def run():
response = requests.get(URL)
soup = BeautifulSoup(response.text, "html5lib")
rows = []
player = date = status = comment = None
for tr in soup.find_all("tbody")[0].children:
row = [td for td in tr if hasattr(tr, "children")]
if player and date and status:
if len(row) == 1 and "Comment:" in row[0].contents[0].string:
Injury.objects.update_or_create(
player=player,
date=date,
comment=unicode(row[0].contents[-1]),
defaults={"status": status},
)
else:
Injury.objects.update_or_create(player=player,
date=date,
defaults={"status": status})
player = date = status = comment = None
if len(row) == 3:
name, status, datestr = [r.string for r in row]
if name != "NAME" and status != "STATUS" and datestr != "DATE":
date = get_date_yearless(datestr)
try:
player = Player.get_by_name(name)
except Player.DoesNotExist:
print(f"Couldn't find player {name}")
def run():
response = requests.get(URL)
soup = BeautifulSoup(response.text, 'html5lib')
rows = []
player = date = status = comment = None
for tr in soup.find_all('tbody')[0].children:
row = [td for td in tr if hasattr(tr, 'children')]
if player and date and status:
if len(row) == 1 and 'Comment:' in row[0].contents[0].string:
Injury.objects.update_or_create(
player=player,
date=date,
comment=unicode(row[0].contents[-1]),
defaults={ 'status': status }
)
else:
Injury.objects.update_or_create(
player=player,
date=date,
defaults={ 'status': status }
)
player = date = status = comment = None
if len(row) == 3:
name, status, datestr = [r.string for r in row]
if name != 'NAME' and status != 'STATUS' and datestr != 'DATE':
date = get_date_yearless(datestr)
try:
player = Player.get_by_name(name)
except Player.DoesNotExist:
print 'Couldn\'t find player %s' % name
def write_salaries_to_db(input_rows, date=datetime.date.today()):
return_rows = []
csvreader = reader(input_rows, delimiter=",", quotechar='"')
try:
for i, row in enumerate(csvreader):
if i != 0 and len(row) == 9: # Ignore possible empty rows
pos, name_and_pid, name, pid, rpos, salary, gameinfo, team, ppg = row
player = Player.get_by_name(name)
dksalary, _ = DKSalary.objects.get_or_create(
player=player, date=date, defaults={"salary": int(salary)})
player.dk_position = pos
player.save()
if dksalary.salary != int(salary):
print("Warning: trying to overwrite salary for %s."
" Ignoring - did not overwrite" % player)
return_rows.append(row)
except UnicodeEncodeError as e:
print(e)
return []
return return_rows
def write_salaries_to_db(input_rows, date=datetime.date.today()):
return_rows = []
csvreader = reader(input_rows, delimiter=',', quotechar='"')
try:
for i, row in enumerate(csvreader):
if i != 0 and len(row) == 6: # Ignore possible empty rows
pos, name, salary, game, ppg, team = row
player = Player.get_by_name(name)
dksalary, _ = DKSalary.objects.get_or_create(
player=player,
date=date,
defaults={ 'salary': int(salary) }
)
player.dk_position = pos
player.save()
if dksalary.salary != int(salary):
print ('Warning: trying to overwrite salary for %s.'
' Ignoring - did not overwrite' % player)
return_rows.append(row)
except UnicodeEncodeError as e:
print e
return []
return return_rows
def get_player_cached(name, player_cache):
if name in player_cache:
result = player_cache[name]
return result
else:
return Player.get_by_name(name)
def get_player_cached(name, player_cache):
if name in player_cache:
result = player_cache[name]
return result
else:
return Player.get_by_name(name)
def print_player_ownerships_timeseries(player_name, percentile=20):
"""
Used to see the difference between player ownerships in lineups in the top
@percentile and all other lineups for a single player over time.
Prints the following table:
Date | # Entries | Top Own % (#) | Bottom Own % (#) | Total Own % | Diff
FPTS | Minutes
"""
def calculate_ownerships(contest):
query = '''
SELECT c.dk_id, c.entries, rank, pg_id, sg_id, sf_id, pf_id, c_id, g_id, f_id,
util_id
FROM nba_dkcontest AS c JOIN nba_dkresult AS r ON c.id=r.contest_id
WHERE c.dk_id='%s'
''' % contest.dk_id
# Executing SQL
cursor = connection.cursor()
cursor.execute(query)
rows = [row for row in cursor.fetchall()]
if len(rows) == 0:
print 'No entries found for contest %s' % contest.dk_id
return
total_entries = rows[0][1]
# Collecting data
topct = botct = toptotal = bottotal = 0
for row in rows:
contest_id, num_entries, rank = (row[0], row[1], row[2])
player_ids = row[3:]
is_top = float(rank) * 100 <= percentile * num_entries
if is_top:
toptotal += 1
else:
bottotal += 1
for player_id in player_ids:
if player_id == player.id:
if is_top:
topct += 1
else:
botct += 1
toppct = float(topct) / toptotal * 100
botpct = float(botct) / bottotal * 100
totalpct = float(topct + botct) / (toptotal + bottotal) * 100
# The player didn't play
if topct == botct and topct == 0:
return
# Printing results
stats = player.get_stats(contest.date, 'min', 'dk_points')
print ('%s\t%d\t%.2f%% (%d)\t%.2f%% (%d)\t%.2f%%\t%.2f%%\t%.2f\t%d'
% (contest.date, toptotal + bottotal, toppct, topct, botpct,
botct, totalpct, toppct - botpct, stats['dk_points'],
stats['min']))
player = Player.get_by_name(player_name)
print ('Contest date\tEntries\tTop % (Entries)\tBot % (Entries)\tTotal %'
'\tDiff %\tFPTS\tMIN')
for date in sorted(set([d.date for d in DKContest.objects.all()])):
contest = DKContest.objects.filter(date=date).order_by('entries').last()
calculate_ownerships(contest)
def print_player_ownerships_timeseries(player_name, percentile=20):
"""
Used to see the difference between player ownerships in lineups in the top
@percentile and all other lineups for a single player over time.
Prints the following table:
Date | # Entries | Top Own % (#) | Bottom Own % (#) | Total Own % | Diff
FPTS | Minutes
"""
def calculate_ownerships(contest):
query = (
f"SELECT c.dk_id, c.entries, rank, pg_id, sg_id, sf_id, pf_id, c_id, g_id, f_id, util_id"
"FROM nba_dkcontest AS c JOIN nba_dkresult AS r ON c.id=r.contest_id"
"WHERE c.dk_id='{contest.dk_id}'")
# Executing SQL
cursor = connection.cursor()
cursor.execute(query)
rows = [row for row in cursor.fetchall()]
if len(rows) == 0:
print("No entries found for contest {contest.dk_id}")
return
total_entries = rows[0][1]
# Collecting data
topct = botct = toptotal = bottotal = 0
for row in rows:
contest_id, num_entries, rank = (row[0], row[1], row[2])
player_ids = row[3:]
is_top = float(rank) * 100 <= percentile * num_entries
if is_top:
toptotal += 1
else:
bottotal += 1
for player_id in player_ids:
if player_id == player.id:
if is_top:
topct += 1
else:
botct += 1
toppct = float(topct) / toptotal * 100
botpct = float(botct) / bottotal * 100
totalpct = float(topct + botct) / (toptotal + bottotal) * 100
# The player didn't play
if topct == botct and topct == 0:
return
# Printing results
stats = player.get_stats(contest.date, "min", "dk_points")
print(
"{}\t{}\t{:.2f}%% ({})\t{:.2f}%% ({})\t{:.2f}%%\t{:.2f}%%\t{:.2f}\t{}"
.format(
contest.date,
toptotal + bottotal,
toppct,
topct,
botpct,
botct,
totalpct,
toppct - botpct,
stats["dk_points"],
stats["min"],
))
player = Player.get_by_name(player_name)
print("Contest date\tEntries\tTop % (Entries)\tBot % (Entries)\tTotal %"
"\tDiff %\tFPTS\tMIN")
for date in sorted(set([d.date for d in DKContest.objects.all()])):
contest = DKContest.objects.filter(
date=date).order_by("entries").last()
calculate_ownerships(contest)
