Pandas padding

gistfile1.txt

import random

import datetime as dt
import pandas as pd

df = pd.DataFrame({'date': ['2021-01-02', '2021-01-03', '2021-01-04',
                            '2021-01-01', '2021-01-03', '2021-01-04',
                            '2021-01-01', '2021-01-02', '2021-01-04',
                            '2021-01-01', '2021-01-02', '2021-01-03'],
                   'plant': [1, 1, 1,
                             2, 2, 2,
                             1, 1, 1,
                             2, 2, 2],
                   'product': [4, 4, 4,
                               5, 5, 5,
                               5, 5, 5,
                               4, 4, 4],
                   'value': [random.randint(1, 100) for x in range(12)]})

exaggerated_start_date = dt.date(2020, 12, 30).strftime('%Y-%m-%d')
exaggerated_end_date = dt.date(2021, 1, 8).strftime('%Y-%m-%d')
extra = pd.DataFrame({'date': [exaggerated_start_date, exaggerated_end_date],
                      'plant': [-1, -1],
                      'product': [-1, -1],
                      'value': [-1, -1]})

df = df.append(extra)

df['date'] = pd.to_datetime(df['date'], format='%Y-%m-%d')
df.set_index(df['date'], inplace=True)

df = (df.groupby(['plant', 'product'])
        .resample('D')['value']
        .first()
        .fillna(0)
        .astype(int)
        .reset_index())

df = (df.pivot(index='date', columns=['plant', 'product'], values='value')
        .unstack()
        .fillna(0)
        .reset_index())

df = df.rename(columns={0: 'value'})

df = df[(df['plant'] != -1) & (df['product'] != -1)]
print(df.head(50))