Question

我正在处理一个Excel文件，其中按month_year（9月18日至10月15日）包含多（36）张工作表，并使用字典进行读取

<div id="container">
   <div class="item">
      Text with a <a href="#">link</a> in it.
      This is text without a link in it.
   </div>

   <div class="item"> 
      Text with a <a href="#">link</a> in it.
      This is text without a link in it.
   </div>
</div>

我需要一次设置所有14个列的名称，但出现ValueError：长度不匹配

在这里，我们可以看到一些工作表仅包含13列

import pandas as pd

fileName = 'project_dropColumnICSv2.xlsx'
df = pd.ExcelFile(fileName)

sheetNames = df.sheet_names
vars_dict = {}

for sheetName in sheetNames:
    vars_dict["the_{0}".format(sheetName)] = pd.read_excel(fileName, sheet_name=sheetName, index_col=False)

mykeys = []

for key, value in vars_dict.items():
    mykeys.append(key)

我尝试仅添加另一列

for mykey in mykeys:
    print("'{}' contains {} columns".format((mykey), len(vars_dict.get(mykey).columns)))

'the_Sep 18' contains 14 columns
'the_Aug 18' contains 14 columns
'the_Jul 18' contains 14 columns
'the_Jun 18' contains 14 columns
'the_May 18' contains 14 columns
'the_April 18' contains 14 columns
'the_March 18' contains 14 columns
'the_February 18' contains 13 columns
'the_January 18' contains 14 columns
'the_December 17' contains 13 columns
'the_November 17' contains 13 columns
'the_October 17' contains 13 columns
'the_September 17' contains 13 columns
'the_August 17' contains 14 columns
'the_July 17' contains 14 columns
'the_June 17' contains 14 columns
'the_May 17' contains 14 columns
'the_April 17' contains 14 columns
'the_MARCH 17' contains 14 columns
'the_February17' contains 14 columns
'the_January17' contains 14 columns
'the_December16' contains 14 columns
'the_November16' contains 14 columns
'the_October 16' contains 14 columns
'the_September' contains 14 columns
'the_August' contains 15 columns
'the_July' contains 14 columns
'the_June' contains 14 columns
'the_May' contains 14 columns
'the_April' contains 14 columns
'the_March' contains 13 columns
'the_February' contains 13 columns
'the_January' contains 13 columns
'the_December' contains 13 columns
'the_November' contains 14 columns
'the_October' contains 13 columns

使用for循环更改列名称，但结果是第一列的字段错误，简而言之，未对齐。

假设我的列名有一个数组，我该如何做呢？

for mykey in mykeys:
    if len(vars_dict.get(mykey).columns) == 13:
        vars_dict.get(mykey)['Another Column'] = 'Nan'

P.S。有一张纸包含15列，只需删除最后一个

Answer 1

我认为需要参数sheet_name=None才能将所有工作表转换为read_excel中的OrderedDict of DataFrames：

fileName = 'project_dropColumnICSv2.xlsx'
dfs = pd.read_excel(fileName, sheet_name=None)

然后使用字典理解来检查列数，并用assign进行设置，并创建新字典：

dfs = {k: v.assign(New = np.nan) if len(v.columns) == 13 else v for k, v in dfs.items()}

如果还需要更改键：

dfs = {f'the_{}'.format(k): v.assign(New = np.nan) 
       if len(v.columns) == 13 
       else v for k, v in dfs.items()}

然后通过键选择每个DataFrame：

print (dfs['Sep 18'])

熊猫不从xlsx文件读取第一列

1 个答案: