在读取 pandas 中的文件时自动转换为 integer

Question

I have a raw data set as.txt file as follows

ID    SI_Number   
1     0010
2     0005

I am pushing this data to postgresql using psycopg2. In the DB I am seeing the following

ID     SI_Number
1      10.0
2      5.0

In order to get rid of unwanted decimal places, I am using the following piece of code.

df['SI_number'] = df['SI_number'].astype(str).str.replace(r'.0$','',regex=True)

With this in the DB, I am seeing

ID    SI_Number
1      10
2      5

But I want to populate the exact value in the DB. Like below:

ID    SI_Number
1      0010
2      0005

I have tried df['SI_Number'].astype(str) but no result.

I am reading the.txt file with the following

df = pd.read_csv(f,usecols=col_lst,sep='|',engine='python',encoding='iso-8859-1',error_bad_lines=False, warn_bad_lines=True)

I am using Linux OS. Also while uploading I am converting everything to VARCHAR().

Am I missing out anything?

Answer 1

You can convert to int -> string and then apply a zfill() (zero fill)

>>> df = pd.DataFrame({"A": [1.0, 2.0, 30.0]})

>>> df
      A
0   1.0
1   2.0
2  30.0

>>> df["A"] = df["A"].astype(int).astype(str).apply(lambda x: x.zfill(4))

>>> df
      A
0  0001
1  0002
2  0030