我有以下输入表，我想获取输出表。
为此，我必须检查从第 14 行到文件末尾的每个单元格内容的最后部分，如果有匹配项，则将整个单元格值移动到第 10、11、12、13 行中（“STA ATT”、“TEC”、“D C P”、“A SPV”） 我必须这样做直到最后一栏
我编写了一个代码来做到这一点，它有效，但它非常混乱、低效且缓慢。我用 openpyxl 编写了它，但我认为这是最慢的方法，并且我认为 pandas 提供了最快的方法。
在下面的例子中： 对于 095 列，我必须将第 16 行中的内容移动到单元格 10，将第 17 行中的内容移动到单元格 11 对于第一列 088，我必须将第 14 行中的内容移动到单元格 12、15 到单元格 13、17 到单元格 11 对于第二列 088，我必须将第 18 行中的内容移动到单元格 12、19 到单元格 10、16 到单元格 10、20 到单元格 11 等等..
从 ROW 14 开始，每列中只有 4 个值需要移动到相应的 ROW 中 这些值不一定是连续的，但从第一个值（可以是空单元格）开始，其余值位于接下来的 3 个单元格中。 例如，在第 095 列中，我将仅移动 2 个值，X XX X_05.04.2024 16:40.STA ATT 和 X XX X_05.04.2024 16:40.TEC，因为前两个单元格为空
你能帮助我提高性能吗？ 使用我编写的代码，我需要 15 分钟才能完成 df 为 482 列和 8k 行的任务
#输入
0 | Unnamed: 0 | 095 | 088 | 088 | 713 | 714 |
1 | APP SPV | | | | | |
2 | DAT | | 11/04/2024 | 11/04/2024 | | |
3 | ST AT | AN | ABA | ABA | | PP |
4 | TEC | TOP | | | CAS | |
5 | DATE | | | | | |
6 | | | | | | |
7 | | | | | | |
8 | | | | | | |
9 | | | | | | |
10 | STA ATT | | | | | |
11 | TEC | | | | | |
12 | A SPV | | | | | |
13 | D C P | | | | | |
14 | A SPV_05.04.2024 16:40 | | X XX X_05.04.2024 16:40.A SPV | | | |
15 | D C P_05.04.2024 16:40 | | X XX X_05.04.2024 16:40.D C P | | | |
16 | STA ATT_05.04.2024 16:40 | X XX X_05.04.2024 16:40.STA ATT | | | X XX X_05.04.2024 16:40.STA ATT | |
17 | TEC_05.04.2024 16:40 | X XX X_05.04.2024 16:40.TEC | X XX X_05.04.2024 16:40.TEC | | | |
18 | A SPV_05.04.2024 18:37 | | | X XX X_05.04.2024 18:37.A SPV | | |
19 | STA ATT_05.04.2024 18:37 | | | X XX X_05.04.2024 18:37.STA ATT | | |
20 | TEC_05.04.2024 18:37 | | | X XX X_05.04.2024 18:37.TEC | | |
21 | A SPV_06.04.2024 10:11 | | | | | X XX X_06.04.2024 10:11.A SPV |
#输出
0 | Unnamed: 0 | 095 | 088 | 088 | 713 | 714 |
1 | APP SPV | | | | | |
2 | DAT | | 11/04/2024 | 11/04/2024 | | |
3 | ST AT | AN | ABA | ABA | | PP |
4 | TEC | TOP | | | CAS | |
5 | DATE | | | | | |
6 | | | | | | |
7 | | | | | | |
8 | | | | | | |
9 | | | | | | |
10 | STA ATT | X XX X_05.04.2024 16:40.STA ATT | | X XX X_05.04.2024 18:37.STA ATT | X XX X_05.04.2024 16:40.STA ATT | |
11 | TEC | X XX X_05.04.2024 16:40.TEC | X XX X_05.04.2024 16:40.TEC | X XX X_05.04.2024 18:37.TEC | | |
12 | A SPV | | X XX X_05.04.2024 16:40.A SPV | X XX X_05.04.2024 18:37.A SPV | |X XX X_06.04.2024 10:11.A SPV |
13 | D C P | | X XX X_05.04.2024 16:40.D C P | | | |
14 | A SPV_05.04.2024 16:40 | | | | | |
15 | D C P_05.04.2024 16:40 | | | | | |
