid:
cid, api = cid.split(':', 1)
self.api_key = api.split(':') if ':' in api else API_KEYS[api]
else:
self.api_key = DEFAULT_API_KEY
if not re.match('^[0-9]+$', cid):
raise InitModuleException(f'Contest id {cid} should be number')
self.cid = cid
def get_standings(self, users=None, statistics=None):
if 'hashcode_scoreboard' in self.info or re.search(r'\bhash.*code\b.*\(round|final\)$', self.name, re.I):
ret = self._hashcode(users, statistics)
elif '/codingcompetitions.withgoogle.com/' in self.url:
ret = self._api_get_standings(users, statistics)
elif '/code.google.com/' in self.url or '/codejam.withgoogle.com/' in self.url:
ret = self._old_get_standings(users)
else:
raise InitModuleException(f'url = {self.url}')
if re.search(r'\bfinal\S*(?:\s+round)?$', self.name, re.I):
ret['options'] = {'medals': [{'name': name, 'count': 1} for name in ('gold', 'silver', 'bronze')]}
return ret
def __init__(self, **kwargs):
super().__init__(**kwargs)
self.is_spectator_ranklist = self.standings_url and 'spectator/ranklist' in self.standings_url
if self.is_spectator_ranklist:
return
cid = self.key
if ':' in cid:
cid, api = cid.split(':', 1)
self.api_key = api.split(':') if ':' in api else API_KEYS[api]
else:
self.api_key = DEFAULT_API_KEY
if not re.match('^[0-9]+$', cid):
raise InitModuleException(f'Contest id {cid} should be number')
self.cid = cid
def __init__(self, **kwargs):
super(Statistic, self).__init__(**kwargs)
if not self.standings_url:
url = 'http://usaco.org/index.php?page=contests'
page = REQ.get(url)
matches = re.finditer(
'<a[^>]*href="(?P<url>[^"]*)"[^>]*>(?P<name>[^<]*[0-9]{4}[^<]*Results)</a>',
page)
month = self.start_time.strftime('%B').lower()
prev_standings_url = None
for match in matches:
name = match.group('name').lower()
if (month in name or self.name.lower() in name) and str(
self.start_time.year) in name:
self.standings_url = urllib.parse.urljoin(
url, match.group('url'))
break
if (month in name or self.name.lower()
in name) and str(self.start_time.year - 1) in name:
prev_standings_url = urllib.parse.urljoin(
url, match.group('url'))
else:
if prev_standings_url is not None:
pred_standings_url = re.sub(
'[0-9]+', lambda m: str(int(m.group(0)) + 1),
prev_standings_url)
url = 'http://usaco.org/'
page = REQ.get(url)
matches = re.finditer(
'<a[^>]*href="?(?P<url>[^"]*)"?[^>]*>here</a>', page)
for match in matches:
standings_url = urllib.parse.urljoin(
url, match.group('url'))
if standings_url == pred_standings_url:
self.standings_url = standings_url
break
if not self.standings_url:
raise InitModuleException(
f'Not found standings url with'
f'month = {month}, '
f'year = {self.start_time.year}, '
f'name = {self.name}')
def get_standings(self, users=None, statistics=None):
result = {}
start_time = self.start_time.replace(tzinfo=None)
if not self.standings_url and datetime.now() - start_time < timedelta(
days=30):
re_round_overview = re.compile(
r'''
(?:<td[^>]*>
(?:
[^<]*<a[^>]*href="(?P<url>[^"]*/stat[^"]*rd=(?P<rd>[0-9]+)[^"]*)"[^>]*>(?P<title>[^<]*)</a>[^<]*|
(?P<date>[0-9]+\.[0-9]+\.[0-9]+)
)</td>[^<]*
){2}
''',
re.VERBOSE,
)
for url in [
'https://www.topcoder.com/tc?module=MatchList&nr=100500',
'https://community.topcoder.com/longcontest/stats/?module=MatchList&nr=100500',
]:
page = REQ.get(url)
matches = re_round_overview.finditer(str(page))
opt = 0.61803398875
for match in matches:
date = datetime.strptime(match.group('date'), '%m.%d.%Y')
if abs(date - start_time) < timedelta(days=2):
title = match.group('title')
intersection = len(
set(title.split()) & set(self.name.split()))
union = len(
set(title.split()) | set(self.name.split()))
iou = intersection / union
if iou > opt:
opt = iou
self.standings_url = urljoin(
