Untitled

thirtymins_dfs = []

for file in sorted(glob('30*.csv')):
    print(file)
    nm = file.split('-')[1].split('.csv')[0]
    data = pd.read_csv(file)
    data['file'] = nm
    
    for col_name in ['ASA','Avg Wait','Avg Handle','Avg Talk','Avg Hold','Avg ACW']:
        data[col_name] = pd.to_timedelta(data[col_name]).apply(lambda x: x.total_seconds() * 1e3)
        data[col_name] = data[col_name].astype('float')
    
    thirtymins_dfs.append(data)
    
    # all non-null ASA and other timed columns are in hh:mm:ss.mmmm, not milliseconds
    
# for each, quickly check if any duplicate date rows
for df in thirtymins_dfs:
    print(df['file'][0])
    print(df['Interval Start'].nunique())
    print('----')
    print(len(df['Interval Start']))
    print('----')
    print(df[df.duplicated(['Interval Start'],keep=False)])
    
# so no direct duplicates against the data