如何根據數組的兩個日期添加列值？

Question

我是編程新手。我遇到了這個要求。我有一個數組，

data= ['2016-1-01', '2016-1-08', '2016-1-15', '2016-1-22', '2016-1-29', '2016-02-05', '2016-02-12', '2016-02-19', '2016-02-26'] 

我有查詢結果如下：

date   a  b  c 
2016-01-19 3  1  5 
2016-01-20 10  4  5 
2016-01-30 1  4  6 

我想生成每週報告的數據。

在本示例中，日期'2016-01-19'和'2016-01-20'位於數據數組的「2016-01-15」和「2016-01-22」之間，因此a， b要加上c & c。

最終的輸出應該是這樣的：

2016-1-01  0  0  0 
2016-1-08  0  0  0 
2016-1-15  13  5  10 
2016-1-22  0  0  0 
2016-1-29  1  4  6 
2016-2-05  0  0  0 
2016-2-12  0  0  0 
2016-2-19  0  0  0 
2016-2-26  0  0  0 

Answer 1

這是一個假設data是有序的，否則使用sorted(data)。

import datetime 

data = [ 
    '2016-1-01', '2016-1-08', '2016-1-15', 
    '2016-1-22', '2016-1-29', '2016-02-05', 
    '2016-02-12', '2016-02-19', '2016-02-26' 
] 

query_result = [ 
    (datetime.date(2016, 1, 19), 3, 1, 5), 
    (datetime.date(2016, 1, 20), 10, 4, 5), 
    (datetime.date(2016, 1, 30), 1, 4, 6) 
] 

# Convert string dates to datetime.date 
date_data = [ datetime.datetime.strptime(date, '%Y-%m-%d').date() 
       for date in data ] 

res = [] 
# zip the dates together in pairs 
for start, end in zip(date_data, date_data[1:]): 
    tally_a = tally_b = tally_c = 0 
    for date, a, b, c in query_result: 
     # if date is in between add values 
     if start <= date <= end: 
      tally_a += a 
      tally_b += b 
      tally_c += c 

    res.append((start, tally_a, tally_b, tally_c)) 

# Output 
for d, a, b, c in res: 
    print(d, a, b, c, sep = '\t') 

2016-01-01 0 0 0 
2016-01-08 0 0 0 
2016-01-15 13 5 10 
2016-01-22 0 0 0 
2016-01-29 1 4 6 
2016-02-05 0 0 0 
2016-02-12 0 0 0 
2016-02-19 0 0 0 

Answer 2

假設data總是排序，沒有重複的元素（你可以做data = sorted(set(data))如果不是這樣），你可以做這樣的事情：

import datetime 

data = ['2016-1-01', '2016-1-08', '2016-1-15', '2016-1-22', '2016-1-29', '2016-02-05', '2016-02-12', '2016-02-19', '2016-02-26'] 
query = [(datetime.date(2016, 1, 19), 3, 1, 5), (datetime.date(2016, 1, 20), 10, 4, 5), (datetime.date(2016, 1, 30), 1, 4, 6)] 

# Convert data to datetime objects 
data = [datetime.datetime.strptime(d, '%Y-%m-%d').date() for d in data] 

output = [] 
query_it = iter(query) 
next_date = data[0] 
next_nums = (0, 0, 0) 
# Iterate through date ranges 
for d_start, d_end in zip(data, data[1:] + [datetime.date.max]): 
    # If the next interesting date is in range 
    if next_date < d_end: 
     nums = next_nums 
     next_nums = (0, 0, 0) 
     for q in query_it: 
      q_date, q_nums = q[0], q[1:] 
      if q_date < d_start: 
       # Ignore dates before the first date in data 
       continue 
      elif q_date < d_end: 
       # Add query numbers to count if in range 
       nums = tuple(n1 + n2 for n1, n2 in zip(nums, q_nums)) 
      else: 
       # When out of range save numbers for next 
       next_date = q_date 
       next_nums = q_nums 
       break 
    else: 
     # Default to zero when no query dates in range 
     nums = (0, 0, 0) 
    # Add result to output 
    output.append((d_start,) + nums) 

for out in output: 
    print(out) 

輸出：

(datetime.date(2016, 1, 1), 0, 0, 0) 
(datetime.date(2016, 1, 8), 0, 0, 0) 
(datetime.date(2016, 1, 15), 13, 5, 10) 
(datetime.date(2016, 1, 22), 0, 0, 0) 
(datetime.date(2016, 1, 29), 1, 4, 6) 
(datetime.date(2016, 2, 5), 0, 0, 0) 
(datetime.date(2016, 2, 12), 0, 0, 0) 
(datetime.date(2016, 2, 19), 0, 0, 0) 
(datetime.date(2016, 2, 26), 0, 0, 0)