return nr[0], int(wl[0]), int(wl[1])
results = []
for line in coaches:
chstr = line['coaches_text'].split(", ")
chid = line['coaches_id'].split('|')
chs = zip(chstr, chid)
order = len(chs)
for c in reversed(chs):
name, wins, losses = parse_coach(c[0])
res = {}
res['team'] = line['team']
res['franchise'] = line['franchise']
res['year'] = line['year']
res['coaches_text'] = line['coaches_text']
res['coach'] = name
res['coach_id'] = c[1]
res['wins'] = wins
res['losses'] = losses
res['games'] = wins + losses
res['order'] = order
results.append(res)
order -= 1
settings.write_csv(
results,
'data/coaches-ordered.csv',
['franchise','team','year','coaches_text','coach','coach_id','wins','losses','games','order']
)
# coaches must appear in the right order for this to work
coaches = sorted(coaches, key=lambda x: (x['franchise'], x['year'], x['order']))
# make a "lookup" such that team_look[team_name][game_num] returns the coach for that game
team_look = {}
for c in coaches:
f = c['franchise']
s = c['year']
g = int(c['games'])
ch = c['coach_id']
if not team_look.get(f):
team_look[f] = {}
if not team_look[f].get(s):
team_look[f][s] = []
for i in range(0, g):
team_look[f][s].append(ch)
with open('data/team-coach-lookup.json', 'w') as outfile:
json.dump(team_look, outfile, indent=2)
# attach the coach's name to each game
games = settings.read_csv('scrape/sportsref-gamelogs.csv')
for g in games:
g['coach'] = team_look[g['franchise']][g['season']][int(g['game_num'])-1]
settings.write_csv(
games,
'data/gamelogs-coaches.csv',
['franchise','team','season','game_num','date','result','coach'],
)
}
for coach in rawcoaches:
poc = note = ''
if coach['coach_id'] in oldpoc:
poc, note = oldpoc[coach['coach_id']]
coach['poc'] = poc
coach['note'] = note
rawcoaches = sorted(rawcoaches,
key=lambda x: (x['poc'], x['coach']),
reverse=True)
settings.write_csv(
rawcoaches,
'categorized/coach-list-poc.csv',
['coach', 'coach_id', 'franchises', 'poc', 'note'],
)
## update NBA player categorization
rawcoaches = settings.read_csv('data/coach-list.csv')
oldplayed = {
l['coach_id']: l['former_nba_player']
for l in settings.read_csv('categorized/coach-list-qualifications.csv')
}
for coach in rawcoaches:
former_player = ''
if coach['coach_id'] in oldplayed:
T.max_date = gm['date']
tenures.append(T)
# summaries of a coach's entire tenure
tenure_summaries = [t.summarize() for t in tenures]
with open('data/tenures.json', 'w') as outfile:
json.dump(tenure_summaries, outfile, indent=2)
settings.write_csv(
tenure_summaries,
'data/tenures-summarized.csv',
[
'slug', 'coach', 'coach_id', 'franchise', 'min_date', 'max_date',
'left_truncated'
],
)
# unique coach names and franchises they coached for
coach_names = set([(t.coach, t.coach_id) for t in tenures])
coach_info = [{
'coach':
n[0],
'coach_id':
n[1],
'franchises':
'|'.join(set(t.franchise for t in tenures if t.coach_id == n[1]))
} for n in coach_names]
continue
elif y <= 2002:
f = 'NORLHP'
s = 'CHH'
elif s == 'NOH':
f = 'NORLHP'
if y >= 2014:
s = 'NOP'
# elif y <= 2002:
# s = 'CHH'
elif y == 2006 or y == 2007:
s = 'NOK'
coachres['team'] = s
coachres['franchise'] = f
coachres['coaches_text'] = t.find(attrs={
"data-stat": "coaches"
}).text
coachres['coaches_id'] = '|'.join(
c['href'].split('/')[-1][:-5]
for c in t.find(attrs={
"data-stat": "coaches"
}).find_all('a'))
coaches.append(coachres)
settings.write_csv(
gamelogs, 'scrape/sportsref-gamelogs.csv',
['franchise', 'team', 'season', 'game_num', 'date', 'result'])
settings.write_csv(coaches, 'scrape/sportsref-coaches.csv',
['franchise', 'team', 'year', 'coaches_text', 'coaches_id'])