url, match.group('url'))
if not self.standings_url:
raise InitModuleException('Not set standings url for %s' %
self.name)
url = self.standings_url + '&nr=100000042'
page = REQ.get(url)
result_urls = re.findall(
r'<a[^>]*href="(?P<url>[^"]*)"[^>]*>Results</a>', str(page), re.I)
if not result_urls: # marathon match
match = re.search(
'<[^>]*>Problem:[^<]*<a[^>]*href="(?P<href>[^"]*)"[^>]*>(?P<name>[^<]*)<',
page)
problem_name = match.group('name').strip()
problems_info = [{
'short':
problem_name,
'url':
urljoin(url,
match.group('href').replace('&', '&'))
}]
rows = etree.HTML(page).xpath(
"//table[contains(@class, 'stat')]//tr")
header = None
for row in rows:
r = parsed_table.ParsedTableRow(row)
if len(r.columns) < 8:
continue
values = [
c.value.strip().replace(u'\xa0', '') for c in r.columns
]
if header is None:
header = values
continue
d = OrderedDict(list(zip(header, values)))
handle = d.pop('Handle').strip()
d = self._dict_as_number(d)
if 'rank' not in d or users and handle not in users:
continue
row = result.setdefault(handle, OrderedDict())
row.update(d)
score = row.pop('final_score' if 'final_score' in
row else 'provisional_score')
row['member'] = handle
row['place'] = row.pop('rank')
row['solving'] = score
row['solved'] = {'solving': 1 if score > 0 else 0}
problems = row.setdefault('problems', {})
problem = problems.setdefault(problem_name, {})
problem['result'] = score
history_index = values.index('submission history')
if history_index:
column = r.columns[history_index]
href = column.node.xpath('a/@href')
if href:
problem['url'] = urljoin(url, href[0])
else: # single round match
matches = re.finditer('<table[^>]*>.*?</table>', page, re.DOTALL)
problems_sets = []
for match in matches:
problems = re.findall(
'<a[^>]*href="(?P<href>[^"]*c=problem_statement[^"]*)"[^>]*>(?P<name>[^/]*)</a>',
match.group(),
re.IGNORECASE,
)
if problems:
problems_sets.append([{
'short': n,
'url': urljoin(url, u)
} for u, n in problems])
problems_info = dict() if len(problems_sets) > 1 else list()
for problems_set, result_url in zip(problems_sets, result_urls):
url = urljoin(self.standings_url,
result_url + '&em=1000000042')
url = url.replace('&', '&')
division = int(parse_qs(url)['dn'][0])
for p in problems_set:
d = problems_info
if len(problems_sets) > 1:
d = d.setdefault('division', OrderedDict())
d = d.setdefault('I' * division, [])
d.append(p)
page = REQ.get(url)
rows = etree.HTML(page).xpath("//tr[@valign='middle']")
header = None
url_infos = []
for row in rows:
r = parsed_table.ParsedTableRow(row)
if len(r.columns) < 10:
continue
values = [c.value for c in r.columns]
if header is None:
header = values
continue
d = OrderedDict(list(zip(header, values)))
handle = d.pop('Coders').strip()
d = self._dict_as_number(d)
if 'division_placed' not in d or users and handle not in users:
continue
row = result.setdefault(handle, OrderedDict())
row.update(d)
if not row.get('new_rating') and not row.get(
'old_rating') and not row.get('rating_change'):
row.pop('new_rating', None)
row.pop('old_rating', None)
row.pop('rating_change', None)
row['member'] = handle
row['place'] = row.pop('division_placed')
row['solving'] = row['point_total']
row['solved'] = {'solving': 0}
row['division'] = 'I' * division
if 'adv.' in row:
row['advanced'] = row.pop('adv.').lower().startswith(
'y')
url_info = urljoin(url,
r.columns[0].node.xpath('a/@href')[0])
url_infos.append(url_info)
def fetch_solution(url):
for i in range(2):
try:
page = REQ.get(url, time_out=60)
match = re.search(
'<td[^>]*class="problemText"[^>]*>(?P<solution>.*?)</td>',
page, re.DOTALL | re.IGNORECASE)
ret = html.unescape(match.group('solution'))
ret = ret.strip()
ret = ret.replace('<BR>', '\n')