16 | STA ATT_05.04.2024 16:40 | | | | | |
17 | TEC_05.04.2024 16:40 | | | | | |
18 | A SPV_05.04.2024 18:37 | | | | | |
19 | STA ATT_05.04.2024 18:37 | | | | | |
20 | TEC_05.04.2024 18:37 | | | | | |
21 | A SPV_06.04.2024 10:11 | | | | | |
#代码
def find_between_r( s, first, last ):
try:
start = s.rindex( first ) + len( first )
end = s.rindex( last, start )
return s[start:end]
except ValueError:
return ""
cont = 0
val_a = ""
val_b = ""
val_c = ""
val_d = ""
val_cel_2 = ""
for colum in range(2,wsa.max_column+1):
cont = 0
Ul_pa_0 = ""
val_cel_2 = ""
for ro in range(15,wsa.max_row+1):
val_cel_2 = ""
In_St_At= 0
In_Te= 0
In_App_SPV = 0
In_Da_Ci_Pr = 0
val_cel_2 = str(wsa.cell(ro,colum).value)
if val_cel_2 != None:
if "_" in val_cel_2:
Ul_pa_0 = find_between_r(val_cel_2, "_",".")
wsa.cell(6,colum).value = Ul_pa_0
if cont < 1:
cont = cont +1
ro = ro
ro1= ro +1
ro2 = ro +2
ro3 = ro +3
Val = wsa.cell(ro,colum).value
Val1 = wsa.cell(ro1,colum).value
Val2 = wsa.cell(ro2,colum).value
Val3 = wsa.cell(ro3,colum).value
if Val == None or Val == '':
Val = "Null"
val_a = "Null"
if Val1 == None or Val1 == '':
Val1 = "Null"
val_b = "Null"
if Val2 == None or Val2 == '':
Val2 = "Null"
val_c = "Null"
if Val3 == None or Val3 == '':
Val3 = "Null"
val_d = "Null"
if Val != None and Val != "Null":
val_a = Val.split("_")[1]
Val = Val.split("_")[0]
if Val1 != None and Val1 != "Null":
val_b = Val1.split("_")[1]
Val1 = Val1.split("_")[0]
if Val2 != None and Val2 != "Null":
val_c = Val2.split("_")[1]
Val2 = Val2.split("_")[0]
if Val3 != None and Val3 != "Null":
val_d = Val3.split("_")[1]
Val3 = Val3.split("_")[0]
if "STA ATT" in val_a and Val != "Null" and In_St_At!= 1:
wsa.cell(11,colum).value = Val
In_St_At= 1
if "STA ATT" in val_b and Val1 != "Null" and In_St_At!= 1:
wsa.cell(11,colum).value = Val1
In_St_At= 1
if "STA ATT" in val_c and Val2 != "Null" and In_St_At!= 1:
wsa.cell(11,colum).value = Val2
In_St_At= 1
if "STA ATT" in val_d and Val3 != "Null" and In_St_At!= 1:
wsa.cell(11,colum).value = Val3
In_St_At= 1
if "TEC" in val_a and Val != "Null" and In_Te!= 1:
wsa.cell(12,colum).value = Val
In_Te= 1
if "TEC" in val_b and Val1 != "Null" and In_Te!= 1:
wsa.cell(12,colum).value = Val1
In_Te= 1
if "TEC" in val_c and Val2 != "Null" and In_Te!= 1:
wsa.cell(12,colum).value = Val2
In_Te= 1
if "TEC" in val_d and Val3 != "Null" and In_Te!= 1:
wsa.cell(12,colum).value = Val3
In_Te= 1
if "A SPV" in val_a and Val != "Null" and In_App_SPV != 1:
wsa.cell(13,colum).value = Val
In_App_SPV = 1
if "A SPV" in val_b and Val1 != "Null" and In_App_SPV != 1:
wsa.cell(13,colum).value = Val1
In_App_SPV = 1
if "A SPV" in val_c and Val2 != "Null" and In_App_SPV != 1:
wsa.cell(13,colum).value = Val2
In_App_SPV = 1
if "A SPV" in val_d and Val3 != "Null" and In_App_SPV != 1:
wsa.cell(13,colum).value = Val3
In_App_SPV = 1
if "D C P" in val_a and Val != "Null" and In_Da_Ci_Pr != 1:
wsa.cell(14,colum).value = Val
In_Da_Ci_Pr = 1
if "D C P" in val_b and Val1 != "Null" and In_Da_Ci_Pr != 1:
wsa.cell(14,colum).value = Val1
In_Da_Ci_Pr = 1
if "D C P" in val_c and Val2 != "Null" and In_Da_Ci_Pr != 1:
wsa.cell(14,colum).value = Val2
In_Da_Ci_Pr = 1
if "D C P" in val_d and Val3 != "Null" and In_Da_Ci_Pr != 1:
wsa.cell(14,colum).value = Val3
In_Da_Ci_Pr = 1
使用 pandas 或更快的方法提高性能
mergeout = df.copy()
ref = df.iloc[10:, 1].reset_index(name='ref')
out.iloc[10:, 2:] = (
df.iloc[14:, 2:]
.apply(lambda s: ref
.merge(s.rename('out'), how='left', left_on='ref',
right_on=s.str.extract(r'([^.]+)$', expand=False))
.set_index('index')['out'])
)
请注意，就像您的描述中一样，要使用的窗口的位置是硬编码的（第 2 列之后、第 10 行之后、第 14 行之后）。
输出：
0 1 2 3 4 5 6
0 0 Unnamed: 0 095 088 088 713 714
1 1 APP SPV NaN NaN NaN NaN NaN
2 2 DAT NaN 11/04/2024 11/04/2024 NaN NaN
3 3 ST AT AN ABA ABA NaN PP
4 4 TEC TOP NaN NaN CAS NaN
5 5 DATE NaN NaN NaN NaN NaN
6 6 NaN NaN NaN NaN NaN NaN
7 7 NaN NaN NaN NaN NaN NaN
8 8 NaN NaN NaN NaN NaN NaN
9 9 NaN NaN NaN NaN NaN NaN
10 10 STA ATT X XX X_05.04.2024 16:40.STA ATT NaN X XX X_05.04.2024 18:37.STA ATT X XX X_05.04.2024 16:40.STA ATT NaN
11 11 TEC X XX X_05.04.2024 16:40.TEC X XX X_05.04.2024 16:40.TEC X XX X_05.04.2024 18:37.TEC NaN NaN
12 12 A SPV NaN X XX X_05.04.2024 16:40.A SPV X XX X_05.04.2024 18:37.A SPV NaN X XX X_06.04.2024 10:11.A SPV
13 13 D C P NaN X XX X_05.04.2024 16:40.D C P NaN NaN NaN
14 14 A SPV_05.04.2024 16:40 NaN NaN NaN NaN NaN
15 15 D C P_05.04.2024 16:40 NaN NaN NaN NaN NaN
16 16 STA ATT_05.04.2024 16:40 NaN NaN NaN NaN NaN
17 17 TEC_05.04.2024 16:40 NaN NaN NaN NaN NaN
18 18 A SPV_05.04.2024 18:37 NaN NaN NaN NaN NaN
19 19 STA ATT_05.04.2024 18:37 NaN NaN NaN NaN NaN
20 20 TEC_05.04.2024 18:37 NaN NaN NaN NaN NaN
21 21 A SPV_06.04.2024 10:11 NaN NaN NaN NaN NaN