Coverage for statistic/adjust_time.py: 94%

1import pandas as pd 

2import sys 

3import os 

4 

5# Определите путь до корня проекта 

6project_root = os.path.abspath(os.path.join(os.path.dirname(__file__), '..')) 

7if project_root not in sys.path: 

8 sys.path.append(project_root) 

9 

10 

11from src.settings import Settings 

12 

13 

14 

15def adjust_time1(df): 

16 """ adjust time in the dataframe if there are more than 2 orders in the same time 

17 if two orders adds 10 minutes to the second order 

18 if three orders removes 10 minutes from the first order and adds 10 minutes to the third order 

19 if four orders removes 20 minutes from the first order and 10 minutes from the second order 

20 and adds 10 minutes to the fourth order 

21 ... 

22 Args: 

23 df (DataFrame pandas): dataframe with the orders 

24 

25 Returns: 

26 DataFrame: dataframe with the adjusted time 

27 """ 

28 max_iterations = 10 # max number of iterations to avoid infinite loop 

29 iteration = 0 

30 

31 while iteration < max_iterations: 

32 # group by time and wenz 

33 time_groups = df.groupby(['time', 'wenz']) 

34 

35 any_changes = False 

36 

37 result = [] 

38 for _ , group in time_groups: 

39 if len(group) == 2: 

40 group.iat[0, group.columns.get_loc('time')] -= pd.Timedelta(minutes=5 + iteration) 

41 any_changes = True 

42 elif len(group) == 3: 

43 group.iat[0, group.columns.get_loc('time')] -= pd.Timedelta(minutes=5 + iteration) 

44 group.iat[2, group.columns.get_loc('time')] += pd.Timedelta(minutes=5 + iteration) 

45 any_changes = True 

46 elif len(group) == 4: 

47 group.iat[0, group.columns.get_loc('time')] -= pd.Timedelta(minutes=10 + iteration) 

48 group.iat[1, group.columns.get_loc('time')] -= pd.Timedelta(minutes=5 + iteration) 

49 group.iat[3, group.columns.get_loc('time')] += pd.Timedelta(minutes=5 + iteration) 

50 any_changes = True 

51 elif len(group) >= 5: 

52 group.iat[0, group.columns.get_loc('time')] -= pd.Timedelta(minutes=10 + iteration) 

53 group.iat[1, group.columns.get_loc('time')] -= pd.Timedelta(minutes=5 + iteration) 

54 group.iat[-2, group.columns.get_loc('time')] += pd.Timedelta(minutes=5 + iteration) 

55 group.iat[-1, group.columns.get_loc('time')] += pd.Timedelta(minutes=10 + iteration) 

56 any_changes = True 

57 result.append(group) 

58 

59 df = pd.concat(result) 

60 

61 if not any_changes: 

62 break # if no changes were made, exit the loop 

63 

64 iteration += 1 

65 

66 return df 

67 

68 

69min_interval = pd.Timedelta(minutes=7) 

70 

71def adjust_time2(df): 

72 """adjust time in the dataframe if interval between the orders is less tha settings.min_interval 

73 if the interval is less than settings.min_interval set it equal to settings.min_interval 

74 

75 Args: 

76 df (DataFrame): dataframe with the orders 

77 

78 Returns: 

79 DataFrame: dataframe with the adjusted time 

80 """ 

81 min_interval = pd.Timedelta(minutes=Settings.min_interval) 

82 

83 df = df.sort_values(by='time').reset_index(drop=True) 

84 for i in range(1, len(df)): 

85 prev_time = df.loc[i - 1, 'time'] 

86 curr_time = df.loc[i, 'time'] 

87 

88 if curr_time < prev_time: 

89 curr_time = prev_time 

90 

91 df.loc[i, 'time'] = curr_time + min_interval 

92 

93 else: 

94 interval = curr_time - prev_time 

95 

96 if interval < min_interval: 

97 df.loc[i, 'time'] = curr_time + (min_interval - interval) 

98 return df 

99 

100def adjust_times(df): 

101 """combination of adjust_time1 and adjust_time2 and sort the dataframe by time, 

102 reset the index and add 1 to the index 

103 

104 Args: 

105 df (DataFrame): dataframe with the orders 

106 

107 Returns: 

108 DataFrame: dataframe with the adjusted time and sorted by time, with index reset and 1 added to the index 

109 """ 

110 

111 # Spread the corses by 10 minutes if they are at the same time 

112 df = adjust_time1(df) 

113 df.sort_values("time", inplace=True) # type: ignore 

114 

115 # Spread the shipments so that the interval between them is not less than Settings.min_interval 

116 df = df.groupby('wenz')[df.columns].apply(adjust_time2).reset_index(drop=True) 

117 df.sort_values("time", inplace=True) # type: ignore 

118 df.reset_index(drop=True, inplace=True) 

119 df.index=df.index+1 

120 

121 return df 

122 

123 

124 

125 

126if __name__ == "__main__": 

127 # import sys, os 

128 # from pathlib import Path 

129 # project_root = os.path.abspath(os.path.join(os.path.dirname(__file__), '..')) 

130 # if project_root not in sys.path: 

131 # sys.path.append(project_root) 

132 

133 

134 data = { 

135 'time': [pd.Timestamp('2023-01-01 10:00:00')] * 15, 

136 'wenz': ['A', 'A', 'A', 'B', 'B', 'A', 'A', 'A', 'B', 'B', 'A', 'B', 'B', 'A', 'A',] 

137 } 

138 df = pd.DataFrame(data) 

139 print(adjust_time1(df)['time'])