ret = ret.replace('\xa0', ' ')
return ret
except FailOnGetResponse:
sleep(i * 10 + 3)
return None
def fetch_info(url):
delay = 3
for _ in range(5):
try:
page = REQ.get(url)
break
except Exception:
sleep(delay)
delay *= 2
else:
return None, None, None
match = re.search(
'class="coderBrackets">.*?<a[^>]*>(?P<handle>[^<]*)</a>',
page, re.IGNORECASE)
handle = html.unescape(match.group('handle').strip())
match = re.search(r' Room\s*(?P<room>[0-9]+)', page)
room = match.group('room') if match else None
matches = re.finditer(
r'''
<td[^>]*>[^<]*<a[^>]*href="(?P<url>[^"]*c=problem_solution[^"]*)"[^>]*>(?P<short>[^<]*)</a>[^<]*</td>[^<]*
<td[^>]*>[^<]*</td>[^<]*
<td[^>]*>[^<]*</td>[^<]*
<td[^>]*>(?P<time>[^<]*)</td>[^<]*
<td[^>]*>(?P<status>[^<]*)</td>[^<]*
<td[^>]*>(?P<result>[^<]*)</td>[^<]*
''', page, re.VERBOSE | re.IGNORECASE)
problems = {}
n_fetch_solution = 0
for match in matches:
d = match.groupdict()
short = d.pop('short')
solution_url = urljoin(url, d['url'])
d['url'] = solution_url
d = self._dict_as_number(d)
if d['status'] in [
'Challenge Succeeded', 'Failed System Test'
]:
d['result'] = -d['result']
if abs(d['result']) < 1e-9:
d.pop('result')
if re.match('^[0.:]+$', d['time']):
d.pop('time')
solution = (statistics or {}).get(handle, {}).get(
'problems', {}).get(short, {}).get('solution')
if not solution:
n_fetch_solution += 1
solution = fetch_solution(solution_url)
d['solution'] = solution
problems[short] = d
challenges = []
matches = re.finditer(
r'''
<td[^>]*>[^<]*<a[^>]*href="[^"]*module=MemberProfile[^"]*"[^>]*>(?P<target>[^<]*)</a>[^<]*</td>[^<]*
<td[^>]*>(?P<problem>[^<]*)</td>[^<]*
<td[^>]*>(?P<status>[^<]*)</td>[^<]*
<td[^>]*>(?P<time>[^<]*)</td>[^<]*
<td[^>]*>(?P<result>[^<]*)</td>[^<]*
<td[^>]*>[^<]*<a[^>]*href="(?P<url>[^"]*)"[^>]*>\s*details\s*</a>[^<]*</td>[^<]*
''', page, re.VERBOSE | re.IGNORECASE)
for match in matches:
d = match.groupdict()
d = {k: v.strip() for k, v in d.items()}
d['result'] = float(d['result'].replace(',', '.'))
d['url'] = urljoin(url, d['url'])
p = problems.setdefault(d['problem'], {})
p.setdefault('extra_score', 0)
p['extra_score'] += d['result']
p.setdefault(
'extra_info',
[]).append(f'{d["target"]}: {d["result"]}')
challenges.append(d)
return url, handle, room, problems, challenges, n_fetch_solution
with PoolExecutor(max_workers=20) as executor, tqdm.tqdm(
total=len(url_infos)) as pbar:
n_fetch_solution = 0
for url, handle, room, problems, challenges, n_sol in executor.map(
fetch_info, url_infos):
n_fetch_solution += n_sol
pbar.set_description(f'div{division} {url}')
pbar.set_postfix(n_solution=n_fetch_solution)
pbar.update()
if handle is not None:
if handle not in result:
LOG.error(
f'{handle} not in result, url = {url}')
result[handle]['url'] = url
if room:
result[handle]['room'] = room
result[handle]['problems'] = problems
result[handle]['challenges'] = challenges
for p in problems.values():
if p.get('result', 0) > 1e-9:
result[handle]['solved']['solving'] += 1
if challenges:
h = result[handle].setdefault(
'hack', {
'title': 'challenges',
'successful': 0,
'unsuccessful': 0,
})
for c in challenges:
h['successful' if c['status'].lower() ==
'yes' else 'unsuccessful'] += 1
standings = {
'result': result,
'url': self.standings_url,
'problems': problems_info,
'options': {
'fixed_fields': [('hack', 'Challenges')],
},
}
if re.search(r'\bfinals?(?:\s+rounds?)?$', self.name, re.I):
standings['options']['medals'] = [{
'name': name,
'count': 1
} for name in ('gold', 'silver', 'bronze')]
return standings
def __init__(self, **kwargs):
super(Statistic, self).__init__(**kwargs)
if not self.name or not self.start_time or not self.url:
raise InitModuleException()
def __init__(self, **kwargs):
