def __init__(self,
db=None,
cache_name='nbacom-agent',
cookies=None,
table_names=None):
'''
Args:
db (NBAPostgres): instance
cache_name (str): for scraper cache_name
cookies: cookie jar
table_names (dict): Database table names
'''
logging.getLogger(__name__).addHandler(logging.NullHandler())
self.scraper = NBAComScraper(cache_name=cache_name, cookies=cookies)
self.parser = NBAComParser()
self.db = db
if table_names:
self.table_names = table_names
else:
self.table_names = {
'pgl': 'player_gamelogs',
'tgl': 'team_gamelogs',
'seas': 'season',
'pl': 'player',
'ps': 'playerstats_daily',
'ts': 'teamstats_daily',
'tod': 'team_opponent_dashboard',
'pbs': 'player_boxscores_combined',
'tbs': 'team_boxscores_combined',
'box2': 'game_boxscores',
'tm': 'team'
}
def __init__(self, db=None):
logging.getLogger(__name__).addHandler(logging.NullHandler())
self._dk_players = []
self.scraper = NBAComScraper()
self.parser = NBAComParser()
if db:
self.nbadb = db
def __init__(self, db=None, cache_name=None, cookies=None):
'''
Arguments:
cache_name: str for scraper cache_name
cookies: cookie jar
db: NBAComPg instance
'''
self.scraper = NBAComScraper(cache_name=cache_name, cookies=cookies)
self.parser = NBAComParser()
if db:
self.db = db
self.insert_db = True
else:
self.insert_db = False
def __init__(self, db=None, cache_name=None, cookies=None):
'''
Arguments:
cache_name: str for scraper cache_name
cookies: cookie jar
db: NBAComPg instance
'''
logging.getLogger(__name__).addHandler(logging.NullHandler())
self.scraper = NBAComScraper(cache_name=cache_name, cookies=cookies)
self.parser = NBAComParser()
if db:
self.db = db
self.insert_db = True
else:
self.insert_db = False
import sys
import time
from configparser import ConfigParser
from nba.scrapers.nbacom import NBAComScraper
from nba.db.nbacom import NBAComPg
from nba.db.queries import missing_games_meta
logging.basicConfig(stream=sys.stdout, level=logging.INFO)
config = ConfigParser()
configfn = os.path.join(os.path.expanduser('~'), '.pgcred')
config.read(configfn)
s = NBAComScraper(cache_name='games-meta')
nbapg = NBAComPg(username=config['nbadb']['username'],
password=config['nbadb']['password'],
database=config['nbadb']['database'])
teams = []
url = 'http://data.nba.com/data/10s/prod/v1/{game_date}/{game_id}_boxscore.json'
headers = ['game_id', 'gamecode', 'game_date', 'team_code', ]
for item in nbapg.select_dict(missing_games_meta()):
try:
content = s.get_json(url.format(game_date=item['game_date'], game_id=item['game_id']))
vteam_code = content['basicGameData']['vTeam']['triCode']
vscore = int(content['basicGameData']['vTeam']['score'])
vls = [int(l['score']) for l in content['basicGameData']['vTeam']['linescore']]
hteam_code = content['basicGameData']['hTeam']['triCode']
hscore = int(content['basicGameData']['hTeam']['score'])
def setUp(self):
logging.getLogger(__name__).addHandler(logging.NullHandler())
self.nbs = NBAComScraper()
def load_data(per_mode, playerpool_size, thresh_gp, thresh_min, lastn=0,
season_code=None, sortcol=None, fn=None):
'''
Loads data from nba.com or csv file
Args:
per_mode (str): 'Totals', 'PerGame', 'Per48'
playerpool_size (int): number of players in pool
thresh_gp (int): minimum number of games played
thresh_min (int): minimum number of minutes played
lastn (int): last number of games, default 0
season_code (str): '2017-18', etc., default None
sortcol (str): sort, default None
fn (str): filename of csv, default None
Returns:
DataFrame
'''
if fn:
df = pd.read_csv(fn)
else:
scraper = NBAComScraper(cache_name='fbasim')
parser = NBAComParser()
if not season_code:
season_code = current_season_code()
content = scraper.playerstats(season_code, per_mode, lastn)
df = pd.DataFrame(parser.playerstats(content, per_mode))
df['MIN'] = df['MIN'].astype(int)
df = df.rename(index=str, columns={"TEAM_ABBREVIATION": "TEAM"})
# account for gp & minutes thresholds
if thresh_gp and thresh_min:
crit = (df.GP >= thresh_gp) & (df.MIN >= thresh_min)
nthresh = len(df[crit])
if nthresh > playerpool_size:
df = df[crit][0:playerpool_size]
else:
df = df[crit]
elif thresh_gp:
crit = df.GP >= thresh_gp
nthresh = len(df[crit])
if nthresh > playerpool_size:
df = df[0:playerpool_size]
else:
df = df[crit]
elif thresh_min:
crit = df.MIN >= thresh_min
nthresh = len(df[crit])
if nthresh > playerpool_size:
df = df[0:playerpool_size]
else:
df = df[crit]
# need index for joins in the sim
df.set_index('PLAYER_ID')
if sortcol:
return df.sort_values(sortcol, ascending=False)
else:
return df
