如何将一系列Pandas数据框行变成具有多个值的一列？

Question

Right now I have an Excel sheet in the following format which I have converted into a Pandas data frame in Python:

      Name            Column2      Unnamed: 2   Datatype   Definition
0   Entity   Accounts Receivable                                    
1    term1                                      char       term1
2    term2                                      numeric    term2
3    term3                                      char       term3
4   Entity      Accounts Payable                                    
5    term4                                      char       term4
6    term5                                      char       term5
7    term6                                      varchar    term6
8    term7                                      numeric    term7

I'm attempting to write a code that will automatically populate the empty cells in Column2 with the corresponding value for 'Entity' next to each term name. So term1, term2, and term3 would be 'Accounts Receivable' and term4, term5, term6, and term7 would be 'Accounts Payable'.

This is the code I've written so far:

   df = pd.read_excel('test.xlsx')

   df = df.replace(np.nan,'')

   values = df.values.tolist()

   ent_list = []

   for values[0] in values:
       if values[0][0] == 'Entity':
           ent_list.append(values[0][1])

   for j in range(len(values)):
       for e in range(len(ent_list)):
           while values[j][1] != ent_list[e]:
               values[j][1] = ent_list[e]
               break
           e += 1

When I print out 'values' though, I get the following:

[['Entity', 'Accounts Payable', '', '', ''], 
 ['term1', 'Accounts Payable', '', 'char', 'term1'], 
 ['term2', 'Accounts Payable', '', 'numeric', 'term2'], 
 ['term3', 'Accounts Payable', '', 'char', 'term3'], 
 ['Entity', 'Accounts Payable', '', '', ''], 
 ['term4', 'Accounts Payable', '', 'char', 'term4'], 
 ['term5', 'Accounts Payable', '', 'char', 'term5'], 
 ['term6', 'Accounts Payable', '', 'varchar', 'term6'], 
 ['term7', 'Accounts Payable', '', 'numeric', 'term7']]

Ideally it should look like this:

[['Entity', 'Accounts Receivable', '', '', ''], 
 ['term1', 'Accounts Receivable', '', 'char', 'term1'], 
 ['term2', 'Accounts Receivable', '', 'numeric', 'term2'], 
 ['term3', 'Accounts Receivable', '', 'char', 'term3'], 
 ['Entity', 'Accounts Payable', '', '', ''], 
 ['term4', 'Accounts Payable', '', 'char', 'term4'], 
 ['term5', 'Accounts Payable', '', 'char', 'term5'], 
 ['term6', 'Accounts Payable', '', 'varchar', 'term6'], 
 ['term7', 'Accounts Payable', '', 'numeric', 'term7']]

Is there a way to achieve this using the method I am currently using? I have to imagine this is possible with VBA but I'm honestly more comfortable using Python. I'm going to keep revising this code but am genuinely stumped as I am not too experienced.

I know I could do it manually but that will take too long as these reports need to generated every so often and usually include between 40,000 and 70,000 rows, and I would much prefer to automate this.

Answer 1

df = df.fillna(method = 'ffill')