super(Statistic, self).__init__(**kwargs)
if not self.url.startswith(
'http://stats.ioinformatics.org/olympiads/'):
raise InitModuleException(
f'Url = {self.url} should be from stats.ioinformatics.org')
def __init__(self, **kwargs):
super(Statistic, self).__init__(**kwargs)
if not self.standings_url:
raise InitModuleException('Not set standings url for %s' %
self.name)
def get_standings(self, users=None, statistics=None):
result = {}
hidden_fields = []
fields_types = {}
order = None
writers = defaultdict(int)
start_time = self.start_time.replace(tzinfo=None)
if not self.standings_url and datetime.now() - start_time < timedelta(days=30):
opt = 0.61803398875
def canonize_title(value):
value = value.lower()
value = re.sub(r'\s+-[^-]+$', '', value)
value = re.sub(r'\bsingle\s+round\s+match\b', 'srm', value)
value = re.sub(r'\bmarathon\s+match\b', 'mm', value)
value = re.sub(r'[0-9]*([0-9]{2})\s*tco(\s+)', r'tco\1\2', value)
value = re.sub(r'tco\s*[0-9]*([0-9]{2})(\s+)', r'tco\1\2', value)
value = re.sub(r'^[0-9]{2}([0-9]{2})(\s+)', r'tco\1\2', value)
return set(re.split('[^A-Za-z0-9]+', value))
def process_match(date, title, url):
nonlocal opt
if abs(date - start_time) > timedelta(days=2):
return
a1 = canonize_title(title)
a2 = canonize_title(self.name)
intersection = 0
for w1 in a1:
for w2 in a2:
if w1.isdigit() or w2.isdigit():
if w1 == w2:
intersection += 1
break
elif w1.startswith(w2) or w2.startswith(w1):
intersection += 1
break
union = len(a1) + len(a2) - intersection
iou = intersection / union
if iou > opt:
opt = iou
self.standings_url = url
url = 'https://www.topcoder.com/tc?module=MatchList&nr=100500'
page = REQ.get(url)
re_round_overview = re.compile(
r'''
(?:<td[^>]*>(?:
[^<]*<a[^>]*href="(?P<url>[^"]*/stat[^"]*rd=(?P<rd>[0-9]+)[^"]*)"[^>]*>(?P<title>[^<]*)</a>[^<]*|
(?P<date>[0-9]+\.[0-9]+\.[0-9]+)
)</td>[^<]*){2}
''',
re.VERBOSE,
)
matches = re_round_overview.finditer(str(page))
for match in matches:
date = datetime.strptime(match.group('date'), '%m.%d.%Y')
process_match(date, match.group('title'), urljoin(url, match.group('url')))
url = 'https://www.topcoder.com/tc?module=BasicData&c=dd_round_list'
page = REQ.get(url)
root = ET.fromstring(page)
for child in root:
data = {}
for field in child:
data[field.tag] = field.text
date = dateutil.parser.parse(data['date'])
url = 'https://www.topcoder.com/stat?c=round_overview&er=5&rd=' + data['round_id']
process_match(date, data['full_name'], url)
for url in self.url, self.standings_url:
if url:
match = re.search('/challenges/(?P<cid>[0-9]+)', url)
if match:
challenge_id = match.group('cid')
break
else:
challenge_id = None
if challenge_id: # marathon match
url = conf.TOPCODER_API_MM_URL_FORMAT.format(challenge_id)
page = REQ.get(url)
data = json.loads(page)
problems_info = []
hidden_fields.extend(['time', 'submits', 'style'])
fields_types = {'delta_rank': ['delta'], 'delta_score': ['delta']}
order = ['place_as_int', '-solving', 'addition__provisional_rank', '-addition__provisional_score']
for row in data:
handle = row.pop('member')
r = result.setdefault(handle, OrderedDict())
r['member'] = handle
r['place'] = row.pop('finalRank', None)
r['provisional_rank'] = row.pop('provisionalRank', None)
r['style'] = row.pop('style')
if r['place'] and r['provisional_rank']:
r['delta_rank'] = r['provisional_rank'] - r['place']
submissions = row.pop('submissions')
has_solution = False
for s in submissions:
score = s.get('finalScore')
if not score or score == '-':
if 'provisional_score' not in r:
p_score = s.pop('provisionalScore', None)
if isinstance(p_score, str):
p_score = asfloat(p_score)
if p_score is not None:
r['provisional_score'] = round(p_score, 2) if p_score >= 0 else False
