用于从 txt 文件中提取数据到 excel 的 Python 脚本

Question

Here is my file:

example.txt

new names_a

    tim
    jeremy; 24 - age

next

new "names_b"

    jordan; 27 - age
    alex; 26 - age
    steven; 24 - age

next

new names_c

    johnny; 20 - age
    ron
    ;joe; 19 - age
    brian; 23 - age

next

Here is my code:

file=open("example.txt", "r")
data=file.read()
categories=data.split('new')
dict_format={}
for categor_data in categories:
    items=categor_data.split('\n')
    item_name=items[0].replace(" ", "")
    item_name=item_name.strip('"')
    dict_format[item_name]=items[1:]

for name in dict_format:
    print(name)

print("Which category to export?")
answer=input()

with open(answer+".csv",'w') as csv:
    for row in dict_format[answer][:-1]:
        if row != "":
            csv.write(row.replace(";",",")+"\n")

    csv.write(dict_format[answer][-1].replace(";",","))
    csv.close()

Example output for the category names_c:

	A	B	C
1	johnny	20 - age
2	ron
3		joe	19 - age
4	brian	23 - age
5
6	next

Question 1:

Is there a way to either:

a) have the code not read the word 'next' as part of the list

b) open the file, delete all entries of the word 'next', save the file, close the file, reopen the file and then run the code

Question 2:

Is there a way to not output entries starting with ';'? ex: ;joe; 19 - age

Question 3:

Is there a way to delete empty rows?

Desired output:

	A	B	C
1	johnny	20 - age
2	ron
3	brian	23 - age

Answer 1

This is one way to do it:

import pandas as pd
# Read data using Pandas
df = pd.read_csv('example.txt',sep = '\n+', header = None, engine='python')
# Drop rows starting with ';' (e.g. ;joe; 19 - age) and 'next'
df = df.drop(df[df[0].str.startswith((';','next'))].index)
# Split categories
df2 = df[0].str.replace('"','').str.split('new ',expand=True)
# Grop dataframe by categories
df3.fillna(method='ffill',inplace=True)
gp = df4.groupby(1)
dfs = [gp.get_group(x).reset_index().drop('index',axis=1).set_index(1)[0].str.split(';',expand=True).iloc[1:] for x in gp.groups]
# save dataframes 
for df in dfs:
    df.to_csv(f"{df.index[0]}.csv",index=False)