在excel Python中过滤一列

Question

我正在尝试通过过滤值来组织一列。换句话说，有成千上万个重复的名称，我想从每个“组”中只取一个名称并将其复制到另一列中。

所以A列是当前情况，be列是我想要得到的结果：

Column A                   Column B

AB Mark Sociedad Ltda      AB Mark Sociedad Ltda
AB Mark Sociedad Ltda      Acosta Acosta Manuel
AB Mark Sociedad Ltda      ALBAGLI, ZALIASNIK 
AB Mark Sociedad Ltda
Acosta Acosta Manuel 
Acosta Acosta Manuel 
Acosta Acosta Manuel
ALBAGLI, ZALIASNIK 
ALBAGLI, ZALIASNIK
ALBAGLI, ZALIASNIK

最后这是我要使用的脚本：

import openpyxl
from openpyxl import load_workbook
import os

os.chdir('path')
workbook = openpyxl.load_workbook('abc.xlsx')
page_i = workbook.get_sheet_names()
sheet = workbook.get_sheet_by_name('Sheet1')

for a in range(1, 10):
    representativex = sheet['A' + str(a)].value
    tuple(sheet['A1':'A10'])

    for row in sheet['A1':'A10']:
        if representativex in row:
            continue
        else:
            sheet['B' + str(a)].value 
            sheet['B' + str(a)] = representativex

        workbook.save('abc.xlsx')

不幸的是，它不起作用。

Answer 1

我并没有真正使用 Python，但这是我相对较快找到的一种粗略方法。

import openpyxl

wb = openpyxl.load_workbook('test.xlsx')
ws1 = wb.active

names = []
for row in ws1.columns[0]:
    names.append(row.value)

names = sorted(list(set(names)))

start = 1
for name in names:
    ws1.cell(row = start, column=2).value = name
    start += 1

wb.save('test.xlsx')

编辑：显然openpyxl 的更新升级需要稍作修改

改变这个：

for row in ws1.columns[0]:
        names.append(row.value)

到这里：

for row in ws1.iter_cols(max_col = 1, min_row=1):
    for cell in row:
        names.append(cell.value)

以防万一您的列不同，

iter_cols(min_col=None, max_col=None, min_row=None, max_row=None)[source]

Returns all cells in the worksheet from the first row as columns.

If no boundaries are passed in the cells will start at A1.

If no cells are in the worksheet an empty tuple will be returned.
Parameters: 

    min_col (int) – smallest column index (1-based index)
    min_row (int) – smallest row index (1-based index)
    max_col (int) – largest column index (1-based index)
    max_row (int) – smallest row index (1-based index)