r['time'] = s['created']
has_solution = True
continue
r['solving'] = score
r['solved'] = {'solving': int(score > 0)}
p_score = s.pop('provisionalScore')
if isinstance(p_score, str):
p_score = asfloat(p_score)
if p_score is not None and p_score > 0:
r['provisional_score'] = round(p_score, 2)
r['delta_score'] = round(score - p_score, 2)
r['time'] = s['created']
has_solution = True
break
if not has_solution:
continue
r['submits'] = len(submissions)
if not result:
raise ExceptionParseStandings('empty standings')
else: # single round match
if not self.standings_url:
raise InitModuleException('Not set standings url for %s' % self.name)
url = self.standings_url + '&nr=100000042'
page = REQ.get(url, time_out=100)
result_urls = re.findall(r'<a[^>]*href="(?P<url>[^"]*)"[^>]*>Results</a>', str(page), re.I)
if not result_urls:
raise ExceptionParseStandings('not found result urls')
dd_round_results = {}
match = re.search('rd=(?P<rd>[0-9]+)', url)
if match:
rd = match.group('rd')
url = f'https://www.topcoder.com/tc?module=BasicData&c=dd_round_results&rd={rd}'
try:
dd_round_results_page = REQ.get(url)
root = ET.fromstring(dd_round_results_page)
for child in root:
data = {}
for field in child:
data[field.tag] = field.text
handle = data.pop('handle')
dd_round_results[handle] = self._dict_as_number(data)
except FailOnGetResponse:
pass
hidden_fields.extend(['coding_phase', 'challenge_phase', 'system_test', 'point_total', 'room'])
matches = re.finditer('<table[^>]*>.*?</table>', page, re.DOTALL)
problems_sets = []
for match in matches:
problems = re.findall(
'<a[^>]*href="(?P<href>[^"]*c=problem_statement[^"]*)"[^>]*>(?P<name>[^/]*)</a>',
match.group(),
re.IGNORECASE,
)
if problems:
problems_sets.append([
{'short': n, 'url': urljoin(url, u)}
for u, n in problems
])
problems_info = dict() if len(problems_sets) > 1 else list()
for problems_set, result_url in zip(problems_sets, result_urls):
url = urljoin(self.standings_url, result_url + '&em=1000000042')
url = url.replace('&', '&')
division = int(parse_qs(url)['dn'][0])
division_str = 'I' * division
with PoolExecutor(max_workers=3) as executor:
def fetch_problem(p):
errors = set()
for attempt in range(3):
try:
page = REQ.get(p['url'], time_out=30)
match = re.search('<a[^>]*href="(?P<href>[^"]*module=ProblemDetail[^"]*)"[^>]*>', page)
page = REQ.get(urljoin(p['url'], match.group('href')), time_out=30)
matches = re.findall(r'<td[^>]*class="statTextBig"[^>]*>(?P<key>[^<]*)</td>\s*<td[^>]*>(?P<value>.*?)</td>', page, re.DOTALL) # noqa
for key, value in matches:
key = key.strip().rstrip(':').lower()
if key == 'categories':
tags = [t.strip().lower() for t in value.split(',')]
tags = [t for t in tags if t]
if tags:
p['tags'] = tags
elif key.startswith('writer') or key.startswith('tester'):
key = key.rstrip('s') + 's'
p[key] = re.findall('(?<=>)[^<>,]+(?=<)', value)
for w in p.get('writers', []):
writers[w] += 1
info = p.setdefault('info', {})
matches = re.finditer('<table[^>]*paddingTable2[^>]*>.*?</table>', page, re.DOTALL)
for match in matches:
html_table = match.group(0)
rows = parsed_table.ParsedTable(html_table)
for row in rows:
key, value = None, None
for k, v in row.items():
if k == "":
key = v.value
elif k and division_str in k.split():
value = v.value
if key and value:
key = re.sub(' +', '_', key.lower())
info[key] = value
if key == 'point_value':
value = toint(value) or asfloat(value)
if value is not None:
p['full_score'] = value
except Exception as e:
errors.add(f'error parse problem info {p}: {e}')
sleep(5 + attempt)
else:
errors = None
if errors:
LOG.error(errors)
return p
for p in tqdm.tqdm(executor.map(fetch_problem, problems_set), total=len(problems_set)):
d = problems_info
if len(problems_sets) > 1:
d = d.setdefault('division', OrderedDict())
d = d.setdefault(division_str, [])
d.append(p)
if not users and users is not None:
continue
page = REQ.get(url)
rows = etree.HTML(page).xpath("//tr[@valign='middle']")
header = None
url_infos = []
for row in rows:
r = parsed_table.ParsedTableRow(row)
if len(r.columns) < 10:
continue
values = [c.value for c in r.columns]
if header is None:
header = values
continue
d = OrderedDict(list(zip(header, values)))
handle = d.pop('Coders').strip()
d = self._dict_as_number(d)
if users and handle not in users:
continue
row = result.setdefault(handle, OrderedDict())
row.update(d)
if not row.get('new_rating') and not row.get('old_rating') and not row.get('rating_change'):
row.pop('new_rating', None)
row.pop('old_rating', None)
row.pop('rating_change', None)
row['member'] = handle
row['place'] = row.pop('division_placed', None)
row['solving'] = row['point_total']
row['solved'] = {'solving': 0}
row['division'] = 'I' * division
if 'adv.' in row:
row['advanced'] = row.pop('adv.').lower().startswith('y')
url_info = urljoin(url, r.columns[0].node.xpath('a/@href')[0])
url_infos.append(url_info)
def fetch_solution(url):
for i in range(2):
try:
page = REQ.get(url, time_out=60)
match = re.search('<td[^>]*class="problemText"[^>]*>(?P<solution>.*?)</td>',
page,
re.DOTALL | re.IGNORECASE)
if not match:
break
ret = html.unescape(match.group('solution'))
ret = ret.strip()
ret = ret.replace('<BR>', '\n')
ret = ret.replace('\xa0', ' ')
return ret
except FailOnGetResponse:
sleep(i * 10 + 3)
return None
n_failed_fetch_info = 0
def fetch_info(url):
nonlocal n_failed_fetch_info
if n_failed_fetch_info > 10:
return
delay = 10
for _ in range(5):
try:
page = REQ.get(url, time_out=delay)
match = re.search('class="coderBrackets">.*?<a[^>]*>(?P<handle>[^<]*)</a>',
page,
re.IGNORECASE)
if match:
break
except Exception:
sleep(delay + _)
else:
n_failed_fetch_info += 1
return
handle = html.unescape(match.group('handle').strip())
match = re.search(r' Room\s*(?P<room>[0-9]+)', page)
room = match.group('room') if match else None
matches = re.finditer(r'''
<td[^>]*>[^<]*<a[^>]*href="(?P<url>[^"]*c=problem_solution[^"]*)"[^>]*>(?P<short>[^<]*)</a>[^<]*</td>[^<]*
<td[^>]*>[^<]*</td>[^<]*
<td[^>]*>[^<]*</td>[^<]*
<td[^>]*>(?P<time>[^<]*)</td>[^<]*
<td[^>]*>(?P<status>[^<]*)</td>[^<]*
<td[^>]*>(?P<result>[^<]*)</td>[^<]*
''', page, re.VERBOSE | re.IGNORECASE)
problems = {}
n_fetch_solution = 0
for match in matches:
d = match.groupdict()
short = d.pop('short')
solution_url = urljoin(url, d['url'])
d['url'] = solution_url
d = self._dict_as_number(d)
if d['status'] in ['Challenge Succeeded', 'Failed System Test']:
d['result'] = -d['result']
if abs(d['result']) < 1e-9:
d.pop('result')
if re.match('^[0.:]+$', d['time']):
d.pop('time')
else:
time_in_seconds = 0
for t in d['time'].split(':'):
time_in_seconds = time_in_seconds * 60 + asfloat(t)
d['time_in_seconds'] = time_in_seconds
solution = (statistics or {}).get(handle, {}).get('problems', {}).get(short, {}).get('solution')
if not solution:
n_fetch_solution += 1
solution = fetch_solution(solution_url)
d['solution'] = solution
problems[short] = d
challenges = []
matches = re.finditer(r'''
<td[^>]*>[^<]*<a[^>]*href="[^"]*module=MemberProfile[^"]*"[^>]*>(?P<target>[^<]*)</a>[^<]*</td>[^<]*
<td[^>]*>(?P<problem>[^<]*)</td>[^<]*
<td[^>]*>(?P<status>[^<]*)</td>[^<]*
<td[^>]*>(?P<time>[^<]*)</td>[^<]*
<td[^>]*>(?P<result>[^<]*)</td>[^<]*
<td[^>]*>[^<]*<a[^>]*href="(?P<url>[^"]*)"[^>]*>\s*details\s*</a>[^<]*</td>[^<]*
''', page, re.VERBOSE | re.IGNORECASE)
for match in matches:
d = match.groupdict()
d = {k: v.strip() for k, v in d.items()}
d['result'] = float(d['result'].replace(',', '.'))
