带循环附加数据框

Question

Morning,

I have 3 excels that i have imported via from excel. I am trying to create a DataFrame which has taken the name ('Ticker') column from each import, add the title of the excel ('Secto') and append it to eachother to create a new DataFrame. This new DataFrame will then be exported to excel.

AA  = ['Aero&Def','REITs', 'Auto&Parts']

File = 'FTSEASX_'+AA[0]+'_Price.xlsx'
xlsx = pd.ExcelFile('C:/Users/Ben/'+File)
df = pd.read_excel(xlsx, 'Price_Data')
df = df[df.Identifier.notnull()]
df.fillna(0)
a = []
b = []
for i in df['Ticker']:
    a.append(i)
    b.append(AA[0])
raw_data = {'Ticker': a, 'Sector': b}
df2 = pd.DataFrame(raw_data, columns = ['Ticker', 'Sector'])

del AA[0]

for j in AA:
    File = 'FTSEASX_'+j+'_Price.xlsx'
    xlsx = pd.ExcelFile('C:/Users/Ben/'+File)
    df3 = pd.read_excel(xlsx, 'Price_Data')
    df3 = df3[df3.Identifier.notnull()]
    df3.fillna(0)
    a = []
    b = []
    for i in df3['Ticker']:
        a.append(i)
        b.append(j)
    raw_data = {'Ticker': a, 'Sector': b}
    df4 = pd.DataFrame(raw_data, columns = ['Ticker', 'Sector'])
    df5 = df2.append(df4)

I am currently getting the below but obviously the 2nd import, titled 'REITs' is not getting captured.

Ticker  Sector
0   AVON-GB Aero&Def
1   BA-GB   Aero&Def
2   COB-GB  Aero&Def
3   MGGT-GB Aero&Def
4   SNR-GB  Aero&Def
5   ULE-GB  Aero&Def
6   QQ-GB   Aero&Def
7   RR-GB   Aero&Def
8   CHG-GB  Aero&Def
0   GKN-GB  Auto&Parts

How would i go about achieving this? or is there a better more pythonic way of achieving this?

Answer 1

I would do it this way:

import pandas as pd

AA  = ['Aero&Def','REITs', 'Auto&Parts']

# assuming that ['Ticker','Sector','Identifier'] columns are in 'B,D,E' Excel columns
xl_cols='B,D,E'

dfs = [ pd.read_excel('FTSEASX_{0}_Price.xlsx'.format(f),
                      'Price_Data',
                      parse_cols=xl_cols,
                     ).query('Identifier == Identifier')
        for f in AA]

df = pd.concat(dfs, ignore_index=True)

print(df[['Ticker', 'Sector']])

Explanation:

.query('Identifier == Identifier') - gives you only those rows where Identifier is NOT NULL (using the fact that value == NaN will always be False )

PS You don't want to loop through your data frames when working with Pandas...