import pandas as pd

m1 = 'maca_point_pyl.csv'
m2 = 'maca_point_white.csv'
w1 = 'wrf_point_pyl.csv'
w2 = 'wrf_point_white.csv'

sdir = '/home/disk/rocinante/DATA/temp/kcp3/scripts/puyallup_summary/list/'
ddir = '/home/disk/rocinante/DATA/temp/kcp3/scripts/puyallup_summary/data/'

dm = 'pool_maca_summary.csv'
dw = 'pool_wrf_summary.csv'

maca_tapps = "data_47.21875_-122.15625"
wrf_tapps = "data_47.23265_-122.14883"    

dfdm = pd.read_csv('{}/{}'.format(ddir, dm))
dfdw = pd.read_csv('{}/{}'.format(ddir, dw))

dfm1 = pd.read_csv('{}/{}'.format(sdir, m1))
dfm2 = pd.read_csv('{}/{}'.format(sdir, m2))
dfw1 = pd.read_csv('{}/{}'.format(sdir, w1))
dfw2 = pd.read_csv('{}/{}'.format(sdir, w2))





def get_data(df_data, df_list, name):
    dq = pd.DataFrame()
    for idx, row in df_list.iterrows():
        q = df_data[(df_data.lat == row.lat) & (df_data.lon == row.lon)]
        dq = pd.concat([dq, q])
        
    dq = dq.groupby([dq.styr, dq.edyr]).mean(numeric_only=True)
    dq['Dataset'] = name
    return dq

d1 = get_data(dfdm, dfm1, 'MACA_pyl')
d2 = get_data(dfdm, dfm2, 'MACA_white')
d3 = get_data(dfdw, dfw1, 'WRF_pyl')
d4 = get_data(dfdw, dfw2, 'WRF_white')

dtm = pd.DataFrame({'lat':[47.21875], 'lon':[-122.15625]})
dtw = pd.DataFrame({'lat':[47.23265], 'lon':[-122.14883]})

d5 = get_data(dfdm, dtm, 'MACA_laketapps')
d6 = get_data(dfdw, dtw, 'WRF_laketapps')
 
df = pd.concat([d1, d2, d3, d4, d5, d6])
df = df.drop(['lat', 'lon'], axis=1)
print(df)

df.to_csv('{}/pool_average_summary.csv'.format(ddir), float_format='%0.4f')