d['url'] = urljoin(url, d['url'])
p = problems.setdefault(d['problem'], {})
p.setdefault('extra_score', 0)
p['extra_score'] += d['result']
p.setdefault('extra_info', []).append(f'{d["target"]}: {d["result"]}')
challenges.append(d)
return url, handle, room, problems, challenges, n_fetch_solution
with PoolExecutor(max_workers=20) as executor, tqdm.tqdm(total=len(url_infos)) as pbar:
n_fetch_solution = 0
for info in executor.map(fetch_info, url_infos):
if info is None:
continue
url, handle, room, problems, challenges, n_sol = info
n_fetch_solution += n_sol
pbar.set_description(f'div{division} {url}')
pbar.set_postfix(n_solution=n_fetch_solution, n_failed_fetch_info=n_failed_fetch_info)
pbar.update()
if handle is not None:
if handle not in result:
LOG.error(f'{handle} not in result, url = {url}')
row = result[handle]
row['url'] = url
if room:
row['room'] = room
row['problems'] = problems
row['challenges'] = challenges
for p in problems.values():
if p.get('result', 0) > 1e-9:
row['solved']['solving'] += 1
if challenges:
h = row.setdefault('hack', {
'title': 'challenges',
'successful': 0,
'unsuccessful': 0,
})
for c in challenges:
h['successful' if c['status'].lower() == 'yes' else 'unsuccessful'] += 1
if dd_round_results:
fields = set()
hidden_fields_set = set(hidden_fields)
for data in result.values():
for field in data.keys():
fields.add(field)
k_mapping = {'new_vol': 'new_volatility', 'advanced': None}
for handle, data in dd_round_results.items():
if handle not in result:
continue
row = result[handle]
for k, v in data.items():
k = k_mapping.get(k, k)
if k and k not in fields:
if k in {'new_rating', 'old_rating'} and not v:
continue
row[k] = v
if k not in hidden_fields_set:
hidden_fields_set.add(k)
hidden_fields.append(k)
ks = k.split('_')
if ks[0] == 'level' and ks[-1] == 'language' and v and v.lower() != 'unspecified':
idx = {'one': 0, 'two': 1, 'three': 2}.get(ks[1], None)
d = problems_info
if len(problems_sets) > 1:
d = d['division'][row['division']]
if idx is not None and 0 <= idx < len(d) and d[idx]['short'] in row['problems']:
row['problems'][d[idx]['short']]['language'] = v
standings = {
'result': result,
'url': self.standings_url,
'problems': problems_info,
'hidden_fields': hidden_fields,
'fields_types': fields_types,
'options': {
'fixed_fields': [('hack', 'Challenges')],
},
}
if writers:
writers = [w[0] for w in sorted(writers.items(), key=lambda w: w[1], reverse=True)]
standings['writers'] = writers
if re.search(r'\bfinals?(?:\s+rounds?)?$', self.name, re.I):
standings['options']['medals'] = [{'name': name, 'count': 1} for name in ('gold', 'silver', 'bronze')]
if order:
standings['options']['order'] = order
return standings
def __init__(self, **kwargs):
super(Statistic, self).__init__(**kwargs)
if '//stats.ioinformatics.org/olympiads/' not in self.url:
raise InitModuleException(f'Url {self.url} should be contains stats.ioinformatics.org/olympiads')
def get_standings(self, users=None, statistics=None):
if '/codingcompetitions.withgoogle.com/' in self.url:
return self._api_get_standings(users, statistics)
if '/code.google.com/' in self.url or '/codejam.withgoogle.com/' in self.url:
return self._old_get_standings(users)
raise InitModuleException(f'url = {self.url}')